Computing power network information notification and routing decision method and device, and medium

Abstract

The application provides a computing power network information notification and routing decision method, device and medium, relates to the technical field of communication, and is used for solving the problem that there is no effective method for realizing information notification and routing decision of each node in the computing power network in the prior art. The method comprises the following steps: acquiring and storing the information of each node in the computing power network in real time; formulating respective issued information for each entry node in the computing power network according to the information of each node; notifying the respective issued information to each entry node, so that each entry node receives the issued information notified to itself, and makes routing decision for the service accessing the computing power network through itself according to the received issued information. The application realizes centralized information notification and distributed routing decision, and can be applied to the demand for simultaneous scheduling of the computing power nodes and the network nodes in the computing power network.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of communication technology, in particular to a computing power network information notification and routing decision method, an arrangement control system, an entry node, a computer device and a computer readable storage medium. BACKGROUND

[0002] When scheduling computing power nodes and network nodes to provide services for access services in a computing power network, it is necessary to make optimal routing decisions for node resource configuration according to the information of each node in combination with service requirements. There are two kinds of implementation modes that can be referred to:

[0003] One is distributed information notification and routing decision, that is, some network node devices collect information of the remaining nodes and make their own routing decisions. The disadvantage is that all information is diffused, related protocols need to be notified synchronously, resulting in bloated data, and the information interaction mechanism between nodes is not mature;

[0004] The other is to refer to the scheduling of network resources in the existing general network system, that is, the upper system collects network node information and makes routing decisions, and then publishes them to network node devices to execute related routing decisions. The disadvantage is that the routing decision calculation is complex, the convergence speed is slow, and the processing mechanism for computing power node information is lacking, which cannot meet the needs of simultaneous scheduling of computing power nodes and network nodes in the computing power network.

[0005] It can be seen that the prior art lacks an effective method for information notification and routing decision of each node in the computing power network. SUMMARY

[0006] The technical problem to be solved by the present application is to provide a computing power network information notification and routing decision method, an arrangement control system, an entry node, a computer device and a computer readable storage medium to solve the problem that the prior art lacks an effective method for information notification and routing decision of each node in the computing power network.

[0007] In a first aspect, the present application provides a computing power network information notification and routing decision method applied to a computing power network arrangement control system, the method comprising:

[0008] Real-time acquisition and storage of information of each node in the computing power network;

[0009] Formulating respective issued information for each entry node in the computing power network according to the information of each node;

[0010] Notifying each entry node of its respective issued information, so that each entry node receives the issued information notified to itself, and makes routing decisions for services accessing the computing power network through itself according to the received issued information.

[0011] Optionally, the node information specifically includes: computing power node information.

[0012] The computing power node information includes: computing power node location information and computing power node resource information.

[0013] Optionally, the node information specifically further includes: network node information and entry node information.

[0014] The network node information includes: network node location information, network node connection information and network node performance information.

[0015] The entry node information includes: entry node location information and access service information.

[0016] Optionally, the real-time acquisition and storage of the node information in the computing power network specifically include:

[0017] Receiving the self-change information of each computing power node, each network node and each entry node in the computing power network after the self information changes, or querying the self information of each computing power node, each network node and each entry node in the computing power network according to a preset frequency;

[0018] Storing each computing power node information received or queried in a computing power node information database, storing each network node information received or queried in a network node information database, and storing each entry node information received or queried in an entry node information database.

[0019] Optionally, the preparation of the respective delivery information for each entry node in the computing power network according to the node information specifically includes:

[0020] Judging whether each entry node has routing decision capability;

[0021] If the first entry node has routing decision capability, aggregating the node information for the first entry node, wherein the first entry node is an entry node having routing decision capability.

[0022] Further judging whether the change amplitude of each parameter in the aggregated node information exceeds a preset threshold, and if so, taking the current value of the parameter whose change amplitude exceeds the preset threshold as the delivery information for the first entry node.

[0023] Optionally, the aggregation of the node information for the first entry node specifically includes:

[0024] According to the node information, dividing each node in the computing power network into a first type of node satisfying a preset association condition with the first entry node and a second type of node not satisfying the preset association condition.

[0025] According to the node information, the nodes in the computing power network are divided into the first type of nodes satisfying the preset association condition with the first entry node and the second type of nodes not satisfying the preset association condition.

[0026] Optionally, according to the node information, the nodes in the computing power network are divided into the first type of nodes satisfying the preset association condition with the first entry node and the second type of nodes not satisfying the preset association condition, specifically including:

[0027] According to the location information in the node information, the part of nodes in the computing power network within the preset proximity range of the first entry node are divided into the first type of nodes associated with the first entry node.

[0028] According to the access service information in the first entry node information, the part of nodes in the computing power network having service association with the first entry node are divided into the first type of nodes associated with the first entry node.

[0029] The remaining nodes in the computing power network are divided into the second type of nodes associated with the first entry node.

[0030] Optionally, according to the node information, the nodes in the computing power network are divided into the first type of nodes satisfying the preset association condition with the first entry node and the second type of nodes not satisfying the preset association condition, specifically including:

[0031] According to the node information, the nodes in the computing power network are divided into the first type of nodes satisfying the preset association condition with the first entry node and the second type of nodes not satisfying the preset association condition.

