Information processing method, device and equipment

Abstract

The invention provides an information processing method, device and equipment, and the information processing method comprises the steps: obtaining calculation demand information and resource state information corresponding to at least one cell; the calculation demand information comprises communication resource demand information and calculation resource demand information of a calculation task, and the resource state information comprises communication resource state information and calculation resource state information; sending the calculation demand information and the resource state information to a second network device; receiving decision information for each cell fed back by the second network equipment; the decision information comprises calculation level information and calculation topology information; and determining resource configuration information of the cell according to the decision information. According to the scheme, resource configuration can be carried out by fusing communication related information and calculation related information, the resource configuration precision is improved, and the problem that in the prior art, a calculation resource scheduling scheme is poor in resource configuration precision is solved.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and in particular to an information processing method, device and equipment. Background Art

[0002] One of the important functions of 6G networks is to provide users with computing services and guarantees with lower latency and jitter and higher overall efficiency based on the efficient collaboration of communication and distributed computing resources. In order to achieve this goal, in the current 6G network architecture, the distribution of network computing power has become increasingly marginalized, and more and more core network functions and computing power have been transferred to many network elements such as DU (distributed unit) and CU (centralized unit) of the access network, and the latency and network resource consumption have been reduced by deploying edge servers. Although this method has an immediate effect in reducing latency and reducing network resource consumption, it has also made computing resource scheduling (including allocation) in edge networks a difficult problem.

[0003] In addition, the computing resource scheduling in the existing technology does not take into account communication-related information, and is difficult to be directly applied to the 6G integrated computing network. There are problems such as poor resource configuration accuracy. Summary of the invention

[0004] The purpose of the present invention is to provide an information processing method, device and equipment to solve the problem of poor resource configuration accuracy in the solution for computing resource scheduling in the prior art.

[0005] In order to solve the above technical problem, an embodiment of the present invention provides an information processing method, which is applied to a first network device, including:

[0006] Obtain computing requirement information and resource status information corresponding to at least one cell; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0007] Sending the computing requirement information and resource status information to the second network device;

[0008] receiving decision information for each of the cells fed back by the second network device; the decision information comprising: computing level information and computing topology information;

[0009] Determine resource configuration information of the cell according to the decision information.

[0010] Optionally, the obtaining computing requirement information and resource status information corresponding to at least one cell includes:

[0011] Using a first granularity and a sample extraction method, obtaining computing demand information and resource status information corresponding to at least one cell;

[0012] The first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity;

[0013] The sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related feature attributes.

[0014] Optionally, the computing demand information further includes: at least one of: the number of computing task requests, the quality of service QoS requirement information of the computing task, and the mobility requirement information of the computing task;

[0015] And / or, the communication resource status information includes: status information of the communication connection between the first network device and the connected network element, the status information includes: at least one of the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, bandwidth utilization, bit error rate or bit error rate or frame error rate of each link, success rate of establishing each link, and link interruption probability; the connected network element refers to a network element that is communicatively connected to the first network device;

[0016] And / or, the computing resource status information includes: at least one of the total amount of computing resources, computing resource utilization, allocable computing capacity, and the number of computing tasks carried of the connected network elements of the first network device;

[0017] And / or, the computing level information includes: first level information, second level information and / or third level information, the first level information includes: information of a third network device to which the cell belongs, the second level information includes: information of the first network device to which the cell belongs and information of all third network devices under the first network device to which the cell belongs, the third level information includes: information of the first network device to which the cell belongs, information of all third network devices under the first network device to which the cell belongs, and information of a multi-access edge computing MEC device adjacent to the first network device to which the cell belongs;

[0018] And / or, the computing topology information includes: information of the network device corresponding to the computing hierarchy information, the connection relationship between the network devices corresponding to the computing hierarchy information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchy information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchy information.

[0019] Optionally, after receiving the decision information for each of the cells fed back by the second network device, the method further includes:

[0020] Obtaining configuration effect information for the decision information;

[0021] Sending the configuration effect information to the second network device;

[0022] The configuration effect information includes at least one of the following:

[0023] Calculate the number of requests accepted or the request acceptance rate of the task;

[0024] Calculate the number of successful executions or the execution success rate of tasks;

[0025] Calculate the interruption probability of the task;

[0026] QoS satisfaction degree information of computing tasks;

[0027] Resource load balancing degree, the resource load balancing degree includes: at least one of computing resource load balancing degree and communication resource load balancing degree.

[0028] Optionally, the QoS satisfaction degree information of the computing task includes: QoS satisfaction degree information corresponding to a single computing task, and at least one of QoS satisfaction degree information corresponding to at least two computing tasks;

[0029] And / or, the computing resource load balance refers to: the balance degree of computing resource utilization of each network element within the first network device and the connected network element; the connected network element refers to the network element that is communicatively connected to the first network device;

[0030] And / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0031] Optionally, before obtaining the computing requirement information and resource status information corresponding to at least one cell, the method further includes:

[0032] receiving a statistics request sent by the second network device;

[0033] The obtaining of computing demand information and resource status information corresponding to at least one cell includes:

[0034] According to the statistical request, obtaining computing demand information and resource status information corresponding to at least one cell;

[0035] The statistical request carries: the first granularity, the sample extraction method, an identifier of the information to be obtained, a calculation task number, a user identifier, and at least one of a cell identifier.

[0036] Optionally, the decision information also carries validity time information of the decision information.

[0037] The embodiment of the present invention further provides an information processing method, which is applied to a second network device, comprising:

[0038] Receiving computing requirement information and resource status information corresponding to at least one cell sent by a first network device; the computing requirement information includes: communication resource requirement information and computing resource requirement information of a computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0039] According to the computing demand information and resource status information, obtaining decision information for each of the cells; the decision information includes: computing level information and computing topology information;

[0040] The decision information is sent to the first network device.

[0041] Optionally, the computing demand information further includes: at least one of: the number of computing task requests, the quality of service QoS requirement information of the computing task, and the mobility requirement information of the computing task;

[0042] And / or, the communication resource status information includes: status information of the communication connection between the first network device and the connected network element, the status information includes: at least one of the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, bandwidth utilization, bit error rate or bit error rate or frame error rate of each link, success rate of establishing each link, and link interruption probability; the connected network element refers to a network element that is communicatively connected to the first network device;

[0043] And / or, the computing resource status information includes: at least one of the total amount of computing resources, computing resource utilization, allocable computing capacity, and the number of computing tasks carried of the connected network elements of the first network device;

[0044] And / or, the computing level information includes: first level information, second level information and / or third level information, the first level information includes: information of a third network device to which the cell belongs, the second level information includes: information of the first network device to which the cell belongs and information of all third network devices under the first network device to which the cell belongs, the third level information includes: information of the first network device to which the cell belongs, information of all third network devices under the first network device to which the cell belongs, and information of a multi-access edge computing MEC device adjacent to the first network device to which the cell belongs;

[0045] And / or, the computing topology information includes: information of the network device corresponding to the computing hierarchy information, the connection relationship between the network devices corresponding to the computing hierarchy information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchy information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchy information.

[0046] Optionally, after sending the decision information to the first network device, the method further includes:

[0047] Receiving configuration effect information sent by the first network device;

[0048] The configuration effect information includes at least one of the following:

[0049] Calculate the number of requests accepted or the request acceptance rate of the task;

[0050] Calculate the number of successful executions or the execution success rate of tasks;

[0051] Calculate the interruption probability of the task;

[0052] QoS satisfaction degree information of computing tasks;

[0053] Resource load balancing degree, the resource load balancing degree includes: at least one of computing resource load balancing degree and communication resource load balancing degree.

[0054] Optionally, the QoS satisfaction degree information of the computing task includes: QoS satisfaction degree information corresponding to a single computing task, and at least one of QoS satisfaction degree information corresponding to at least two computing tasks;

[0055] And / or, the computing resource load balance refers to: the balance degree of computing resource utilization of each network element within the first network device and the connected network element; the connected network element refers to the network element that is communicatively connected to the first network device;

[0056] And / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0057] Optionally, before receiving the computing requirement information and resource status information corresponding to at least one cell sent by the first network device, the method further includes:

[0058] Sending a statistics request to the first network device;

[0059] The receiving computing requirement information and resource status information corresponding to at least one cell sent by the first network device includes:

[0060] receiving computing requirement information and resource status information corresponding to at least one cell fed back by the first network device according to the statistical request;

[0061] The statistical request carries: at least one of the first granularity, the sample extraction method, the identifier of the information to be obtained, the calculation task number, the user identifier, and the cell identifier;

[0062] The first granularity refers to the granularity used by the first network device to obtain computing demand information and resource status information, and the sample extraction method refers to the sample extraction method used by the first network device to obtain computing demand information and resource status information;

[0063] The first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity;

[0064] The sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related feature attributes.

[0065] Optionally, also include:

[0066] Marking corresponding timestamps for the computing requirement information, resource status information, and configuration effect information respectively;

[0067] The acquiring decision information for each of the cells according to the computing demand information and the resource status information includes:

[0068] Decision information for each of the cells is obtained according to the computing requirement information, resource status information and configuration effect information marked with a timestamp.

[0069] Optionally, acquiring decision information for each of the cells according to the computing requirement information and the resource status information includes:

[0070] Obtaining information on the degree of change corresponding to the computing demand information and the resource status information;

[0071] When the change degree information is greater than or equal to a threshold, obtaining decision information for each of the cells according to the computing demand information and the resource status information;

[0072] When the change degree information is less than a threshold value, the decision information obtained last time is used as the decision information for this time.

[0073] Optionally, the decision information also carries validity time information of the decision information.

[0074] The embodiment of the present invention further provides an information processing device, which is applied to a first network device, and includes:

[0075] A first acquisition module is used to acquire computing demand information and resource status information corresponding to at least one cell; the computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0076] A first sending module, used for sending the computing requirement information and resource status information to the second network device;

[0077] A first receiving module, configured to receive decision information for each of the cells fed back by the second network device; the decision information includes: computing level information and computing topology information;

[0078] The first determination module is used to determine the resource configuration information of the cell according to the decision information.

[0079] Optionally, the obtaining computing requirement information and resource status information corresponding to at least one cell includes:

[0080] Using a first granularity and a sample extraction method, obtaining computing demand information and resource status information corresponding to at least one cell;

[0081] The first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity;

[0082] The sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related feature attributes.

[0083] Optionally, the computing demand information further includes: at least one of: the number of computing task requests, the quality of service QoS requirement information of the computing task, and the mobility requirement information of the computing task;

[0084] And / or, the communication resource status information includes: status information of the communication connection between the first network device and the connected network element, the status information includes: at least one of the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, bandwidth utilization, bit error rate or bit error rate or frame error rate of each link, success rate of establishing each link, and link interruption probability; the connected network element refers to a network element that is communicatively connected to the first network device;

[0085] And / or, the computing resource status information includes: at least one of the total amount of computing resources, computing resource utilization, allocable computing capacity, and the number of computing tasks carried of the connected network elements of the first network device;

[0086] And / or, the computing level information includes: first level information, second level information and / or third level information, the first level information includes: information of a third network device to which the cell belongs, the second level information includes: information of the first network device to which the cell belongs and information of all third network devices under the first network device to which the cell belongs, the third level information includes: information of the first network device to which the cell belongs, information of all third network devices under the first network device to which the cell belongs, and information of a multi-access edge computing MEC device adjacent to the first network device to which the cell belongs;

[0087] And / or, the computing topology information includes: information of the network device corresponding to the computing hierarchy information, the connection relationship between the network devices corresponding to the computing hierarchy information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchy information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchy information.

[0088] Optionally, also include:

[0089] A second acquisition module is used to acquire configuration effect information for the decision information after receiving the decision information for each of the cells fed back by the second network device;

[0090] A second sending module, used to send the configuration effect information to the second network device;

[0091] The configuration effect information includes at least one of the following:

[0092] Calculate the number of requests accepted or the request acceptance rate of the task;

[0093] Calculate the number of successful executions or the execution success rate of tasks;

[0094] Calculate the interruption probability of the task;

[0095] QoS satisfaction degree information of computing tasks;

[0096] Resource load balancing degree, the resource load balancing degree includes: at least one of computing resource load balancing degree and communication resource load balancing degree.