[0032] According to the node information, the nodes in the computing power network are divided into the first type of nodes satisfying the preset association condition with the first entry node and the second type of nodes not satisfying the preset association condition.

[0033] The second preset time interval is greater than the first preset time interval, and the second preset location area is greater than the first preset location area.

[0034] Optionally, the method further comprises: judging whether the variation amplitude of each parameter in the aggregated node information exceeds a preset threshold value, and if so, taking the current value of the parameter whose variation amplitude exceeds the preset threshold value as the delivery information for the first entry node.

[0035] In response to the node in the computing power network being a computing power node, it is determined whether a variation range of each computing power node resource information parameter in the aggregated computing power node information exceeds a first preset threshold value, and if so, a current value of the computing power node resource information parameter whose variation range exceeds the first preset threshold value is taken as the delivery information of the first entry node.

[0036] In response to the node in the computing power network being a network node, each network link from the first entry node to a certain computing power node connected by network node connection information in the aggregated network node information is determined, and it is determined whether a variation range of the network node performance information parameter of each network link exceeds a second preset threshold value, and if so, a current value of the network node performance information parameter whose variation range exceeds the second preset threshold value is taken as the delivery information of the first entry node.

[0037] Optionally, the variation range of the network node performance information parameter of each network link is specifically calculated according to the following formula:

[0038]

[0039] In the formula, S 2 is the variation range of a certain network node performance information parameter of a certain network link, n is the number of network nodes of the certain network link, Zi=|Xi-Yi| is the deviation of the current value Xi of the certain network node performance information parameter of the i∈(1, n)th network node from the historical value Yi of the last aggregation, and μ is the expected value of the certain network node performance information parameter.

[0040] Optionally, after the determination of whether each entry node has the routing decision capability, the method further includes:

[0041] If the second entry node does not have the routing decision capability, a routing decision scheme is formulated for the access service of the second entry node according to each node information, and the formulated routing decision scheme is taken as the delivery information of the second entry node, wherein the second entry node is an entry node without the routing decision capability.

[0042] In a second aspect, the application provides a computing power network information notification and routing decision method, applied to a computing power network entry node, and the method includes:

[0043] Receiving delivery information notified by an arrangement and control system of a computing power network, wherein the delivery information is formulated and notified by the arrangement and control system for each entry node in the computing power network according to each node information acquired and stored in real time.

[0044] Making routing decisions for services accessing the computing power network through the entry node according to the received delivery information.

[0045] Optionally, the routing decision is made for the service accessing the computing power network through the service itself according to the received distributed information, specifically including:

[0046] The services accessing the computing power network through the services themselves are sorted according to service categories and user priorities;

[0047] The routing decision is made for each sorted service in turn according to the received distributed information.

[0048] Optionally, the routing decision is made for each sorted service in turn according to the received distributed information, specifically including:

[0049] According to the computing power node information and the network node information in the received distributed information, a computing power node is selected for each sorted service in turn, and a network link composed of network nodes to the selected computing power node is selected.

[0050] In a third aspect, the present application provides a computing power network arrangement and control system, comprising:

[0051] A resource collection module is configured to acquire and store node information in the computing power network in real time;

[0052] A resource management module is connected to the resource collection module and configured to formulate distributed information for each entry node in the computing power network according to the node information;

[0053] A resource configuration module is connected to the resource management module and configured to notify each entry node of the distributed information, so that each entry node receives the distributed information notified to itself and makes a routing decision for the service accessing the computing power network through the service itself according to the received distributed information.

[0054] In a fourth aspect, the present application provides a computing power network entry node, comprising:

[0055] A resource receiving module is configured to receive distributed information notified by an arrangement and control system of the computing power network, wherein the distributed information is formulated and notified by the arrangement and control system for each entry node in the computing power network according to the node information acquired and stored in real time by the arrangement and control system;

[0056] A service scheduling module is connected to the resource receiving module and configured to make a routing decision for the service accessing the computing power network through the service itself according to the received distributed information.

[0057] In a fifth aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor runs the computer program stored in the memory, the processor executes the computing power network information notification and routing decision method as described above.

[0058] In a sixth aspect, the present application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the computing power network information notification and routing decision method.

[0059] The present application provides a computing power network information notification and routing decision method, an arrangement and control system, an entry node, a computer device and a computer readable storage medium. Through the arrangement and control system of the computing power network, the information of each node in the computing power network is acquired and stored in real time, and the information of each node is used to formulate the information issued by each entry node in the computing power network, and the information issued by each entry node is notified to each entry node, and each entry node makes routing decision for the service passing through the computing power network through itself, and the centralized information notification and distributed routing decision method is used to realize the information notification by using the mature information interaction mechanism, the node information is issued to each entry node, the data is avoided to be bulky, and meanwhile, the routing decision calculation complexity is reduced, the convergence speed is improved, and the demand for the scheduling of the computing power node and the network node in the computing power network is met. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 is a flowchart of a computing power network information notification and routing decision method in an embodiment of the present application;

[0061] Figure 2 is a structural schematic diagram of a computing power network in an embodiment of the present application;

[0062] Figure 3 is a flowchart of another computing power network information notification and routing decision method in an embodiment of the present application;

[0063] Figure 4 is a structural schematic diagram of a computing power network arrangement and control system in an embodiment of the present application;

[0064] Figure 5 is a structural schematic diagram of a computing power network entry node in an embodiment of the present application;

[0065] Figure 6 is a structural schematic diagram of a computer device in an embodiment of the present application. DETAILED DESCRIPTION

[0066] In order for those skilled in the art to better understand the technical solutions of the present application, the embodiments of the present application will be further described in detail below with reference to the drawings.