[0097] Optionally, the QoS satisfaction degree information of the computing task includes: QoS satisfaction degree information corresponding to a single computing task, and at least one of QoS satisfaction degree information corresponding to at least two computing tasks;

[0098] And / or, the computing resource load balance refers to: the balance degree of computing resource utilization of each network element within the first network device and the connected network element; the connected network element refers to the network element that is communicatively connected to the first network device;

[0099] And / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0100] Optionally, also include:

[0101] A second receiving module, configured to receive a statistical request sent by the second network device before obtaining the computing requirement information and resource status information corresponding to at least one cell;

[0102] The obtaining of computing demand information and resource status information corresponding to at least one cell includes:

[0103] According to the statistical request, obtaining computing demand information and resource status information corresponding to at least one cell;

[0104] The statistical request carries: the first granularity, the sample extraction method, an identifier of the information to be obtained, a calculation task number, a user identifier, and at least one of a cell identifier.

[0105] Optionally, the decision information also carries validity time information of the decision information.

[0106] The embodiment of the present invention further provides an information processing device, which is applied to a second network device, including:

[0107] A third receiving module is used to receive computing requirement information and resource status information corresponding to at least one cell sent by the first network device; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0108] A third acquisition module is used to acquire decision information for each of the cells according to the computing demand information and the resource status information; the decision information includes: computing level information and computing topology information;

[0109] The third sending module is used to send the decision information to the first network device.

[0110] Optionally, the computing demand information further includes: at least one of: the number of computing task requests, the quality of service QoS requirement information of the computing task, and the mobility requirement information of the computing task;

[0111] And / or, the communication resource status information includes: status information of the communication connection between the first network device and the connected network element, the status information includes: at least one of the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, bandwidth utilization, bit error rate or bit error rate or frame error rate of each link, success rate of establishing each link, and link interruption probability; the connected network element refers to a network element that is communicatively connected to the first network device;

[0112] And / or, the computing resource status information includes: at least one of the total amount of computing resources, computing resource utilization, allocable computing capacity, and the number of computing tasks carried of the connected network elements of the first network device;

[0113] And / or, the computing level information includes: first level information, second level information and / or third level information, the first level information includes: information of a third network device to which the cell belongs, the second level information includes: information of the first network device to which the cell belongs and information of all third network devices under the first network device to which the cell belongs, the third level information includes: information of the first network device to which the cell belongs, information of all third network devices under the first network device to which the cell belongs, and information of a multi-access edge computing MEC device adjacent to the first network device to which the cell belongs;

[0114] And / or, the computing topology information includes: information of the network device corresponding to the computing hierarchy information, the connection relationship between the network devices corresponding to the computing hierarchy information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchy information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchy information.

[0115] Optionally, also include:

[0116] a fourth receiving module, configured to receive configuration effect information sent by the first network device after sending the decision information to the first network device;

[0117] The configuration effect information includes at least one of the following:

[0118] Calculate the number of accepted requests or the acceptance rate of the task;

[0119] Calculate the number of successful executions or the execution success rate of tasks;

[0120] Calculate the interruption probability of the task;

[0121] QoS satisfaction degree information of computing tasks;

[0122] Resource load balancing degree, the resource load balancing degree includes: at least one of computing resource load balancing degree and communication resource load balancing degree.

[0123] Optionally, the QoS satisfaction degree information of the computing task includes: QoS satisfaction degree information corresponding to a single computing task, and at least one of QoS satisfaction degree information corresponding to at least two computing tasks;

[0124] And / or, the computing resource load balance refers to: the balance degree of computing resource utilization of each network element within the first network device and the connected network element; the connected network element refers to the network element that is communicatively connected to the first network device;

[0125] And / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0126] Optionally, also include:

[0127] a fourth sending module, configured to send a statistical request to the first network device before receiving the computing requirement information and resource status information corresponding to at least one cell sent by the first network device;

[0128] The receiving computing requirement information and resource status information corresponding to at least one cell sent by the first network device includes:

[0129] receiving computing requirement information and resource status information corresponding to at least one cell fed back by the first network device according to the statistical request;

[0130] The statistical request carries: at least one of the first granularity, the sample extraction method, the identifier of the information to be obtained, the calculation task number, the user identifier, and the cell identifier;

[0131] The first granularity refers to the granularity used by the first network device to obtain computing demand information and resource status information, and the sample extraction method refers to the sample extraction method used by the first network device to obtain computing demand information and resource status information;

[0132] The first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity;

[0133] The sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related feature attributes.

[0134] Optionally, also include:

[0135] A first processing module, used for marking corresponding timestamps for the computing requirement information, resource status information and configuration effect information respectively;

[0136] The acquiring decision information for each of the cells according to the computing demand information and the resource status information includes:

[0137] Decision information for each of the cells is obtained according to the computing requirement information, resource status information and configuration effect information marked with a timestamp.

[0138] Optionally, acquiring decision information for each of the cells according to the computing requirement information and the resource status information includes:

[0139] Obtaining information on the degree of change corresponding to the computing demand information and the resource status information;

[0140] When the change degree information is greater than or equal to a threshold, obtaining decision information for each of the cells according to the computing demand information and the resource status information;

[0141] When the change degree information is less than a threshold value, the decision information obtained last time is used as the decision information for this time.

[0142] Optionally, the decision information also carries validity time information of the decision information.

[0143] An embodiment of the present invention further provides an information processing device, the information processing device being a first network device, comprising: a processor and a transceiver;

[0144] The processor is used to obtain computing requirement information and resource status information corresponding to at least one cell; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0145] Sending the computing requirement information and resource status information to the second network device via the transceiver;

[0146] Receiving, through the transceiver, decision information for each of the cells fed back by the second network device; the decision information includes: computing level information and computing topology information;

[0147] Determine resource configuration information of the cell according to the decision information.

[0148] Optionally, the obtaining computing requirement information and resource status information corresponding to at least one cell includes:

[0149] Using a first granularity and a sample extraction method, obtaining computing demand information and resource status information corresponding to at least one cell;

[0150] The first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity;

[0151] The sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related feature attributes.

[0152] Optionally, the computing demand information further includes: at least one of: the number of computing task requests, the quality of service QoS requirement information of the computing task, and the mobility requirement information of the computing task;

[0153] And / or, the communication resource status information includes: status information of the communication connection between the first network device and the connected network element, the status information includes: at least one of the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, bandwidth utilization, bit error rate or bit error rate or frame error rate of each link, success rate of establishing each link, and link interruption probability; the connected network element refers to a network element that is communicatively connected to the first network device;

[0154] And / or, the computing resource status information includes: at least one of the total amount of computing resources, computing resource utilization, allocable computing capacity, and the number of computing tasks carried of the connected network elements of the first network device;

[0155] And / or, the computing level information includes: first level information, second level information and / or third level information, the first level information includes: information of a third network device to which the cell belongs, the second level information includes: information of the first network device to which the cell belongs and information of all third network devices under the first network device to which the cell belongs, the third level information includes: information of the first network device to which the cell belongs, information of all third network devices under the first network device to which the cell belongs, and information of a multi-access edge computing MEC device adjacent to the first network device to which the cell belongs;

[0156] And / or, the computing topology information includes: information of the network device corresponding to the computing hierarchy information, the connection relationship between the network devices corresponding to the computing hierarchy information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchy information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchy information.

[0157] Optionally, the processor is further configured to:

[0158] After receiving the decision information for each of the cells fed back by the second network device, acquiring configuration effect information for the decision information;

[0159] Sending the configuration effect information to the second network device through the transceiver;

[0160] The configuration effect information includes at least one of the following:

[0161] Calculate the number of requests accepted or the request acceptance rate of the task;

[0162] Calculate the number of successful executions or the execution success rate of tasks;

[0163] Calculate the interruption probability of the task;

[0164] QoS satisfaction degree information of computing tasks;

[0165] Resource load balancing degree, the resource load balancing degree includes: at least one of computing resource load balancing degree and communication resource load balancing degree.

[0166] Optionally, the QoS satisfaction degree information of the computing task includes: QoS satisfaction degree information corresponding to a single computing task, and at least one of QoS satisfaction degree information corresponding to at least two computing tasks;

[0167] And / or, the computing resource load balance refers to: the balance degree of computing resource utilization of each network element within the first network device and the connected network element; the connected network element refers to the network element that is communicatively connected to the first network device;

[0168] And / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0169] Optionally, the processor is further configured to:

[0170] Before acquiring the computing demand information and resource status information corresponding to at least one cell, receiving, by the transceiver, a statistical request sent by the second network device;

[0171] The obtaining of computing demand information and resource status information corresponding to at least one cell includes:

[0172] According to the statistical request, obtaining computing demand information and resource status information corresponding to at least one cell;

[0173] The statistical request carries: the first granularity, the sample extraction method, an identifier of the information to be obtained, a calculation task number, a user identifier, and at least one of a cell identifier.

[0174] Optionally, the decision information also carries validity time information of the decision information.

[0175] An embodiment of the present invention further provides an information processing device, which is a second network device and includes: a processor and a transceiver;

[0176] The processor is configured to receive, through the transceiver, computing requirement information and resource status information corresponding to at least one cell sent by the first network device; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0177] According to the computing demand information and resource status information, obtaining decision information for each of the cells; the decision information includes: computing level information and computing topology information;

[0178] The decision information is sent to the first network device through the transceiver.

[0179] Optionally, the computing demand information further includes: at least one of: the number of computing task requests, the quality of service QoS requirement information of the computing task, and the mobility requirement information of the computing task;

[0180] And / or, the communication resource status information includes: status information of the communication connection between the first network device and the connected network element, the status information includes: at least one of the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, bandwidth utilization, bit error rate or bit error rate or frame error rate of each link, success rate of establishing each link, and link interruption probability; the connected network element refers to a network element that is communicatively connected to the first network device;

[0181] And / or, the computing resource status information includes: at least one of the total amount of computing resources, computing resource utilization, allocable computing capacity, and the number of computing tasks carried of the connected network elements of the first network device;

[0182] And / or, the computing level information includes: first level information, second level information and / or third level information, the first level information includes: information of a third network device to which the cell belongs, the second level information includes: information of the first network device to which the cell belongs and information of all third network devices under the first network device to which the cell belongs, the third level information includes: information of the first network device to which the cell belongs, information of all third network devices under the first network device to which the cell belongs, and information of a multi-access edge computing MEC device adjacent to the first network device to which the cell belongs;

[0183] And / or, the computing topology information includes: information of the network device corresponding to the computing hierarchy information, the connection relationship between the network devices corresponding to the computing hierarchy information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchy information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchy information.

[0184] Optionally, the processor is further configured to:

[0185] After sending the decision information to the first network device, receiving, by the transceiver, configuration effect information sent by the first network device;

[0186] The configuration effect information includes at least one of the following:

[0187] Calculate the number of accepted requests or the acceptance rate of the task;

[0188] Calculate the number of successful executions or the execution success rate of tasks;

[0189] Calculate the interruption probability of the task;

[0190] QoS satisfaction degree information of computing tasks;

[0191] Resource load balancing degree, the resource load balancing degree includes: at least one of computing resource load balancing degree and communication resource load balancing degree.

[0192] Optionally, the QoS satisfaction degree information of the computing task includes: QoS satisfaction degree information corresponding to a single computing task, and at least one of QoS satisfaction degree information corresponding to at least two computing tasks;

[0193] And / or, the computing resource load balance refers to: the balance degree of computing resource utilization of each network element within the first network device and the connected network element; the connected network element refers to the network element that is communicatively connected to the first network device;

[0194] And / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0195] Optionally, the processor is further configured to:

[0196] Before receiving the computing requirement information and resource status information corresponding to at least one cell sent by the first network device, sending a statistics request to the first network device through the transceiver;

[0197] The receiving computing requirement information and resource status information corresponding to at least one cell sent by the first network device includes:

[0198] receiving computing requirement information and resource status information corresponding to at least one cell fed back by the first network device according to the statistical request;

[0199] The statistical request carries: at least one of the first granularity, the sample extraction method, the identifier of the information to be obtained, the calculation task number, the user identifier, and the cell identifier;

[0200] The first granularity refers to the granularity used by the first network device to obtain computing demand information and resource status information, and the sample extraction method refers to the sample extraction method used by the first network device to obtain computing demand information and resource status information;

[0201] The first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity;

[0202] The sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related feature attributes.

[0203] Optionally, the processor is further configured to:

[0204] Marking corresponding timestamps for the computing requirement information, resource status information, and configuration effect information respectively;

[0205] The acquiring decision information for each of the cells according to the computing demand information and the resource status information includes:

[0206] Decision information for each of the cells is obtained according to the computing requirement information, resource status information and configuration effect information marked with a timestamp.