[0067] It can be understood that the specific embodiments and drawings described herein are only used to explain the present application, and not to limit the present application.

[0068] It can be understood that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0069] It can be understood that, for the convenience of description, only parts related to the present application are shown in the drawings of the present application, and parts irrelevant to the present application are not shown in the drawings.

[0070] It can be understood that each unit and module involved in the embodiments of the present application can correspond to only one entity structure, or can be composed of multiple entity structures, or multiple units and modules can be integrated into one entity structure.

[0071] It can be understood that the functions and steps marked in the flowcharts and block diagrams of the present application can occur in an order different from that marked in the drawings without conflict.

[0072] It can be understood that in the flowcharts and block diagrams of the present application, the architecture, functions and operations of possible implementations of the system, device, equipment and method according to the embodiments of the present application are shown. Each block in the flowchart or block diagram can represent a unit, module, program segment, code, which contains executable instructions for realizing the specified functions. Moreover, each block or combination of blocks in the block diagram and flowchart can be realized by a hardware-based system for realizing the specified functions, or by a combination of hardware and computer instructions.

[0073] It can be understood that the units and modules involved in the embodiments of the present application can be realized in the form of software or hardware, for example, the units and modules can be located in a processor.

[0074] Embodiment 1

[0075] As shown in Figure 1 Embodiment 1 of the present application provides a computing power network information notification and routing decision method, which comprises steps S11, S12 and S13.

[0076] Specifically, in the present embodiment, the method is applied to the orchestration and control system 1 of the computing power network as shown in Figure 2 The computing power network further comprises: an entry node 2, a network node 3 and a computing power node 4, the entry node 2 can be specifically an entry gateway of an access service device 5; a service requesting a computing power network service issued by a certain service device 5 accesses the computing power network through a certain entry node 2, and reaches a certain computing power node 4 using a network link formed by part of the network nodes 3, and after obtaining the computing power service of the certain computing power node 4, returns the computing power service result to the certain service device 5 through the network link; it should be noted that Figure 2For a simplified example of the computing power network of the present application, the actual computing power network structure contains more nodes and more service devices 5 than Figure 2 much more.

[0077] As shown in Figure 1 step S11, the information of each node in the computing power network is acquired and stored in real time.

[0078] Optionally, the information of each node specifically includes information of each computing power node 4.

[0079] The information of each computing power node 4 includes location information and resource information of the computing power node 4.

[0080] Optionally, the information of each node specifically further includes information of each network node 3 and information of each entry node 2.

[0081] The information of each network node 3 includes location information, connection information and performance information of the network node 3.

[0082] The information of each entry node 2 includes location information and access service information.

[0083] Optionally, the acquiring and storing of the information of each node in the computing power network in real time specifically includes:

[0084] receiving the self-change information of each computing power node 4, each network node 3 and each entry node 2 in the computing power network after the self-change information changes, or querying the self-change information of each computing power node 4, each network node 3 and each entry node 2 in the computing power network at a preset frequency;

[0085] storing the received or queried information of each computing power node 4 in a computing power node information database, storing the received or queried information of each network node 3 in a network node information database, and storing the received or queried information of each entry node 2 in an entry node information database.

[0086] Specifically, in the present embodiment, in the initial stage of the computing power network establishment, the computing power nodes 4, the network nodes 3 and the entry nodes 2 respectively report all their information to the arrangement and control system 1, or the arrangement and control system 1 collects all the information of the computing power nodes 4, the network nodes 3 and the entry nodes 2; in the process of the computing power network establishment and operation, the computing power nodes 4, the network nodes 3 and the entry nodes 2 respectively report their changed information to the arrangement and control system 1 in the case of their own information change, so as to reduce the data transmission amount, or the arrangement and control system 1 collects all the information of the computing power nodes 4, the network nodes 3 and the entry nodes 2 according to a preset frequency, for example, once every 5s / 30s / 1min / 5min / 30min, and the specific interval time is set according to the network demand; the arrangement and control system 1 pre-establishes the computing power node information database, the network node information database and the entry node information database, and stores all the collected information in the corresponding database; the information collected for the computing power nodes 4 includes the location information such as the location of the resource pool where the computing power nodes 4 are located, and the resource information such as the service type, the resource type, the total amount / remaining amount / utilization rate of the computing resource, the total amount / remaining amount / utilization rate of the storage resource, the number of service connections, the total amount / remaining amount / utilization rate of the external connection bandwidth, and the number of external connection links; the information collected for the network nodes 3 includes the location information such as the IP (Internet Protocol) address of the network nodes 3, the connection information such as the network link formed by the network nodes 3 and the nodes at both ends of the network link, and the performance information such as the link delay, the link packet loss rate, and the total amount / remaining amount / utilization rate of the link bandwidth; the information collected for the entry nodes 2 includes the location information such as the IP address of the entry nodes 2, and in the process of the computing power network operation, the information also includes the access service information such as the service name, the service type, the computing power nodes and the network links used by the service accessed by the entry nodes 2; when the entry nodes 2 are access gateways, they are essentially a kind of network nodes 3, and can also collect relevant information as network nodes 3 and store the information in the network node information database. It should be noted that the present application can also take into account the existing network nodes 3 of the computing power network which have arranged the distributed route advertisement protocol, at this time, since the network nodes 3 already have the capability of network information advertisement and diffusion, the network nodes 3 use the existing route advertisement protocol for advertisement and diffusion, the arrangement and control system 1 does not collect the information of each network node 3, and each network node 3 does not report its own information to the arrangement and control system 1, that is, each node information does not include the information of each network node 3, the arrangement and control system 1 processes the information of each computing power node 4 according to the method of the present application, and delivers the processed information of each computing power node 4 to each entry node 2.