[0207] Optionally, acquiring decision information for each of the cells according to the computing requirement information and the resource status information includes:

[0208] Obtaining information on the degree of change corresponding to the computing demand information and the resource status information;

[0209] When the change degree information is greater than or equal to a threshold, obtaining decision information for each of the cells according to the computing demand information and the resource status information;

[0210] When the change degree information is less than a threshold value, the decision information obtained last time is used as the decision information for this time.

[0211] Optionally, the decision information also carries validity time information of the decision information.

[0212] An embodiment of the present invention also provides an information processing device, including a memory, a processor, and a program stored in the memory and executable on the processor; when the processor executes the program, the information processing method on the first network device side or the second network device side is implemented.

[0213] An embodiment of the present invention further provides a readable storage medium on which a program is stored. When the program is executed by a processor, the steps in the information processing method on the first network device side or the second network device side are implemented.

[0214] The beneficial effects of the above technical solution of the present invention are as follows:

[0215] In the above scheme, the information processing method obtains computing demand information and resource status information corresponding to at least one cell; the computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; sends the computing demand information and resource status information to the second network device; receives decision information for each of the cells fed back by the second network device; the decision information includes: computing level information and computing topology information; determines the resource configuration information of the cell based on the decision information; can support the implementation of resource configuration by integrating communication-related information and computing-related information, improve resource configuration accuracy, and solve the problem of poor resource configuration accuracy in the existing technology for computing resource scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0216] Figure 1 The information processing method of the embodiment of the present invention is shown in FIG. Figure 1 ;

[0217] Figure 2 The information processing method of the embodiment of the present invention is shown in FIG. Figure 2 ;

[0218] Figure 3 It is a schematic diagram of the overall framework of the information processing method according to an embodiment of the present invention;

[0219] Figure 4 The information processing method of the embodiment of the present invention is specifically implemented as follows: Figure 1 ;

[0220] Figure 5 The information processing method of the embodiment of the present invention is specifically implemented as follows: Figure 2 ;

[0221] Figure 6 The information processing method of the embodiment of the present invention is specifically implemented as follows: Figure 3 ;

[0222] Figure 7 The information processing method of the embodiment of the present invention is specifically implemented as follows: Figure 4 ;

[0223] Figure 8 A schematic diagram of a topological connection relationship in an embodiment of the present invention;

[0224] Fig. 9 Detailed topology information of an embodiment of the present invention is shown in FIG. Figure 1 ;

[0225] Fig.10 Detailed topology information of an embodiment of the present invention is shown in FIG. Figure 2 ;

[0226] Fig.11 The structure of the information processing device according to the embodiment of the present invention is shown in FIG. Figure 1 ;

[0227] Fig.12 The structure of the information processing device according to the embodiment of the present invention is shown in FIG. Figure 2 ;

[0228] Fig.13 The structure of the information processing device according to the embodiment of the present invention is shown in FIG. Figure 1 ;

[0229] Fig.14 The structure of the information processing device according to the embodiment of the present invention is shown in FIG. Figure 2 . DETAILED DESCRIPTION

[0230] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0231] The following first introduces the relevant contents of this plan.

[0232] At present, the computing resource allocation method adopted by the industry is usually based on computing tasks, fairness or priority, so as to ensure that the computing resources of each node in the network are fully utilized and all tasks in the network can be completed fairly. At present, the more commonly used technical solutions in the industry include the fair scheduler and capacity scheduler in the YARN (Yet Another Resource Negotiator) scheduler. Specifically:

[0233] The fair scheduler is a multi-user, multi-job scheduler. User tasks are assigned to multiple resource pools. Each resource pool sets the minimum guarantee and maximum limit for resource allocation. Administrators can specify the priority of resource pools. During scheduling, some resources are allocated according to the minimum guarantee of each resource pool. The remaining resources are then allocated to each resource pool in proportion according to the specified priority of the resource pool. Finally, in each resource pool, resources are allocated to each job according to the job priority or fairness policy.

[0234] The capacity scheduler is also a multi-user, multi-job scheduler. Compared with the fair scheduler, it emphasizes the fairness of resources between users rather than between jobs. It divides users into multiple queues, and sets the minimum guarantee and usage limit of resource allocation for each queue. When a queue has surplus resources, it can temporarily share them with other queues. When scheduling, resources are allocated to the queue with the highest resource utilization rate first, and the queue is scheduled according to FIFO (first in, first out).

[0235] However, the fair scheduler and capacity scheduler have the following disadvantages:

[0236] (1) Disadvantages of the fair scheduler: The fair scheduler ensures fairness among tasks by setting up resource pools during scheduling, but does not allow excess resources to be scheduled between resource pools. It adopts a hard isolation approach, which leads to a lack of flexibility in resource scheduling and a certain loss in the utilization of network resources.

[0237] (2) Disadvantages of the capacity scheduler: The capacity scheduler ensures fairness by establishing queues for users during scheduling. However, its user-level construction concept will impose a heavy burden on the network when too many users access the network. It is also difficult to deeply couple with communications, making it inconvenient to use in a network with integrated computing.

[0238] Both of the above solutions have certain defects in computing resource allocation, and both are proposed for computing resources without considering communication-related statistics, making it difficult to directly apply them to 6G computing fusion networks.

[0239] Based on the above, the present invention aims at the problem of poor resource configuration accuracy in the existing technology for computing resource scheduling, and provides an information processing method applied to a first network device, such as Figure 1 As shown, including:

[0240] Step 11: Obtain computing requirement information and resource status information corresponding to at least one cell; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0241] Step 12: Sending the computing requirement information and resource status information to the second network device;

[0242] Step 13: receiving decision information for each of the cells fed back by the second network device; the decision information includes: calculation level information and calculation topology information;

[0243] Step 14: Determine resource configuration information of the cell according to the decision information.

[0244] Among them, the first network device can be implemented as a centralized unit CU, and / or, the first network device can be implemented as an AIMO (artificial intelligence AI management and orchestration) entity, and / or, step 12 can also include: sending target configuration effect information, that is, the configuration effect information corresponding to the decision information corresponding to the computing requirement information and resource status information sent previously, and / or, the resource configuration information may include: the resource range (pre-configured resources) information available to the cell, and / or, the decision information can be obtained by the second network device using a neural network algorithm, and / or, the resource allocation of the cell can be performed by the first network device; but it is not limited to this.

[0245] The information processing method provided in the embodiment of the present invention obtains computing demand information and resource status information corresponding to at least one cell; the computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; sends the computing demand information and resource status information to a second network device; receives decision information for each of the cells fed back by the second network device; the decision information includes: computing level information and computing topology information; determines resource configuration information of the cell based on the decision information; can support the implementation of resource configuration by integrating communication-related information and computing-related information, improve resource configuration accuracy, and solve the problem of poor resource configuration accuracy in the prior art for computing resource scheduling solutions.

[0246] Among them, the obtaining of computing demand information and resource status information corresponding to at least one cell includes: adopting a first granularity and a sample extraction method to obtain the computing demand information and resource status information corresponding to at least one cell; wherein, the first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity; the sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and at least one of extraction by computing task-related characteristic attributes.

[0247] This allows accurate acquisition of computing demand information and resource status information.

[0248] Further, the computing demand information also includes: at least one of the number of computing task requests, the quality of service QoS requirement information of the computing tasks, and the mobility requirement information of the computing tasks; and / or, the communication resource status information includes: the status information of the communication connection between the first network device and the connected network element, and the status information includes: the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, the bandwidth utilization, the bit error rate or bit error rate or frame error rate of each link, the success rate of each link establishment, and at least one of the link interruption probability; the connected network element refers to the network element that is communicatively connected to the first network device; and / or, the computing resource status information includes: the total amount of computing resources of the connected network elements of the first network device, the computing resource utilization, the allocable computing capacity, and at least one of the number of computing tasks carried; and / or, the computing level information includes: the first level information, the second level information hierarchical information and / or third-level information, the first-level information includes: information about the third network device to which the cell belongs, the second-level information includes: information about the first network device to which the cell belongs and information about all third network devices under the first network device to which the cell belongs, the third-level information includes: information about the first network device to which the cell belongs, information about all third network devices under the first network device to which the cell belongs, and information about the multi-access edge computing MEC device adjacent to the first network device to which the cell belongs; and / or, the computing topology information includes: information about the network device corresponding to the computing hierarchical information, the connection relationship between the network devices corresponding to the computing hierarchical information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchical information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchical information.

[0249] In this way, computing demand information, communication resource status information, computing resource status information, computing level information and / or computing topology information can be clarified. The third network device can be implemented as a distributed unit DU, but is not limited thereto.

[0250] Furthermore, after receiving the decision information for each of the cells fed back by the second network device, it also includes: obtaining configuration effect information for the decision information; sending the configuration effect information to the second network device; wherein the configuration effect information includes at least one of the following: the number of accepted requests or the request acceptance rate of the computing task; the number of successful executions or the execution success rate of the computing task; the interruption probability of the computing task; the QoS satisfaction degree information of the computing task; and the resource load balancing degree, wherein the resource load balancing degree includes: at least one of the computing resource load balancing degree and the communication resource load balancing degree.

[0251] This can assist the second network device to accurately obtain the next round of decision information and update the relevant algorithm. Wherein, the obtaining of configuration effect information for the decision information may include: obtaining configuration effect information for the decision information according to the first granularity, but is not limited thereto.

[0252] Among them, the QoS satisfaction degree information of the computing task includes: the QoS satisfaction degree information corresponding to a single computing task, and at least one item of the QoS satisfaction degree information corresponding to at least two computing tasks; and / or, the computing resource load balancing degree refers to: the balance degree of computing resource utilization of each network element within the scope of the first network device and the connected network elements; the connected network elements refer to the network elements that are communicatively connected to the first network device; and / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0253] In this way, QoS satisfaction information, computing resource load balance and / or communication resource load balance can be clarified.

[0254] Furthermore, before obtaining the computing requirement information and resource status information corresponding to at least one cell, it also includes: receiving a statistical request sent by the second network device; the obtaining of the computing requirement information and resource status information corresponding to at least one cell includes: obtaining the computing requirement information and resource status information corresponding to at least one cell according to the statistical request; wherein the statistical request carries: at least one of the first granularity, the sample extraction method, the identification of the information to be obtained, the computing task number, the user identification and the cell identification.

[0255] In this way, relevant information can be collected and reported according to the needs of the second network device.

[0256] The decision information also carries the validity time information of the decision information.

[0257] This can ensure the timeliness and effectiveness of related processing operations.

[0258] The embodiment of the present invention also provides an information processing method, which is applied to a second network device, such as Figure 2 As shown, including:

[0259] Step 21: receiving computing requirement information and resource status information corresponding to at least one cell sent by the first network device; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0260] Step 22: acquiring decision information for each of the cells according to the computing demand information and the resource status information; the decision information includes: computing level information and computing topology information;

[0261] Step 23: Send the decision information to the first network device.

[0262] Wherein, when obtaining the decision information for each of the cells, it may be the computing demand information and resource status information sorted out at the cell granularity; but it is not limited thereto.

[0263] The information processing method provided by the embodiment of the present invention receives computing demand information and resource status information corresponding to at least one cell sent by a first network device; the computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; according to the computing demand information and resource status information, decision information for each of the cells is obtained; the decision information includes: computing level information and computing topology information; the decision information is sent to the first network device; it can support the implementation of integrated communication-related information and computing-related information to perform resource configuration, improve resource configuration accuracy, and solve the problem of poor resource configuration accuracy in the existing technology for computing resource scheduling solutions.

[0264] Wherein, the computing demand information also includes: at least one of the number of computing task requests, the service quality QoS requirement information of the computing task and the mobility requirement information of the computing task; and / or, the communication resource status information includes: the status information of the communication connection between the first network device and the connected network element, and the status information includes: the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, the bandwidth utilization, the bit error rate or bit error rate or frame error rate of each link, the success rate of each link establishment, and at least one of the link interruption probability; the connected network element refers to the network element that is communicatively connected to the first network device; and / or, the computing resource status information includes: the total amount of computing resources of the connected network elements of the first network device, the computing resource utilization, the allocable computing capacity and at least one of the number of computing tasks carried; and / or, the computing level information includes: the first level information, the second level information The first-level information and / or third-level information, the first-level information includes: information about the third network device to which the cell belongs, the second-level information includes: information about the first network device to which the cell belongs and information about all third network devices under the first network device to which the cell belongs, the third-level information includes: information about the first network device to which the cell belongs, information about all third network devices under the first network device to which the cell belongs, and information about the multi-access edge computing MEC device adjacent to the first network device to which the cell belongs; and / or, the computing topology information includes: information about the network device corresponding to the computing level information, connection relationships between the network devices corresponding to the computing level information, a maximum computing capacity allocated to the cell for the network device corresponding to the computing level information, and at least one of a maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to a link between the network devices corresponding to the computing level information.

[0265] In this way, computing demand information, communication resource status information, computing resource status information, computing level information and / or computing topology information can be clarified. The third network device can be implemented as a distributed unit DU, but is not limited thereto.

[0266] Furthermore, after sending the decision information to the first network device, it also includes: receiving configuration effect information sent by the first network device; wherein the configuration effect information includes at least one of the following: the number of accepted requests or the request acceptance rate of the computing task; the number of successful executions or the execution success rate of the computing task; the interruption probability of the computing task; the QoS satisfaction degree information of the computing task; and the resource load balancing degree, wherein the resource load balancing degree includes: at least one of the computing resource load balancing degree and the communication resource load balancing degree.

[0267] This can assist the second network device to accurately obtain the next round of decision information and update the relevant algorithm. This solution may also include: updating the algorithm for obtaining decision information (such as the following neural network algorithm) according to the configuration effect information, but is not limited to this.

[0268] Among them, the QoS satisfaction degree information of the computing task includes: the QoS satisfaction degree information corresponding to a single computing task, and at least one item of the QoS satisfaction degree information corresponding to at least two computing tasks; and / or, the computing resource load balancing degree refers to: the balance degree of computing resource utilization of each network element within the scope of the first network device and the connected network elements; the connected network elements refer to the network elements that are communicatively connected to the first network device; and / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0269] In this way, QoS satisfaction information, computing resource load balance and / or communication resource load balance can be clarified.

[0270] Furthermore, before receiving the computing demand information and resource status information corresponding to at least one cell sent by the first network device, it also includes: sending a statistical request to the first network device; the receiving of the computing demand information and resource status information corresponding to at least one cell sent by the first network device includes: receiving the computing demand information and resource status information corresponding to at least one cell fed back by the first network device according to the statistical request; wherein the statistical request carries: a first granularity, a sample extraction method, an identifier of the information to be obtained, a computing task number, a user identifier, and at least one item of the cell identifier; the first granularity refers to the granularity used by the first network device to obtain the computing demand information and resource status information, and the sample extraction method refers to the sample extraction method used by the first network device to obtain the computing demand information and resource status information; the first granularity includes: at least one item of task granularity, user granularity, cell granularity, and network element granularity; the sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related characteristic attributes.

[0271] This can support the first network device to collect and report relevant information according to the needs of the second network device. The sending of the statistical request to the first network device may include: sending the statistical request to the first network device regularly (periodically) or according to the demand trigger information, which is not limited here.

[0272] In an embodiment of the present invention, the information processing method further includes: marking corresponding timestamps for the computing requirement information, resource status information and configuration effect information respectively; obtaining decision information for each of the cells based on the computing requirement information and resource status information, including: obtaining decision information for each of the cells based on the computing requirement information, resource status information and configuration effect information after marking the timestamp; specifically may include: using a neural network algorithm to obtain decision information for each of the cells based on the computing requirement information, resource status information and configuration effect information after marking the timestamp, but is not limited to this, for example, a non-neural network algorithm may also be used to obtain decision information for each of the cells based on the computing requirement information, resource status information and configuration effect information after marking the timestamp; or according to rules, etc., the decision information for each of the cells may be obtained based on the computing requirement information, resource status information and configuration effect information after marking the timestamp.

[0273] In this way, the computing requirement information, resource status information and configuration effect information of different batches (or different rounds) can be accurately distinguished, so as to accurately obtain the decision information of each of the cells; wherein, in "obtaining the decision information for each of the cells according to the computing requirement information, resource status information and configuration effect information after marking the timestamp", the "computing requirement information and resource status information" used may be the information of the current batch, and the configuration effect information used may be the feedback information of the decision information of the previous batch (such as the previous batch or the previous round), and the decision information of the previous batch is obtained according to the "computing requirement information and resource status information" of the previous batch. In this scheme, the information processing method may also include: updating the neural network algorithm according to the configuration effect information and target decision information after marking the timestamp; wherein, the target decision information is obtained according to the computing requirement information and resource status information whose timestamp matches the configuration effect information (that is, the computing requirement information and resource status information belonging to the same batch or the same round as the configuration effect information).

[0274] Among them, obtaining decision information for each of the cells based on the computing demand information and the resource status information includes: obtaining change degree information corresponding to the computing demand information and the resource status information; when the change degree information is greater than or equal to a threshold, obtaining decision information for each of the cells based on the computing demand information and the resource status information; when the change degree information is less than a threshold, using the decision information obtained last time as the decision information this time.

[0275] In this way, processing resources can be saved while ensuring accurate acquisition of decision information. The change degree information may refer to the change degree of the current batch information relative to the previous batch information (such as the change degree of the current batch of computing demand information relative to the previous batch of computing demand information), but is not limited to this. "When the change degree information is greater than or equal to a threshold, the decision information for each of the cells is obtained based on the computing demand information and resource status information", which may specifically include: when the change degree information is greater than or equal to a threshold, a neural network algorithm is used to obtain the decision information for each of the cells based on the computing demand information and resource status information, but is not limited to this.

[0276] In the embodiment of the present invention, the decision information also carries the validity time information of the decision information.

[0277] This can ensure the timeliness and effectiveness of related processing operations.

[0278] It is to be noted that the related contents of the first network device side and the second network device side mentioned above can refer to each other, and the repeated contents will not be repeated.

[0279] The information processing method provided in an embodiment of the present invention is illustrated below by taking an AIMO (artificial intelligence AI management and orchestration) entity deployed on a southbound network management device OMC-R as an example (the AIMO entity can also be deployed on other devices or set up independently), the second network device takes CU as an example, and the third network device takes DU as an example.

[0280] In response to the above technical problems, an embodiment of the present invention provides an information processing method, which can be specifically implemented as a method for determining the computing capacity level of a cell in a wireless network. The method is oriented to the new 6G computing fusion network, and proposes a resource scheduling solution that completes computing resource allocation by dividing the computing level in units of cells; specifically, it can also be understood that this solution proposes a resource configuration solution that integrates communication and computing task scheduling to complete the computing resource allocation of the cell by dividing the computing level. Among them, the solution mainly includes three parts: AIMO interacts with CU to obtain relevant statistics (corresponding to the above computing demand information and resource status information), AIMO determines the computing level and computing topology for each cell under CU (corresponding to the above decision information), and AIMO returns the computing level and computing topology to CU, and finally optimizes the scheduling of resources. The interaction between AIMO and CU can be as follows: Figure 3 As shown, it includes: AIMO sends a statistics request to CU (corresponding to the above statistics request), CU reports statistics accordingly, and AIMO sends the cell computing capacity level and topology configuration (i.e. computing level and computing topology) to CU. Among them, CRM represents computing resource management entity; Figure 3 The OMC-R is only a possible example of AIMO deployment. The actual deployment location of AIMO may have many possibilities, which may be a new network element (that is, AIMO can also be deployed separately), or a centralized point of a network domain, which is not limited here.

[0281] Specifically, AIMO can sample or read relevant statistics from the CU (including local statistics of the CU or statistics obtained by the CU from other network elements (such as DU)), and then determine the computing level and computing topology of each cell under the CU through algorithms (such as neural network algorithms such as reinforcement learning), and AIMO allocates the computing level and computing topology to the CU. For example: when the reinforcement learning algorithm is used, the important parameters of reinforcement learning include the state space S, the action space A and the reward function R; S and R are respectively composed of some statistics reported from the CU, and A is composed of the computing level and topological decision results (i.e., computing topology) of one or more cells. A is obtained by AIMO based on S and R (part of the output of R will be given to S for algorithm calculation).

[0282] The following is an introduction to each part of this plan.

[0283] [Part 1] Statistics, configuration information and interaction processes transmitted between CU and AIMO;

[0284] The statistics include computing demand statistics (corresponding to the computing demand information), resource status statistics (corresponding to the resource status information) and configuration effect statistics (corresponding to the configuration effect information). The statistical granularity can be task level, user level, cell level, network element level (corresponding to the first granularity including at least one of task granularity, user granularity, cell granularity and network element granularity), and the sample extraction method can be according to the task number, the computing task type, the specified time period, the computing task related characteristic attributes (such as the computing resource type used, etc.) or the above rules (corresponding to the above extraction method including at least one of the extraction method according to the task number, the extraction method according to the computing task type, the extraction method according to the specified time period, and the extraction method according to the computing task related characteristic attributes).

[0285] The configuration information (corresponding to the above-mentioned decision information) includes the calculation level and topology decision results (corresponding to the above-mentioned calculation level information and calculation topology information) of one or more cells.

[0286] [Part 1-1] Computing demand statistics reported by CU to AIMO;

[0287] The calculation of the demand statistics may include at least one of the following:

[0288] (1) The number of computing task requests; this number can be reflected by the type of computing task and the number of accesses (of each type), for example: the number of computing task requests = the type of computing task × the number of accesses (i.e., the number of computing task requests of all computing task types), or the number of computing task requests = the number of accesses of one type of computing task (i.e., the number of computing task requests of each computing task type), but is not limited thereto;

[0289] (2) Resource requirements of computing tasks, including communication resources (corresponding to the above-mentioned communication resource requirement information) and computing resources (corresponding to the above-mentioned computing resource requirement information); specifically, it may be the total amount of resources (for example, the total amount of communication resources and computing resources corresponding to all tasks of a cell within a period of time), or it may be information such as a computing task scheduling diagram that can be used to infer the requirements for communication and computing resources. This information can support the calculation of the resource requirements for a computing task, but is not limited to this.

[0290] (3) QoS requirements of computing tasks (including privacy requirements, computing accuracy requirements, latency requirements, continuity requirements, etc.), corresponding to the above QoS requirement information;

[0291] (4) Mobility requirements of computing tasks (including mobile computing, non-mobile computing, or estimation of mobile speed, etc.), corresponding to the above mobility requirement information;

[0292] Among them, task persistence is defined as: the time degree of the network actually calculating the task during the task calculation process. For example, it can be measured by the ratio of the non-interruption time (that is, the actual calculation time after excluding the interruption time) to the total calculation time.

[0293] [Part 1-2] Resource status statistics reported by CU to AIMO;

[0294] The resource status statistics may include at least one of the following:

[0295] (1) computing resource status statistics (corresponding to the computing resource status information): at least one of the total amount of computing resources, computing resource utilization, allocable computing capacity, and the number of computing tasks carried by the network elements connected to the CU (including the CU, neighboring DUs, neighboring CUs, and neighboring MECs (mobile edge computing or multi-access edge computing)); corresponding to the computing resource status information, including: at least one of the total amount of computing resources, computing resource utilization, allocable computing capacity, and the number of computing tasks carried by the network elements connected to the first network device;

[0296] (2) Communication resource status statistics (corresponding to the above-mentioned communication resource status information): Status statistics of the communication connection between the CU and its connected network elements (including DU, neighboring CU, and neighboring MEC) (corresponding to the status information of the communication connection between the above-mentioned first network device and the connected network elements), including: at least one of the statistics such as the total transmission bandwidth or throughput of each link, the allocable bandwidth or throughput, the bandwidth utilization, the bit error rate or bit error rate or frame error rate of each link, the success rate of each link establishment, and the probability of link interruption; corresponding to the above-mentioned status information, including: at least one of the statistics such as the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, the bandwidth utilization, the bit error rate or bit error rate or frame error rate of each link, the success rate of each link establishment, and the probability of link interruption.

[0297] [Part 1-3] Configuration effect statistics reported by CU to AIMO (the statistics may be feedback on the previous round of configuration information);

[0298] Specifically, the statistic is statistically measured after the CU completes the calculation level (ie, the cell calculation capacity level) and topology configuration according to the decision information of AIMO; the statistic may include at least one of the following:

[0299] (1) Calculate the number of task requests accepted or the acceptance rate, the number of successful executions or the success rate, and the interruption probability;

[0300] (2) QoS satisfaction degree of computing tasks (corresponding to the above-mentioned QoS satisfaction degree information), including: QoS satisfaction degree of a single computing task (corresponding to the above-mentioned QoS satisfaction degree information corresponding to the single computing task), and multi-task statistics based on task QoS satisfaction degree (corresponding to the above-mentioned QoS satisfaction degree information corresponding to at least two computing tasks);

[0301] (3) Resource load balance, including computing resource load balance, communication resource load balance, etc.