[0087] As shown in Figure 1 Step S12, formulating respective delivery information for each entry node 2 in the computing power network according to each node information.

[0088] Optionally, the respective issuing information for each entry node 2 in the computing power network is formulated according to the node information, and specifically includes:

[0089] determining whether each entry node 2 has routing decision capability;

[0090] if the first entry node 2 has routing decision capability, aggregating the node information for the first entry node 2, wherein the first entry node 2 is an entry node 2 with routing decision capability;

[0091] further determining whether the change amplitude of each parameter in the aggregated node information exceeds a preset threshold, and if so, taking the current value of the parameter with the change amplitude exceeding the preset threshold as the issuing information for the first entry node 2.

[0092] Optionally, the aggregation of the node information for the first entry node 2 specifically includes:

[0093] according to the node information, dividing each node in the computing power network into a first type of node that meets a preset association condition with the first entry node 2 and a second type of node that does not meet the preset association condition;

[0094] for the first entry node 2, aggregating the node information of the first type of node according to a first granularity, and aggregating the node information of the second type of node according to a second granularity, the second granularity being greater than the first granularity.

[0095] Optionally, the division of each node in the computing power network into a first type of node that meets a preset association condition with the first entry node 2 and a second type of node that does not meet the preset association condition according to the node information specifically includes:

[0096] according to the location information in the node information, dividing part of the nodes in the computing power network within a preset proximity range of the first entry node 2 into the first type of node associated with the first entry node 2;

[0097] according to the access service information in the first entry node 2 information, dividing part of the nodes in the computing power network that have service association with the first entry node 2 into the first type of node associated with the first entry node 2;

[0098] dividing the remaining nodes in the computing power network into the second type of node associated with the first entry node 2.

[0099] Optionally, the node information of the first type of nodes is aggregated according to a first granularity, and the node information of the second type of nodes is aggregated according to a second granularity greater than the first granularity, for the first entry node 2, specifically including:

[0100] The node information of the first type of nodes is aggregated according to a first preset time interval and a first preset location area, for the first entry node 2.

[0101] The node information of the second type of nodes is aggregated according to a second preset time interval and a second preset location area, for the first entry node 2.

[0102] The second preset time interval is greater than the first preset time interval, and the second preset location area is greater than the first preset location area.

[0103] Optionally, the change amplitude of each parameter in the aggregated node information is determined, and if the change amplitude of each parameter exceeds a preset threshold, the current value of the parameter whose change amplitude exceeds the preset threshold is taken as the delivery information of the first entry node 2, specifically including:

[0104] If the node in the computing power network is a computing power node 4, the change amplitude of each computing power node 4 resource information parameter in the aggregated computing power node 4 information is determined, and if the change amplitude of each computing power node 4 resource information parameter exceeds a first preset threshold, the current value of the computing power node 4 resource information parameter whose change amplitude exceeds the first preset threshold is taken as the delivery information of the first entry node 2.

[0105] If the node in the computing power network is a network node, each network link from the first entry node 2 to a certain computing power node 4 connected by the network node 3 connection information in the aggregated network node 3 information is determined, and the change amplitude of the network node 3 performance information parameter of each network link is determined, and if the change amplitude of the network node 3 performance information parameter of each network link exceeds a second preset threshold, the current value of the network node 3 performance information parameter whose change amplitude exceeds the second preset threshold is taken as the delivery information of the first entry node 2.

[0106] Optionally, the change amplitude of the network node 3 performance information parameter of each network link is calculated according to the following formula:

[0107]

[0108] In the formula, S 2Xi-Yi| is the deviation of the current value Yi of the performance information parameter of the i-th network node 3 in (1, n) from the historical value Xi aggregated last time, and μ is the expected value of the performance information parameter of the network node 3.

[0109] Specifically, in the present embodiment, after acquiring and storing each node information, the arrangement and control system 1 reads the stored data from each database and performs aggregation processing; in the early stage of the establishment of the computing power network, the aggregation method can be performed in the time dimension and the position dimension, for example, the information collection is collected by the arrangement and control system 1 every 5s, the average value of the values in the information collected every 1min is calculated, a plurality of computing power nodes 4 in a certain preset position region are aggregated to obtain the total resource information, a plurality of network nodes 3 directly connected to a certain computing power node 4 are aggregated to obtain the total performance information, etc., and the aggregation can reduce the data amount transmitted from the arrangement and control system 1 to the entry node 2; different granularity of issued information can also be formulated for each entry node 2, for example, for a certain entry node 2, a plurality of computing power nodes 4 located in a certain set adjacent region thereof and the network nodes 3 of the network link from the certain entry node 2 to the plurality of computing power nodes 4 are determined as nodes associated with the certain entry node 2 with high degree (first type of nodes), the nodes with high degree of association are aggregated with smaller time interval, the resource information and performance information of the nodes are aggregated with smaller number of nodes, at this time, the aggregated information has smaller granularity (first granularity) and can represent more detailed node information, which is used as the issued information of the certain entry node 2 with the first granularity, the remaining nodes (second type of nodes) can be considered as nodes associated with the certain entry node 2 with low degree, and the related node information can be aggregated with longer time interval and more number of nodes, at this time, the aggregated information has larger granularity (second granularity) and can represent more generalized node information, in addition, in the process of running the computing power network, the nodes associated with the certain entry node 2 with high degree can also be determined according to the association condition of the service accessed by the certain entry node 2, for example, the computing power nodes 4 and the network nodes 3 whose services are being used by the service should notify the corresponding entry node 2 of more detailed node information, i.e., smaller granularity information, and the second type of nodes can be notified of larger granularity information, so that the transmission data amount is reduced while ensuring the effectiveness of the issued information.

[0110] Specifically, in the present embodiment, after obtaining and storing the node information, the orchestration and control system 1 reads the stored data from each database and determines whether data needs to be issued according to the change amplitude of the stored data before and after; Specifically, it can include calculating the change amplitude of a single computing power node 4, for example, the current value of a certain parameter of the resource information of a certain computing power node 4 is C2, and the last stored historical value is C1, the deviation of the two is calculated according to the following formula: If b>r, it is considered that the value of C2 needs to be issued, where r is the change ratio coefficient of a certain parameter value of a certain computing power node 4, which can be determined as a constant according to the characteristics of each parameter value, for example, the computing resource utilization rate is taken as 10% for the parameter, and the change of the computing resource utilization rate is defined as more than 10%, which is considered to be large and needs to be issued data, if b Where: Zi=|Xi-Yi|, if S 2 >=k, it is considered that the value of (Y1, Y2, Y3, …, Yn) needs to be issued, where k is the deviation variance coefficient of the newly collected value of a certain parameter of the network node 3 and the original value, which is determined as a constant according to different parameter types, for example, the link delay is taken as k=0.05ms, when the deviation degree of the newly collected delay value is more than 0.05ms in total, it is judged that the delay data of the network link changes greatly, and data needs to be issued, otherwise, it does not need to be issued. Data; using data change amplitude to control whether to issue data can also be combined with the above-mentioned information aggregation process, and the above calculation is performed according to the parameter value in the aggregated information, and information notification is performed based on variable frequency and granularity, etc., to obtain optimal processing efficiency, better reduce the amount of transmitted data and ensure the effectiveness of the issued information. For the case where each node information does not include network node 3 information, only the above method is used to process each computing power node 4 information.

[0111] As shown in Figure 1 Step S13, each entry node 2 is notified of the respective issued information, so that each entry node 2 receives the issued information notified to itself and makes routing decisions for services accessing the computing power network through itself according to the received issued information.

[0112] Specifically, in the present embodiment, the distribution information is announced by the orchestration control system 1 to each entry node 2, for example, the distribution information announcement can be realized through the way of south-north interface, so that the complex protocol mode such as BGP (Border Gateway Protocol), IGP (Interior Gateway Protocol), OSPF (Open Short Path First) and the like is not needed to be adopted to realize, thereby reducing the requirement for node device; the route decision refers to that when a certain service needs to use computing resource, a certain or certain computing nodes 4 meeting the computing resource requirement of the service are determined for the service, and a path from the entry node 2 accessed by the service to the computing node 4 needing to be used is selected, the route decision of each entry node 2 is performed for the route decision of its own service, thereby improving the calculation efficiency and convergence speed of the route decision, realizing fast traffic access and path optimization, and improving the service use experience.

[0113] Optionally, after judging whether each entry node 2 has the route decision capability, the method further comprises:

[0114] If the second entry node 2 does not have the route decision capability, a route decision scheme is made for the access service of the second entry node 2 according to the node information, and the made route decision scheme is taken as the distribution information for the second entry node 2, wherein the second entry node 2 is an entry node 2 without the route decision capability.

[0115] Specifically, in the present embodiment, considering that there may be entry nodes 2 without the route decision capability in the existing arranged computing power network, for these entry nodes 2, a step of judging whether each entry node 2 has the route decision capability can be added by the orchestration control system 1, if there is an entry node 2 without the route decision capability (second entry node 2), a route decision scheme is made for it by the orchestration control system 1, and then the route decision scheme is directly distributed to it, without the need of distributing the corresponding node information (i.e. steps S12 and S13 are not executed), and for the entry nodes 2 with the route decision capability (first entry nodes 2), steps S12 and S13 are continued to be executed.