[0302] Among them, the computing resource load balance is defined as: the balance degree of computing resource utilization of each network element within the CU and the network elements connected to the CU (including the CU, neighboring DU, neighboring CU, and neighboring MEC), for example, it can be measured by the statistical variance value. The smaller the variance, the more balanced the load distribution; corresponding to the above computing resource load balance refers to: the balance degree of computing resource utilization of each network element within the first network device and the connected network element; the connected network element refers to the network element that is communicatively connected to the first network device;

[0303] The communication resource load balancing degree is defined as: the degree of balance of resource utilization of each communication link within the CU and the network elements connected to the CU (including the CU, neighboring DU, neighboring CU, and neighboring MEC). For example, it can be measured by the statistical variance value. The smaller the variance, the more balanced the load distribution. The corresponding communication resource load balancing degree refers to: the degree of balance of resource utilization of each communication link within the first network device and the connected network elements.

[0304] [Part 1-4] Configuration information sent by AIMO to CU (corresponding to the above-mentioned decision information for each of the cells);

[0305] The configuration information may include: a list of configured cell IDs (identifications), computing level information corresponding to each cell, and computing topology decision results (corresponding to the above computing topology information);

[0306] The computing level information may include: levels A, B and C (corresponding to the above computing level information including: first level information, second level information and / or third level information). Level A includes the DU to which the cell belongs (corresponding to the above first level information including: information of the third network device to which the cell belongs), level B includes the CU to which the cell belongs and all DUs under it (corresponding to the above second level information including: information of the first network device to which the cell belongs and information of all third network devices under the first network device to which the cell belongs), and level C includes the CU to which the cell belongs and all DUs under it and the MEC node adjacent to the CU (corresponding to the above third level information including: information of the first network device to which the cell belongs, information of all third network devices under the first network device to which the cell belongs, and information of the multi-access edge computing MEC device adjacent to the first network device to which the cell belongs).

[0307] The calculation topology decision result may include: the network element nodes allocated within the above-mentioned hierarchy, the connection relationship between the nodes, and may also include the maximum computing capacity allocated by the node to the cell and the maximum communication bandwidth allocated by the link between nodes to the cell; the corresponding calculation topology information includes: the information of the network device corresponding to the calculation hierarchy information, the connection relationship between the network devices corresponding to the calculation hierarchy information, the maximum computing capacity allocated to the cell for the network device corresponding to the calculation hierarchy information, and at least one of the maximum communication bandwidth allocated to the cell for the link between nodes.

[0308] It is explained here that the above parts 1-1 and 1-2 can be the information sent to AIMO in this round, which is used to obtain configuration information; part 1-3 can be feedback information for the configuration information received in the previous round (such as the previous round), corresponding to R calculation, which can be used to update the algorithm of AIMO (used to obtain configuration information), etc.; correspondingly, in the information sent to AIMO this time, the timestamp of part 1-3 is different from the timestamp of parts 1-1 and 1-2.

[0309] [Part 1-5] Interaction process, messages and content between AIMO and CU;

[0310] Specific examples include Figures 4 to 6 As shown, the statistics acquisition process is divided into: resource status process, computing task status process (corresponding to computing demand statistics) and configuration and effect feedback (corresponding to configuration effect statistics) process. These three processes can be integrated and executed (for example, the resource status statistics request and the computing task status statistics request are sent together, and the feedback of these two requests is received together) or executed separately and independently, which is not limited here. Among them, the above-mentioned statistical request may include an overall request, or at least two sub-requests, such as a resource status statistics request and a computing task status statistics request, but it is not limited to this.

[0311] Among them, (1) in the resource status class process, the resource status statistics request (i.e., resource status class statistics request) may include: statistical granularity (such as network element level), sample extraction method (such as according to a specified time period), statistics list (selected from the above-mentioned resource status class statistics (i.e., resource status statistics), which may include default options), network element ID list (which may correspond to the above-mentioned network element connected to the CU), resource type (one or more of computing resources and communication resources), etc. At least one of the following information; the statistics list may correspond to the identification of the above-mentioned information to be obtained; but it is not limited to this.

[0312] Resource status statistics feedback (i.e. resource status statistics feedback): includes corresponding statistics and values ​​as required in the request message.

[0313] (2) In the computing task status process, the computing task status statistics request may include: statistical granularity (such as task level, user level or cell level), sample extraction method (such as by task number, by computing task type, by specified time period, by computing task related characteristic attributes or a combination of the above rules), statistics list (selected from the above computing task statistics (i.e., computing demand statistics), which may include default options), task number list (corresponding to the above computing task number), user ID list (corresponding to the above user identifier), cell ID list (corresponding to the above cell identifier) ​​and other information. The statistics list may correspond to the identifier of the above information to be obtained; but is not limited to this.

[0314] Compute task status statistics feedback: Contains the corresponding statistics and values ​​as required in the request message.

[0315] (3) In the configuration and effect feedback process, the calculation level and topology configuration may include: a list of configured cell IDs, corresponding calculation level information and calculation topology decision results, a list of state statistics (selected from the above-mentioned state class statistics (corresponding to the above-mentioned configuration effect statistics) and may include default options), statistical granularity, sample extraction method and other information.

[0316] New status and effect statistics feedback (i.e. configuration effect statistics): contains corresponding statistics and values ​​according to the status statistics related requirements in the above message.

[0317] [Part 2] The process and method of AIMO decision-making calculation hierarchy and topology configuration, as well as the calculation and use of action validity duration (corresponding to the above validity time information) and data smoothness;

[0318] Among them, the neural network algorithm takes the reinforcement learning algorithm as an example, which involves A calculation (i.e. action calculation), S calculation (i.e. state calculation), R calculation (i.e. feedback calculation or reward calculation) and C calculation (i.e. preprocessing (or general) calculation), as shown in the following figure. Figure 7 As shown in the diagram of the relationship between reinforcement learning algorithms and processes in AIMO, the operations performed by AIMO include the following:

[0319] 1. AIMO periodically extracts and counts the statistics uploaded by CU, and performs time smoothing and classification processing in C calculation.

[0320] The time interval of the timed extraction should be much shorter than the effective duration of the action to ensure the real-time and reliability of the sample. Specifically, the timed extraction can be performed by sending a resource status statistics request and a computing task status statistics request, and receiving resource status statistics feedback and computing task status statistics feedback; but it is not limited to this.

[0321] Regarding time smoothing, for example, if the timing extraction period F1 is longer, while the interface sample generation period F2 on the CU side (such as computing demand information and resource status information) is shorter, the interface samples generated in F1 can be smoothed, such as sample averaging or weighted averaging, etc., which is not limited here.

[0322] The classification may include: resource status statistics and computing task status statistics should be stored and distinguished separately to facilitate subsequent management.

[0323] The figure does not show the "feedback of new status and effect statistics". This feedback is also given to C calculation and processed by C calculation.

[0324] 2. When the previous round of action expires and the data smoothness is higher than the threshold, C will copy the data, stamp it with different timestamps, and upload it to the current round of S calculation and the previous round of R calculation respectively; and report the time interval of this data upload to A calculation (that is, the time interval between this data upload and the previous data upload) to assist A calculation in generating the following action validity duration.

[0325] Among them, one round may include one S calculation + A calculation (corresponding to executing once the computing requirement information and resource status information corresponding to at least one cell are obtained;...determine the resource configuration information of the cell according to the decision information); the R calculation belongs to the previous round, including: feedback calculation for the last configuration to update the algorithm, and also includes: calculation of state-type statistics (such as part of the resource status-type statistics) to assist the S calculation, the state-type statistics can be statistics that can reflect the relevant status after the last round of configuration, such as: computing resource utilization and bandwidth utilization, etc.; the statistics used in the R calculation can be included in the above-mentioned computing requirement statistics and / or resource status statistics, or can be separately included in the configuration effect statistics, which is not limited here.

[0326] The expiration of the previous round of actions may refer to that the current upload time interval (ie, the time interval for uploading the data this time) is greater than the effective duration specified in the previous round of actions (ie, the effective duration of the actions corresponding to the previous round).

[0327] The data smoothness may refer to the fluctuation size of the data (i.e., statistics) after time smoothing (corresponding to the above-mentioned degree of change information). When the fluctuation is less than the threshold, the network will ignore the current round of adjustment (i.e., no algorithm reasoning will be performed this time) to save resources; corresponding to the above-mentioned case where the degree of change information is less than the threshold, the decision information obtained last time will be used as the decision information for this time. The data smoothness can be obtained by AIMO itself.

[0328] The different timestamps may mean that the data transmitted to S for calculation should be marked with the timestamp of this round, and the data transmitted to R for calculation should be marked with the timestamp of the previous round, and the timestamps are used to distinguish the rounds to which the variables belong; corresponding to the above-mentioned timestamps for marking the computing requirement information, resource status information, and configuration effect information respectively.

[0329] 3. The previous round of R calculation processes some resource status statistics and outputs feedback values, which are timestamped and sent to the current round of S calculation.

[0330] The feedback value may refer to the feedback value (such as the gap value, which may be equal to the expected value minus the actual calculated value, etc.) for the QoS satisfaction of the above-mentioned computing tasks, the network resource utilization rate, and the network load balance.

[0331] 4. This round of S calculation counts the resource status statistics, performs dimensionality reduction and normalization on the computing task status statistics, and uses the processed results as the state space S for reinforcement learning.

[0332] The statistics of resource status statistics may refer to memorizing and storing resource status statistics, so as to facilitate the network and managers to judge the changing trend and distribution of the overall network resources. In addition, since the data of computing task status statistics may be much larger than that of resource status statistics, the computing task status statistics calculated by S need to be processed by dimension reduction and normalization.

[0333] The dimension reduction and normalization processing may refer to the further operation of the computing task state statistics after time smoothing uploaded by the computing C, which can compress the data size of the computing task state statistics and improve the accuracy and representativeness of the calculation. The dimension reduction may adopt the mainstream principal component analysis (PCA) algorithm, and the normalization may also adopt the mainstream methods such as Z-Score (z score, also called standard score) normalization or linear normalization.

[0334] 5. The reinforcement learning result (i.e., the inference result output by the reinforcement learning model) is output to A calculation. After A calculation converts the result format, it adds the effective duration of the action and sends it to CU for execution; that is, AIMO sends the computing level and topology (configuration) and the effective time (i.e., the effective duration of the action, which can be understood as the effective duration of the computing level and topology, but is not limited to this) to CU.

[0335] The effective duration of the action may refer to the duration of the calculation level in this action. Usually, the duration is the same as the C calculation upload time interval (ie, the time interval for AIMO to extract statistics regularly).

[0336] The converting of the result form may refer to expressing the topological connection relationship and the topological specific information into a specific form that is easy for CU to identify, such as expressing the topological connection relationship as Figure 8 The topological relationship matrix shown (1 represents connection, 0 represents no connection) represents the topological specific information as Fig. 9 The node information representation shown and Fig.10 The link information is shown in the form of representation, but is not limited to this.

[0337] Among them, the algorithm used by AIMO is embodied as follows Figure 7 The reinforcement learning model in , the use of algorithms to obtain computing levels and topologies can be understood as reinforcement learning model reasoning; the update of the algorithm can be understood as reinforcement learning model training.

[0338] The following is a detailed introduction to the reinforcement learning algorithm used by AIMO.

[0339] The S, A, and R involved in the reinforcement learning algorithm include the following statistics and decision quantities respectively:

[0340] S (status): includes two types of statistics: resource status and computing task status. The resource status category may include: the total bandwidth of the CU and its connected network elements (corresponding to the total transmission bandwidth of each link mentioned above), throughput (corresponding to the allocable throughput mentioned above), resource utilization (corresponding to the bandwidth utilization mentioned above), bit error rate or bit error rate or frame error rate of each link, and the load balance degree for the previous round of feedback (corresponding to the resource load balance degree mentioned above); the computing task status category may include: the type of computing task, the number of access times, the amount of computing resources applied for (corresponding to the resource requirements of the above computing tasks), latency requirements, privacy requirements, (computing) accuracy requirements, continuity requirements, and the degree of satisfaction of the computing task QoS for the previous round of feedback.

[0341] A (Action): includes the computing hierarchy of each cell and the computing topology decision of the network element nodes (the topology decision should include the network element nodes allocated within the above hierarchy, the connection relationship between the nodes, and may also include the maximum computing capacity allocated by the node for the cell and the maximum communication bandwidth allocated by the link between nodes for the cell).

[0342] R (reward): includes the QoS satisfaction of computing tasks, network resource utilization and network load balance (corresponding to the resource load balance mentioned above) and the priority setting between these three types of statistics. The statistics included in the QoS satisfaction of computing tasks are: privacy, computing accuracy, latency, and continuity; the statistics included in the network resource utilization are: computing resource utilization and communication resource utilization (corresponding to the bandwidth utilization mentioned above); the statistics included in the network load balance are: computing resource load balance and communication resource load balance; examples of the priority setting of the reward function are as follows:

[0343] Condition 1. The computing level should enable the task to be completed within the latency tolerance.

[0344] Condition 2. The computing level should ensure the accuracy requirements of the task on the basis of meeting condition 1.

[0345] Condition 3. The computing level should ensure the concealment requirements of the task on the basis of satisfying conditions 1 and 2.

[0346] Condition 4. The computing layer should ensure maximum network resource utilization and optimal load balancing on the basis of meeting conditions 1, 2, and 3.

[0347] The specific forms of S, A and R are as follows:

[0348] (1) State space S:

[0349] S: [task, resource, Error];

[0350] Among them, resource represents resource status statistics, task represents task status statistics, and Error represents the QoS satisfaction degree of the computing task fed back in the previous round. The representation forms of resource and task are as follows:

[0351]

[0352] Where type, num access 、computility exp 、delay exp , BER exp 、con exp They represent: the type of computing task, the number of accesses, the amount of computing resources requested, latency requirements, privacy requirements (which can be represented by the number of nodes used for computing, the more nodes, the worse the privacy), accuracy requirements (which can be represented by bit error rate or bit error rate or frame error rate, the lower the bit error rate, the higher the accuracy requirement), and continuity requirements. These seven variables belong to the statistics of computing task status and can be directly given by the terminal uploading the CU.

[0353] resource: [bandwith, trans, UTE, BER];

[0354] Bandwith, trans, UTE, and BER represent the total bandwidth, throughput, resource utilization, and bit error rate or frame error rate of each link during sampling. These four variables belong to resource status statistics and can be directly given by the CU reading the network.

[0355] (2) Action space A:

[0356] A:[level,Topology];

[0357] Level represents the computing level of AIMO decision, and Topology represents the specific network topology under this computing level (i.e. computing topology decision). These two are generated by AIMO based on the reinforcement learning algorithm.

[0358] (3) Reward function R (including reward):

[0359]

[0360] Which corresponds to f(delay real , delay exp)、f(BER real , BER exp ), f(con real ,con exp ), the function f(x, y) is defined as follows:

[0361]

[0362] Among them, delay real , BER real , con real It represents: the actual delay, bit error rate or frame error rate, privacy, and continuity fed back by the CU after the execution of this round of actions. These four pieces of information are read by the CU and fed back to AIMO. bal represents the balance of the network this time (which can correspond to the above network load balance). This value is calculated by the CU and fed back to AIMO. bal can be calculated by the resource load balance bal cal and communication resource load balancing degree bal trans Weighted, bal cal The calculation formula of can refer to the supplementary embodiment below, bal trans With bal cal The calculation formula is basically the same, and no more examples are given here. In addition, other related parameters can refer to the above explanations of S and A, and no more details are given here.

[0363] Supplementary embodiment:

[0364] (1) The QoS satisfaction of a computing task can be expressed by the error between the required QoS of the computing task and the actual QoS of the computing task. The specific expression is as follows:

[0365] Error=[Error pri , Error delay , Error BER , Error con ];

[0366]

[0367] Error represents the QoS error of a single task; Error total Represents the task satisfaction statistics of the entire community. Here, the mean of all task satisfaction is used as an example; num task Represents the number of computing tasks for the cell. pri , Error delay , Error BER , Error conThey represent the errors between AIMO's expected and actual feedback in terms of privacy, latency, calculation accuracy, and persistence. The specific formulas are as follows:

[0368] Error delay =delay exp -delay real ;

[0369] Where delay exp represents the delay required by the user (i.e. the delay expected by AIMO), delay real Represents the actual feedback delay.

[0370] (2) Calculation of resource utilization can be achieved using the following formula:

[0371]

[0372] where cal node Represents the computing amount (i.e. computing resources) currently used by the network element node. total Represents the total computing capacity of the network element node.

[0373] (3) The computing resource load balance can be measured by the statistical variance of computing resource utilization. The smaller the variance, the more balanced the load distribution. The specific formula is as follows:

[0374]

[0375]

[0376] Among them, UTE node Represents the computing resource utilization of the node (i.e., the above-mentioned UTE); UTE total Represents the overall computing resource utilization of the network (all nodes corresponding to the topology). Here, all UTEs in the network are used as an example. node The weighted value is used to measure; num node Represents the number of network element nodes (that is, the number of nodes corresponding to the topology). bal Represents the (expected value of computing resource load balancing - actual calculated value) of the previous round. The expected value can be pre-configured by network operation and maintenance personnel.

[0377] In addition, in this solution, the agent of AIMO reinforcement learning is the cell. Since the computing power of network elements in the network is shared by multiple cells, this process belongs to multi-agent reinforcement learning (that is, one reinforcement learning model can correspond to at least one cell). The reinforcement learning algorithm used in AIMO can be one or more combinations of multi-agent reinforcement learning algorithms such as BiCNet (Bilateral Complementary Network), CommNet (Communication Neural Net), MADDPG (Multi-Agent Deep Deterministic Policy Gradient), etc., which are not limited here.

[0378] As mentioned above, all existing solutions have certain defects in computing resource allocation, and are all proposed for pure computing networks without considering communication-related statistics, which cannot meet the needs of 6G computing fusion systems. In this regard, this solution is aimed at the new 6G computing fusion network and proposes a resource scheduling algorithm that completes computing resource allocation by dividing the computing level in cells to meet the needs of 6G computing fusion systems. In addition, this solution also has the following advantages:

[0379] 1. The establishment of the computing layer provides the possibility of hard isolation, but also allows resource overlap between cells and allows redundant resources between cells to be scheduled with each other, thus ensuring resource scheduling reference while taking into account resource scheduling flexibility.

[0380] 2. Computing resources are bound to cells during scheduling, facilitating deep coupling and coordinated management of communication and computing.

[0381] 3. Before allocating resources, computing tasks can be divided into optional network element nodes according to the computing level, which makes the search space for resource allocation smaller and reduces the delay and energy consumption of the network resource allocation process to a certain extent.

[0382] The embodiment of the present invention further provides an information processing device, which is applied to a first network device, such as Fig.11 As shown, including:

[0383] The first acquisition module 111 is used to acquire computing requirement information and resource status information corresponding to at least one cell; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0384] A first sending module 112, configured to send the computing requirement information and resource status information to the second network device;

[0385] The first receiving module 113 is used to receive the decision information for each of the cells fed back by the second network device; the decision information includes: calculation level information and calculation topology information;

[0386] The first determining module 114 is configured to determine resource configuration information of the cell according to the decision information.

[0387] The information processing device provided in the embodiment of the present invention obtains computing demand information and resource status information corresponding to at least one cell; the computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; sends the computing demand information and resource status information to a second network device; receives decision information for each of the cells fed back by the second network device; the decision information includes: computing level information and computing topology information; determines resource configuration information of the cell based on the decision information; can support the implementation of integrated communication-related information and computing-related information to perform resource configuration, improve resource configuration accuracy, and solve the problem of poor resource configuration accuracy in the prior art for computing resource scheduling solutions.

[0388] Among them, the obtaining of computing demand information and resource status information corresponding to at least one cell includes: adopting a first granularity and a sample extraction method to obtain the computing demand information and resource status information corresponding to at least one cell; wherein, the first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity; the sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and at least one of extraction by computing task-related characteristic attributes.

[0389] In an embodiment of the present invention, the computing demand information also includes: at least one of: the number of computing task requests, the service quality QoS requirement information of the computing task, and the mobility requirement information of the computing task; and / or, the communication resource status information includes: the status information of the communication connection between the first network device and the connected network element, and the status information includes: the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, the bandwidth utilization, the bit error rate or bit error rate or frame error rate of each link, the success rate of each link establishment, and at least one of the link interruption probability; the connected network element refers to the network element that is communicatively connected to the first network device; and / or, the computing resource status information includes: the total amount of computing resources of the connected network elements of the first network device, the computing resource utilization, the allocable computing capacity, and at least one of the number of computing tasks carried; and / or, the computing level information includes: the first level information, The second-level information and / or the third-level information, the first-level information includes: information about the third network device to which the cell belongs, the second-level information includes: information about the first network device to which the cell belongs and information about all third network devices under the first network device to which the cell belongs, the third-level information includes: information about the first network device to which the cell belongs, information about all third network devices under the first network device to which the cell belongs, and information about the multi-access edge computing MEC device adjacent to the first network device to which the cell belongs; and / or, the computing topology information includes: information about the network device corresponding to the computing level information, the connection relationship between the network devices corresponding to the computing level information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing level information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing level information.

[0390] Furthermore, the information processing device also includes: a second acquisition module, used to obtain configuration effect information for the decision information after receiving the decision information for each of the cells fed back by the second network device; a second sending module, used to send the configuration effect information to the second network device; wherein the configuration effect information includes at least one of the following: the number of accepted requests or the request acceptance rate of the computing task; the number of successful executions or the execution success rate of the computing task; the interruption probability of the computing task; the QoS satisfaction degree information of the computing task; and the resource load balancing degree, wherein the resource load balancing degree includes: at least one of the computing resource load balancing degree and the communication resource load balancing degree.

[0391] Among them, the QoS satisfaction degree information of the computing task includes: the QoS satisfaction degree information corresponding to a single computing task, and at least one item of the QoS satisfaction degree information corresponding to at least two computing tasks; and / or, the computing resource load balancing degree refers to: the balance degree of computing resource utilization of each network element within the scope of the first network device and the connected network elements; the connected network elements refer to the network elements that are communicatively connected to the first network device; and / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0392] Furthermore, the information processing device also includes: a second receiving module, used to receive a statistical request sent by the second network device before obtaining the computing requirement information and resource status information corresponding to at least one cell; the obtaining of the computing requirement information and resource status information corresponding to at least one cell includes: obtaining the computing requirement information and resource status information corresponding to at least one cell according to the statistical request; wherein the statistical request carries: at least one of the first granularity, the sample extraction method, the identification of the information to be obtained, the computing task number, the user identification and the cell identification.

[0393] In the embodiment of the present invention, the decision information also carries the validity time information of the decision information.

[0394] Among them, the implementation embodiments of the information processing method on the first network device side are all applicable to the embodiments of the information processing device and can achieve the same technical effects.

[0395] The embodiment of the present invention further provides an information processing device, which is applied to a second network device, such as Fig.12 As shown, including:

[0396] The third receiving module 121 is used to receive computing requirement information and resource status information corresponding to at least one cell sent by the first network device; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0397] The third acquisition module 122 is used to acquire decision information for each of the cells according to the computing demand information and the resource status information; the decision information includes: computing level information and computing topology information;

[0398] The third sending module 123 is used to send the decision information to the first network device.

[0399] The information processing device provided by the embodiment of the present invention receives computing demand information and resource status information corresponding to at least one cell sent by a first network device; the computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; according to the computing demand information and resource status information, decision information for each of the cells is obtained; the decision information includes: computing level information and computing topology information; the decision information is sent to the first network device; it can support the implementation of fusion of communication-related information and computing-related information to perform resource configuration, improve resource configuration accuracy, and solve the problem of poor resource configuration accuracy in the existing technology for computing resource scheduling solutions.