[0116] Embodiment 1 realizes centralized information announcement from the perspective of the computing power network orchestration control system 1, adopts a mature information interaction mechanism, and distributes the node information to each entry node in a targeted manner to avoid data bloating, and then realizes distributed route decision by each entry node to realize the scheduling of the computing nodes and network nodes in the computing power network.

[0117] Embodiment 2:

[0118] AsFigure 3 As shown, the embodiment 2 of the present application provides a computing power network information notification and routing decision method, applied to a computing power network entry node, the method comprising:

[0119] S21, receiving the information issued by the arrangement and control system of the computing power network, the information issued by the arrangement and control system is the information of each node in the computing power network, which is acquired and stored in real time, and is respectively formulated and announced by each entry node in the computing power network according to the information of each node;

[0120] S22, according to the received information, making routing decision for the service through the access of the computing power network.

[0121] Optionally, the routing decision for the service through the access of the computing power network according to the received information, specifically includes:

[0122] sorting the service through the access of the computing power network according to the service category and the user priority;

[0123] According to the received information, making routing decision for each service in turn.

[0124] Optionally, the routing decision for each service in turn according to the received information, specifically includes:

[0125] According to the computing power node information and the network node information in the received information, selecting the computing power node for each service in turn, and selecting the network link composed of the network node to the selected computing power node.

[0126] Specifically, in the embodiment, the method corresponds to the method described in embodiment 1, and is specifically applied to a computing power network entry node as Figure 2The entry node 2 in the computing power network shown in this embodiment mainly refers to an entry node 2 with routing decision-making capabilities. It makes routing decisions for services accessing the computing power network by receiving the information issued by the orchestration and control system 1. When multiple services are accessed, the routing decisions can be sorted according to the service category and user priority. For example, when the service is a latency-sensitive service, the routing decision needs to be prioritized for the service, or when the user of the service has subscribed to priority service, the routing decision needs to be prioritized for the service, etc. Since the orchestration and control system 1 can only issue the corresponding information when the change of information of each node exceeds a preset threshold, the entry node 2 should also have a certain recording capability to record the node information required for its own routing decision. The routing decision is usually to arrange the computing power node 4 that meets the computing power requirements, has the best resources, and is the closest to the service, and at the same time select the network link composed of network nodes 3 with the best network performance to reach the computing power node 4. In Example 2, the computing power network entry node interacts with the orchestration and control system of Example 1 to complete computing power network information notification and routing decision. The specific process can be understood in conjunction with Example 1. From the perspective of the computing power network entry node, Example 2 realizes the reception of information notifications from the orchestration and control system, performs distributed routing decision, reduces the computational complexity of routing decision, improves the convergence speed, and realizes the scheduling of computing power nodes and network nodes in the computing power network.

[0127] Example 3:

[0128] like Figure 4 As shown, Embodiment 3 of the present invention provides a computing power network orchestration and control system, comprising:

[0129] Resource acquisition module 11 is used to acquire and store information of each node in the computing power network in real time;

[0130] Resource management module 12, connected to resource acquisition module 11, is used to formulate the respective information to be sent to each entry node in the computing power network based on the information of each node.

[0131] The resource configuration module 13, connected to the resource management module 12, is used to notify each entry node of its own distributed information, so that each entry node can receive the distributed information notified to itself and make routing decisions for services accessing the computing power network through itself based on the received distributed information.

[0132] Optionally, the node information specifically includes: information about each computing power node;

[0133] The computing node information includes: computing node location information and computing node resource information.

[0134] Optionally, the node information may further include: network node information and entry node information;

[0135] The network node information includes network node location information, network node connection information, and network node performance information.

[0136] The entry node information includes entry node location information and access service information.

[0137] Optionally, the resource collection module 11 specifically includes:

[0138] The real-time acquisition unit is configured to receive self-change information of each computing power node, each network node, and each entry node in the computing power network, which is reported in real time after the self-change information changes, or query the self information of each computing power node, each network node, and each entry node in the computing power network in real time according to a preset frequency.

[0139] The storage unit is connected with the real-time acquisition unit, and is configured to store each computing power node information received or queried in a computing power node information database, store each network node information received or queried in a network node information database, and store each entry node information received or queried in an entry node information database.

[0140] Optionally, the resource management module 12 specifically includes:

[0141] The judgment unit is configured to judge whether each entry node has routing decision capability.

[0142] The aggregation unit is connected with the judgment unit, and is configured to aggregate each node information for a first entry node if the first entry node has routing decision capability, wherein the first entry node is an entry node having routing decision capability.

[0143] The formulation unit is connected with the aggregation unit, and is configured to further judge whether a variation amplitude of each parameter in the aggregated node information exceeds a preset threshold value, and if so, take a current value of the parameter whose variation amplitude exceeds the preset threshold value as the delivery information for the first entry node.

[0144] Optionally, the aggregation unit specifically includes:

[0145] The division sub-unit is configured to divide each node in the computing power network into a first type of node satisfying a preset association condition with the first entry node and a second type of node not satisfying the preset association condition according to each node information.