[0400] Wherein, the computing demand information also includes: at least one of the number of computing task requests, the service quality QoS requirement information of the computing task and the mobility requirement information of the computing task; and / or, the communication resource status information includes: the status information of the communication connection between the first network device and the connected network element, and the status information includes: the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, the bandwidth utilization, the bit error rate or bit error rate or frame error rate of each link, the success rate of each link establishment, and at least one of the link interruption probability; the connected network element refers to the network element that is communicatively connected to the first network device; and / or, the computing resource status information includes: the total amount of computing resources of the connected network elements of the first network device, the computing resource utilization, the allocable computing capacity and at least one of the number of computing tasks carried; and / or, the computing level information includes: the first level information, the second level information The first-level information and / or third-level information, the first-level information includes: information about the third network device to which the cell belongs, the second-level information includes: information about the first network device to which the cell belongs and information about all third network devices under the first network device to which the cell belongs, the third-level information includes: information about the first network device to which the cell belongs, information about all third network devices under the first network device to which the cell belongs, and information about the multi-access edge computing MEC device adjacent to the first network device to which the cell belongs; and / or, the computing topology information includes: information about the network device corresponding to the computing level information, connection relationships between the network devices corresponding to the computing level information, a maximum computing capacity allocated to the cell for the network device corresponding to the computing level information, and at least one of a maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to a link between the network devices corresponding to the computing level information.

[0401] Furthermore, the information processing device also includes: a fourth receiving module, used to receive configuration effect information sent by the first network device after sending the decision information to the first network device; wherein the configuration effect information includes at least one of the following: the number of accepted requests or the request acceptance rate of the computing task; the number of successful executions or the execution success rate of the computing task; the interruption probability of the computing task; the QoS satisfaction degree information of the computing task; and the resource load balancing degree, wherein the resource load balancing degree includes: at least one of the computing resource load balancing degree and the communication resource load balancing degree.

[0402] Among them, the QoS satisfaction degree information of the computing task includes: the QoS satisfaction degree information corresponding to a single computing task, and at least one item of the QoS satisfaction degree information corresponding to at least two computing tasks; and / or, the computing resource load balancing degree refers to: the balance degree of computing resource utilization of each network element within the scope of the first network device and the connected network elements; the connected network elements refer to the network elements that are communicatively connected to the first network device; and / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0403] Furthermore, the information processing device also includes: a fourth sending module, which is used to send a statistical request to the first network device before receiving the computing demand information and resource status information corresponding to at least one cell sent by the first network device; the receiving of the computing demand information and resource status information corresponding to at least one cell sent by the first network device includes: receiving the computing demand information and resource status information corresponding to at least one cell fed back by the first network device according to the statistical request; wherein the statistical request carries: at least one of the first granularity, the sample extraction method, the identifier of the information to be obtained, the computing task number, the user identifier and the cell identifier; the first granularity refers to the granularity used by the first network device to obtain the computing demand information and the resource status information, and the sample extraction method refers to the sample extraction method used by the first network device to obtain the computing demand information and the resource status information; the first granularity includes: at least one of the task granularity, the user granularity, the cell granularity and the network element granularity; the sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related characteristic attributes.

[0404] In an embodiment of the present invention, the information processing device further includes: a first processing module, used to mark corresponding timestamps for the computing requirement information, resource status information and configuration effect information respectively; obtaining decision information for each of the cells based on the computing requirement information and resource status information, including: obtaining decision information for each of the cells based on the computing requirement information, resource status information and configuration effect information after marking the timestamps.

[0405] Among them, obtaining decision information for each of the cells based on the computing demand information and the resource status information includes: obtaining change degree information corresponding to the computing demand information and the resource status information; when the change degree information is greater than or equal to a threshold, obtaining decision information for each of the cells based on the computing demand information and the resource status information; when the change degree information is less than a threshold, using the decision information obtained last time as the decision information this time.

[0406] Furthermore, the decision information also carries validity time information of the decision information.

[0407] Among them, the implementation embodiments of the information processing method on the second network device side are all applicable to the embodiments of the information processing device and can achieve the same technical effects.

[0408] The embodiment of the present invention further provides an information processing device, wherein the information processing device is a first network device, such as Fig.13 As shown, it includes: a processor 131 and a transceiver 132;

[0409] The processor 131 is configured to obtain computing requirement information and resource status information corresponding to at least one cell; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0410] Sending the computing requirement information and resource status information to the second network device via the transceiver 132;

[0411] Receiving, through the transceiver 132, decision information for each of the cells fed back by the second network device; the decision information includes: computing level information and computing topology information;

[0412] Determine resource configuration information of the cell according to the decision information.

[0413] The information processing device provided in the embodiment of the present invention obtains computing demand information and resource status information corresponding to at least one cell; the computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; sends the computing demand information and resource status information to a second network device; receives decision information for each of the cells fed back by the second network device; the decision information includes: computing level information and computing topology information; determines resource configuration information of the cell based on the decision information; can support the implementation of integrated communication-related information and computing-related information to perform resource configuration, improve resource configuration accuracy, and solve the problem of poor resource configuration accuracy in the existing technology for computing resource scheduling solutions.

[0414] Among them, the obtaining of computing demand information and resource status information corresponding to at least one cell includes: adopting a first granularity and a sample extraction method to obtain the computing demand information and resource status information corresponding to at least one cell; wherein, the first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity; the sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and at least one of extraction by computing task-related characteristic attributes.

[0415] Further, the computing demand information also includes: at least one of the number of computing task requests, the quality of service QoS requirement information of the computing tasks, and the mobility requirement information of the computing tasks; and / or, the communication resource status information includes: the status information of the communication connection between the first network device and the connected network element, and the status information includes: the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, the bandwidth utilization, the bit error rate or bit error rate or frame error rate of each link, the success rate of each link establishment, and at least one of the link interruption probability; the connected network element refers to the network element that is communicatively connected to the first network device; and / or, the computing resource status information includes: the total amount of computing resources of the connected network elements of the first network device, the computing resource utilization, the allocable computing capacity, and at least one of the number of computing tasks carried; and / or, the computing level information includes: the first level information, the second level information hierarchical information and / or third-level information, the first-level information includes: information about the third network device to which the cell belongs, the second-level information includes: information about the first network device to which the cell belongs and information about all third network devices under the first network device to which the cell belongs, the third-level information includes: information about the first network device to which the cell belongs, information about all third network devices under the first network device to which the cell belongs, and information about the multi-access edge computing MEC device adjacent to the first network device to which the cell belongs; and / or, the computing topology information includes: information about the network device corresponding to the computing hierarchical information, the connection relationship between the network devices corresponding to the computing hierarchical information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchical information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchical information.

[0416] In an embodiment of the present invention, the processor is also used to: after receiving the decision information for each of the cells fed back by the second network device, obtain configuration effect information for the decision information; send the configuration effect information to the second network device through the transceiver; wherein the configuration effect information includes at least one of the following: the number of accepted requests or the request acceptance rate of the computing task; the number of successful executions or the execution success rate of the computing task; the interruption probability of the computing task; the QoS satisfaction degree information of the computing task; and the resource load balancing degree, wherein the resource load balancing degree includes: at least one of the computing resource load balancing degree and the communication resource load balancing degree.

[0417] Among them, the QoS satisfaction degree information of the computing task includes: the QoS satisfaction degree information corresponding to a single computing task, and at least one item of the QoS satisfaction degree information corresponding to at least two computing tasks; and / or, the computing resource load balancing degree refers to: the balance degree of computing resource utilization of each network element within the scope of the first network device and the connected network elements; the connected network elements refer to the network elements that are communicatively connected to the first network device; and / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0418] Furthermore, the processor is also used to: before obtaining the computing requirement information and resource status information corresponding to at least one cell, receive a statistical request sent by the second network device through the transceiver; the obtaining of the computing requirement information and resource status information corresponding to at least one cell includes: obtaining the computing requirement information and resource status information corresponding to at least one cell according to the statistical request; wherein the statistical request carries: at least one of the first granularity, the sample extraction method, the identification of the information to be obtained, the computing task number, the user identification and the cell identification.

[0419] In the embodiment of the present invention, the decision information also carries the validity time information of the decision information.

[0420] Among them, the implementation embodiments of the information processing method on the first network device side are all applicable to the embodiments of the information processing device and can achieve the same technical effects.

[0421] The embodiment of the present invention further provides an information processing device, wherein the information processing device is a second network device, such as Fig.14 As shown, it includes: a processor 141 and a transceiver 142;

[0422] The processor 141 is configured to receive computing requirement information and resource status information corresponding to at least one cell sent by the first network device through the transceiver 142; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information;

[0423] According to the computing demand information and resource status information, obtaining decision information for each of the cells; the decision information includes: computing level information and computing topology information;

[0424] The decision information is sent to the first network device through the transceiver 142.

[0425] The information processing device provided in the embodiment of the present invention receives computing requirement information and resource status information corresponding to at least one cell sent by a first network device; the computing requirement information includes: communication resource requirement information and computing resource requirement information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; according to the computing requirement information and resource status information, decision information for each of the cells is obtained; the decision information includes: computing level information and computing topology information; the decision information is sent to the first network device; it can support the implementation of fusion of communication-related information and computing-related information to perform resource configuration, improve resource configuration accuracy, and solve the problem of poor resource configuration accuracy in the existing technology for computing resource scheduling solutions.

[0426] Further, the computing demand information also includes: at least one of the number of computing task requests, the quality of service QoS requirement information of the computing tasks, and the mobility requirement information of the computing tasks; and / or, the communication resource status information includes: the status information of the communication connection between the first network device and the connected network element, and the status information includes: the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, the bandwidth utilization, the bit error rate or bit error rate or frame error rate of each link, the success rate of each link establishment, and at least one of the link interruption probability; the connected network element refers to the network element that is communicatively connected to the first network device; and / or, the computing resource status information includes: the total amount of computing resources of the connected network elements of the first network device, the computing resource utilization, the allocable computing capacity, and at least one of the number of computing tasks carried; and / or, the computing level information includes: the first level information, the second level information hierarchical information and / or third-level information, the first-level information includes: information about the third network device to which the cell belongs, the second-level information includes: information about the first network device to which the cell belongs and information about all third network devices under the first network device to which the cell belongs, the third-level information includes: information about the first network device to which the cell belongs, information about all third network devices under the first network device to which the cell belongs, and information about the multi-access edge computing MEC device adjacent to the first network device to which the cell belongs; and / or, the computing topology information includes: information about the network device corresponding to the computing hierarchical information, the connection relationship between the network devices corresponding to the computing hierarchical information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchical information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchical information.

[0427] In an embodiment of the present invention, the processor is also used to: after sending the decision information to the first network device, receive the configuration effect information sent by the first network device through the transceiver; wherein the configuration effect information includes at least one of the following: the number of accepted requests or the request acceptance rate of the computing task; the number of successful executions or the execution success rate of the computing task; the interruption probability of the computing task; the QoS satisfaction degree information of the computing task; and the resource load balancing degree, wherein the resource load balancing degree includes: at least one of the computing resource load balancing degree and the communication resource load balancing degree.

[0428] Among them, the QoS satisfaction degree information of the computing task includes: the QoS satisfaction degree information corresponding to a single computing task, and at least one item of the QoS satisfaction degree information corresponding to at least two computing tasks; and / or, the computing resource load balancing degree refers to: the balance degree of computing resource utilization of each network element within the scope of the first network device and the connected network elements; the connected network elements refer to the network elements that are communicatively connected to the first network device; and / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

[0429] Further, the processor is also used to: before receiving the computing demand information and resource status information corresponding to at least one cell sent by the first network device, send a statistical request to the first network device through the transceiver; the receiving of the computing demand information and resource status information corresponding to at least one cell sent by the first network device includes: receiving the computing demand information and resource status information corresponding to at least one cell fed back by the first network device according to the statistical request; wherein the statistical request carries: at least one of the first granularity, the sample extraction method, the identifier of the information to be obtained, the computing task number, the user identifier and the cell identifier; the first granularity refers to the granularity used by the first network device to obtain the computing demand information and resource status information, and the sample extraction method refers to the sample extraction method used by the first network device to obtain the computing demand information and resource status information; the first granularity includes: at least one of the task granularity, the user granularity, the cell granularity and the network element granularity; the sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related characteristic attributes.

[0430] In an embodiment of the present invention, the processor is also used to: respectively mark corresponding timestamps for the computing requirement information, resource status information and configuration effect information; and obtain decision information for each of the cells based on the computing requirement information and resource status information, including: obtaining decision information for each of the cells based on the computing requirement information, resource status information and configuration effect information after marking the timestamps.