[0146] The aggregation sub-unit is connected with the division sub-unit, and is configured to aggregate node information of the first type of node according to a first granularity and aggregate node information of the second type of node according to a second granularity for the first entry node, wherein the second granularity is greater than the first granularity.

[0147] Optionally, the dividing sub-unit is specifically used for:

[0148] dividing, according to position information in the node information, part of nodes in the computing power network within a preset proximity range of the first entry node as first-type nodes associated with the first entry node;

[0149] dividing, according to access service information in the first entry node information, part of nodes in the computing power network having a service association with the first entry node as first-type nodes associated with the first entry node;

[0150] dividing the rest of the nodes in the computing power network as second-type nodes associated with the first entry node.

[0151] Optionally, the aggregating sub-unit is specifically used for:

[0152] aggregating, for the first entry node, node information of the first-type nodes according to a first preset time interval and a first preset location area;

[0153] aggregating, for the first entry node, node information of the second-type nodes according to a second preset time interval and a second preset location area;

[0154] wherein the second preset time interval is greater than the first preset time interval, and the second preset location area is greater than the first preset location area.

[0155] Optionally, the formulating unit specifically includes:

[0156] a first formulating sub-unit configured to, in response to the nodes in the computing power network being computing power nodes, determine whether a variation amplitude of each computing power node resource information parameter in the aggregated computing power node information exceeds a first preset threshold value, and if so, take a current value of the computing power node resource information parameter whose variation amplitude exceeds the first preset threshold value as the first entry node's delivery information;

[0157] a second formulating sub-unit configured to, in response to the nodes in the computing power network being network nodes, determine each network link from the first entry node to a certain computing power node connected by network node connection information in the aggregated network node information, determine whether a variation amplitude of a network node performance information parameter of each network link exceeds a second preset threshold value, and if so, take a current value of the network node performance information parameter whose variation amplitude exceeds the second preset threshold value as the first entry node's delivery information.

[0158] Optionally, the variation range of network node performance information parameters in each network link of the second formulation subunit is specifically calculated according to the following formula:

[0159]

[0160] In the formula, S 2 Let be the variation range of a performance information parameter of a certain network node in a certain network link, n be the number of network nodes in the certain network link, Zi = |Xi - Yi| be the deviation difference between the current value Xi of the performance information parameter of the i-th network node in (1, n) and the historical value Yi of the last aggregation, and μ be the expected value of the performance information parameter of the certain network node.

[0161] Optionally, the resource management formulation module 12 is further used for:

[0162] If the second entry node does not have routing decision-making capabilities, a routing decision-making scheme is formulated for the access services of the second entry node based on the information of each node, and the formulated routing decision-making scheme is sent as information to the second entry node, wherein the second entry node is an entry node that does not have routing decision-making capabilities.

[0163] Example 3 provides a computing power network orchestration and control system that enables centralized information notification and adopts a mature information interaction mechanism. Node information is sent to each entry node in a targeted manner to avoid data bloat. Then, each entry node can make distributed routing decisions to achieve scheduling of computing power nodes and network nodes in the computing power network.

[0164] Example 4:

[0165] like Figure 5 As shown, Embodiment 4 of the present invention provides a computing power network entry node, including:

[0166] The resource receiving module 21 is used to receive the information issued by the orchestration and control system of the computing power network. The information is formulated and issued by the orchestration and control system for each entry node in the computing power network based on the information of each node after the system has acquired and stored the information of each node in the computing power network in real time.

[0167] The service scheduling module 22 is connected to the resource receiving module 21 and is used to make routing decisions for services that access the computing power network through itself based on the received information.

[0168] Optionally, the service scheduling module 22 specifically includes:

[0169] The sorting unit is used to sort services that access the computing network through itself according to service category and user priority.

[0170] A decision unit is connected with the sorting unit, and is configured to make routing decisions for each service in sequence according to the received information.

[0171] Optionally, the decision unit is specifically configured to:

[0172] According to the computing power node information and the network node information in the received information, a computing power node is selected for each service in sequence after sorting, and a network link composed of network nodes to the selected computing power node is selected.

[0173] Embodiment 4 provides a computing power network entry node, which receives information notification of an orchestration management system, makes distributed routing decisions, reduces routing decision calculation complexity, improves convergence speed, and realizes scheduling of computing power nodes and network nodes in the computing power network.

[0174] Embodiment 5:

[0175] As shown in Figure 6 Embodiment 5 of the present application provides a computer device, which includes a memory 10 and a processor 20, the memory 10 stores a computer program, and when the processor 20 runs the computer program stored in the memory 10, the processor 20 executes the computing power network information notification and routing decision method as described in Embodiment 1 or 2.

[0176] The memory 10 is connected with the processor 20, the memory 10 can adopt a flash memory or a read-only memory or other memories, and the processor 20 can adopt a central processing unit or a single-chip microcomputer.

[0177] Embodiment 5 provides a computer device for realizing centralized information notification, using a mature information interaction mechanism, and issuing node information in a targeted manner to avoid data bloating, and realizing distributed routing decisions, reducing routing decision calculation complexity, improving convergence speed, and realizing scheduling of computing power nodes and network nodes in the computing power network.