[0431] Among them, obtaining decision information for each of the cells based on the computing demand information and the resource status information includes: obtaining change degree information corresponding to the computing demand information and the resource status information; when the change degree information is greater than or equal to a threshold, obtaining decision information for each of the cells based on the computing demand information and the resource status information; when the change degree information is less than a threshold, using the decision information obtained last time as the decision information this time.

[0432] Furthermore, the decision information also carries validity time information of the decision information.

[0433] Among them, the implementation embodiments of the information processing method on the second network device side are all applicable to the embodiments of the information processing device and can achieve the same technical effects.

[0434] An embodiment of the present invention also provides an information processing device, including a memory, a processor, and a program stored in the memory and executable on the processor; when the processor executes the program, the information processing method on the first network device side or the second network device side is implemented.

[0435] Among them, the implementation embodiments of the information processing method on the first network device side or the second network device side are all applicable to the embodiments of the information processing device and can achieve the same technical effects.

[0436] An embodiment of the present invention further provides a readable storage medium on which a program is stored. When the program is executed by a processor, the steps in the information processing method on the first network device side or the second network device side are implemented.

[0437] Among them, the implementation embodiments of the information processing method on the first network device side or the second network device side are all applicable to the embodiments of the readable storage medium and can achieve the same technical effect.

[0438] It should be noted that many functional components described in this specification are referred to as modules in order to more particularly emphasize the independence of their implementation methods.

[0439] In the embodiment of the present invention, module can be implemented with software so that it can be executed by various types of processors. For example, an executable code module of an identification can include one or more physical or logical blocks of computer instructions, for example, it can be constructed as an object, process or function. Nevertheless, the executable code of the identified module does not need to be physically located together, but can include different instructions stored in different positions, and when these instructions are logically combined together, it constitutes a module and realizes the specified purpose of the module.

[0440] In fact, executable code module can be a single instruction or many instructions, and can even be distributed on a plurality of different code segments, distributed among different programs, and distributed across a plurality of memory devices. Similarly, operating data can be identified in the module, and can be implemented and organized in the data structure of any appropriate type according to any appropriate form. The operating data can be collected as a single data set, or can be distributed in different locations (including on different storage devices), and can only be present on a system or network as an electronic signal at least in part.

[0441] When a module can be implemented by software, considering the level of existing hardware technology, a person skilled in the art can build a corresponding hardware circuit to implement the corresponding function of the module that can be implemented by software without considering the cost. The hardware circuit includes a conventional very large scale integration (VLSI) circuit or gate array and existing semiconductors such as logic chips, transistors, or other discrete components. The module can also be implemented by a programmable hardware device, such as a field programmable gate array, a programmable array logic, a programmable logic device, etc.

[0442] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary personnel in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An information processing method, applied to a first network device, characterized in that: include: Obtain computing demand information and resource status information corresponding to at least one cell; The computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; Sending the computing requirement information and resource status information to the second network device; receiving decision information for each of the cells fed back by the second network device; the decision information comprising: computing level information and computing topology information; Determine resource configuration information of the cell according to the decision information.

2. The information processing method according to claim 1, characterized in that: The obtaining of computing demand information and resource status information corresponding to at least one cell includes: Using a first granularity and a sample extraction method, obtaining computing demand information and resource status information corresponding to at least one cell; The first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity; The sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related feature attributes.

3. The information processing method according to claim 1, characterized in that: The computing demand information also includes at least one of: the number of computing task requests, the quality of service QoS requirement information of the computing task, and the mobility requirement information of the computing task; And / or, the communication resource status information includes: status information of the communication connection between the first network device and the connected network element, the status information includes: at least one of the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, bandwidth utilization, bit error rate or bit error rate or frame error rate of each link, success rate of establishing each link, and link interruption probability; the connected network element refers to a network element that is communicatively connected to the first network device; And / or, the computing resource status information includes: at least one of the total amount of computing resources, computing resource utilization, allocable computing capacity, and the number of computing tasks carried of the connected network elements of the first network device; And / or, the computing level information includes: first level information, second level information and / or third level information, the first level information includes: information of a third network device to which the cell belongs, the second level information includes: information of the first network device to which the cell belongs and information of all third network devices under the first network device to which the cell belongs, the third level information includes: information of the first network device to which the cell belongs, information of all third network devices under the first network device to which the cell belongs, and information of a multi-access edge computing MEC device adjacent to the first network device to which the cell belongs; And / or, the computing topology information includes: information of the network device corresponding to the computing hierarchy information, the connection relationship between the network devices corresponding to the computing hierarchy information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchy information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchy information.

4. The information processing method according to claim 1, characterized in that: After receiving the decision information for each of the cells fed back by the second network device, the method further includes: Obtaining configuration effect information for the decision information; Sending the configuration effect information to the second network device; The configuration effect information includes at least one of the following: Calculate the number of accepted requests or the acceptance rate of the task; Calculate the number of successful executions or the execution success rate of tasks; Calculate the interruption probability of the task; QoS satisfaction degree information of computing tasks; Resource load balancing degree, the resource load balancing degree includes: at least one of computing resource load balancing degree and communication resource load balancing degree.

5. The information processing method according to claim 4, characterized in that: The QoS satisfaction degree information of the computing task includes: QoS satisfaction degree information corresponding to a single computing task, and at least one of QoS satisfaction degree information corresponding to at least two computing tasks; And / or, the computing resource load balance refers to: the balance degree of computing resource utilization of each network element within the first network device and the connected network element; the connected network element refers to the network element that is communicatively connected to the first network device; And / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

6. The information processing method according to claim 2, characterized in that: Before obtaining the computing requirement information and resource status information corresponding to at least one cell, the method further includes: receiving a statistics request sent by the second network device; The obtaining of computing demand information and resource status information corresponding to at least one cell includes: According to the statistical request, obtaining computing demand information and resource status information corresponding to at least one cell; The statistical request carries: the first granularity, the sample extraction method, an identifier of the information to be obtained, a calculation task number, a user identifier, and at least one of a cell identifier.

7. The information processing method according to claim 1, characterized in that: The decision information also carries the validity time information of the decision information.

8. An information processing method, applied to a second network device, characterized in that: include: receiving computing requirement information and resource status information corresponding to at least one cell sent by the first network device; The computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; According to the computing demand information and resource status information, obtaining decision information for each of the cells; the decision information includes: computing level information and computing topology information; The decision information is sent to the first network device.

9. The information processing method according to claim 8, characterized in that: The computing demand information also includes at least one of: the number of computing task requests, the quality of service QoS requirement information of the computing task, and the mobility requirement information of the computing task; And / or, the communication resource status information includes: status information of the communication connection between the first network device and the connected network element, the status information includes: at least one of the total transmission bandwidth or total transmission throughput of each link, the allocable bandwidth or allocable throughput, bandwidth utilization, bit error rate or bit error rate or frame error rate of each link, success rate of establishing each link, and link interruption probability; the connected network element refers to a network element that is communicatively connected to the first network device; And / or, the computing resource status information includes: at least one of the total amount of computing resources, computing resource utilization, allocable computing capacity, and the number of computing tasks carried of the connected network elements of the first network device; And / or, the computing level information includes: first level information, second level information and / or third level information, the first level information includes: information of a third network device to which the cell belongs, the second level information includes: information of the first network device to which the cell belongs and information of all third network devices under the first network device to which the cell belongs, the third level information includes: information of the first network device to which the cell belongs, information of all third network devices under the first network device to which the cell belongs, and information of a multi-access edge computing MEC device adjacent to the first network device to which the cell belongs; And / or, the computing topology information includes: information of the network device corresponding to the computing hierarchy information, the connection relationship between the network devices corresponding to the computing hierarchy information, the maximum computing capacity allocated to the cell for the network device corresponding to the computing hierarchy information, and at least one of the maximum communication bandwidth allocated to the cell for the inter-node link; the inter-node link refers to the link between the network devices corresponding to the computing hierarchy information.

10. The information processing method according to claim 8, characterized in that: After sending the decision information to the first network device, the method further includes: Receiving configuration effect information sent by the first network device; The configuration effect information includes at least one of the following: Calculate the number of accepted requests or the acceptance rate of the task; Calculate the number of successful executions or the execution success rate of tasks; Calculate the interruption probability of the task; QoS satisfaction degree information of computing tasks; Resource load balancing degree, the resource load balancing degree includes: at least one of computing resource load balancing degree and communication resource load balancing degree.

11. The information processing method according to claim 10, characterized in that: The QoS satisfaction degree information of the computing task includes: QoS satisfaction degree information corresponding to a single computing task, and at least one of QoS satisfaction degree information corresponding to at least two computing tasks; And / or, the computing resource load balance refers to: the balance degree of computing resource utilization of each network element within the first network device and the connected network element; the connected network element refers to the network element that is communicatively connected to the first network device; And / or, the communication resource load balancing degree refers to: the balance degree of resource utilization of each communication link within the scope of the first network device and the connected network elements.

12. The information processing method according to claim 8, characterized in that: Before receiving the computing requirement information and resource status information corresponding to at least one cell sent by the first network device, the method further includes: Sending a statistics request to the first network device; The receiving computing requirement information and resource status information corresponding to at least one cell sent by the first network device includes: receiving computing requirement information and resource status information corresponding to at least one cell fed back by the first network device according to the statistical request; The statistical request carries: at least one of the first granularity, the sample extraction method, the identifier of the information to be obtained, the calculation task number, the user identifier, and the cell identifier; The first granularity refers to the granularity used by the first network device to obtain computing demand information and resource status information, and the sample extraction method refers to the sample extraction method used by the first network device to obtain computing demand information and resource status information; The first granularity includes: at least one of task granularity, user granularity, cell granularity and network element granularity; The sample extraction method includes: extraction by task number, extraction by computing task type, extraction by specified time period, and extraction by computing task-related feature attributes.

13. The information processing method according to claim 10, characterized in that: Also includes: Marking corresponding timestamps for the computing requirement information, resource status information, and configuration effect information respectively; The acquiring decision information for each of the cells according to the computing demand information and the resource status information includes: Decision information for each of the cells is obtained according to the computing requirement information, resource status information and configuration effect information marked with a timestamp.

14. The information processing method according to claim 8, characterized in that: The acquiring decision information for each of the cells according to the computing demand information and the resource status information includes: Obtaining information on the degree of change corresponding to the computing demand information and the resource status information; When the change degree information is greater than or equal to a threshold, obtaining decision information for each of the cells according to the computing demand information and the resource status information; When the change degree information is less than a threshold value, the decision information obtained last time is used as the decision information for this time.

15. The information processing method according to claim 8, characterized in that: The decision information also carries the validity time information of the decision information.

16. An information processing device, applied to a first network device, characterized in that: include: A first acquisition module, used to acquire computing demand information and resource status information corresponding to at least one cell; The computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; A first sending module, used for sending the computing requirement information and resource status information to the second network device; A first receiving module, configured to receive decision information for each of the cells fed back by the second network device; the decision information includes: computing level information and computing topology information; The first determination module is used to determine the resource configuration information of the cell according to the decision information.

17. An information processing device, applied to a second network device, characterized in that: include: A third receiving module, configured to receive computing requirement information and resource status information corresponding to at least one cell sent by the first network device; The computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; A third acquisition module is used to acquire decision information for each of the cells according to the computing demand information and the resource status information; the decision information includes: computing level information and computing topology information; The third sending module is used to send the decision information to the first network device.

18. An information processing device, the information processing device being a first network device, characterized in that: include: processor and transceiver; The processor is used to obtain computing demand information and resource status information corresponding to at least one cell; The computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; Sending the computing requirement information and resource status information to the second network device via the transceiver; Receiving, through the transceiver, decision information for each of the cells fed back by the second network device; the decision information includes: computing level information and computing topology information; Determine resource configuration information of the cell according to the decision information.

19. An information processing device, the information processing device being a second network device, characterized in that: include: processor and transceiver; The processor is configured to receive, through the transceiver, computing requirement information and resource status information corresponding to at least one cell sent by the first network device; The computing demand information includes: communication resource demand information and computing resource demand information of the computing task, and the resource status information includes: communication resource status information and computing resource status information; According to the computing demand information and resource status information, obtaining decision information for each of the cells; the decision information includes: computing level information and computing topology information; The decision information is sent to the first network device through the transceiver.

20. An information processing device, comprising a memory, a processor, and a program stored in the memory and executable on the processor; characterized in that: When the processor executes the program, the information processing method according to any one of claims 1 to 15 is implemented.

21. A readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps of the information processing method according to any one of claims 1 to 15 are implemented.

Images