[0178] Embodiment 6:

[0179] Embodiment 6 of the present application provides a computer readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computing power network information notification and routing decision method as described in Embodiment 1 or 2 is realized.

[0180] The computer-readable storage medium includes volatile or non-volatile, removable or non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, computer program modules or other data. The computer-readable storage medium includes, but is not limited to, RAM (Random Access Memory), ROM (Read-Only Memory), EEPROM (Electrically Erasable Programmable read only memory), flash memory or other memory technology, CD-ROM (Compact Disc Read-Only Memory), digital versatile disc (DVD) or other optical disc storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer.

[0181] Embodiment 6 provides a computer-readable storage medium for storing a computer program, which implements centralized information notification, adopts a mature information interaction mechanism, and node information is targetedly issued to avoid data bloating, and distributed routing decision is realized to reduce routing decision calculation complexity, improve convergence speed, and realize scheduling of computing power nodes and network nodes in a computing power network.

[0182] Embodiments 1-6 of the present application provide a computing power network information notification and routing decision method, an arrangement and control system, an entry node, a computer device and a computer-readable storage medium. Through the arrangement and control system of the computing power network, each node information in the computing power network is acquired and stored in real time, each entry node in the computing power network is formulated with respective issued information according to each node information, each entry node is notified with respective issued information, each entry node respectively performs routing decision for a service passing through the entry node and accessing the computing power network, centralized information notification and distributed routing decision are adopted, a mature information interaction mechanism is adopted to realize information notification, node information is targetedly issued to each entry node to avoid data bloating, and at the same time, routing decision for a service of each entry node is performed by the entry node, routing decision calculation complexity is reduced, convergence speed is improved, and the demand for simultaneous scheduling of computing power nodes and network nodes in the computing power network can be met.

[0183] It can be understood that the above embodiments are only exemplary embodiments adopted for illustrating the principles of the present application, and the present application is not limited thereto. Various modifications and improvements can be made by those skilled in the art without departing from the spirit and essence of the present application, and these modifications and improvements are also regarded as the protection scope of the present application.

Claims

1. A method for computing power network information notification and routing decision, characterized in that, The method is applied to a computing power network arrangement management and control system, and comprises the following steps: Real-time acquisition and storage of node information in the computing power network; Formulation of respective delivery information for each entry node in the computing power network according to the node information, wherein the delivery information formulated for a first entry node with routing decision capability is targeted node information; Announcement of the respective delivery information to each entry node, so that each entry node receives the delivery information announced to itself, and the first entry node makes routing decisions for services accessing the computing power network through itself according to the received delivery information.

2. The method of claim 1, wherein, The node information specifically comprises: computing power node information; The computing power node information comprises: computing power node location information and computing power node resource information.

3. The method of claim 2, wherein, The node information specifically further comprises: network node information and entry node information; The network node information comprises: network node location information, network node connection information and network node performance information; The entry node information comprises: entry node location information and access service information.

4. The method of claim 3, wherein, The real-time acquisition and storage of node information in the computing power network specifically comprises: Real-time reporting of self-change information by each computing power node, each network node and each entry node in the computing power network after a change in self-information, or real-time querying of the information of each computing power node, each network node and each entry node in the computing power network according to a preset frequency; Storage of the received or queried computing power node information in a computing power node information database, storage of the received or queried network node information in a network node information database, and storage of the received or queried entry node information in an entry node information database.

5. The method according to any one of claims 1 to 4, characterized in that, The formulation of respective delivery information for each entry node in the computing power network according to the node information, wherein the delivery information formulated for a first entry node with routing decision capability is targeted node information, specifically comprises: Judgment of whether each entry node has routing decision capability; If the first entry node has routing decision capability, aggregation of the node information for the first entry node, wherein the first entry node is an entry node with routing decision capability; Further judgment of whether the change amplitude of each parameter in the aggregated node information exceeds a preset threshold, and if so, the current value of the parameter with the change amplitude exceeding the preset threshold is taken as the delivery information for the first entry node.

6. The method of claim 5, wherein, The aggregation of the node information for the first entry node specifically comprises: According to the node information, the nodes in the computing power network are divided into first-type nodes satisfying a preset association condition with the first entry node and second-type nodes not satisfying the preset association condition; For the first entry node, the node information of the first-type nodes is aggregated at a first granularity, and the node information of the second-type nodes is aggregated at a second granularity, wherein the second granularity is greater than the first granularity.

7. The method of claim 6, wherein, The division of the nodes in the computing power network into first-type nodes satisfying a preset association condition with the first entry node and second-type nodes not satisfying the preset association condition according to the node information specifically comprises: According to the position information in the node information, nodes in the computing power network within a preset proximity range of the first entry node are divided into a first type of nodes associated with the first entry node; According to the access service information in the first entry node information, nodes in the computing power network having a service association with the first entry node are divided into a first type of nodes associated with the first entry node; The remaining nodes in the computing power network are divided into a second type of nodes associated with the first entry node.

8. The method of claim 6, wherein, For the first entry node, the node information of the first type of nodes is aggregated at a first granularity, and the node information of the second type of nodes is aggregated at a second granularity, the second granularity being greater than the first granularity, specifically including: For the first entry node, the node information of the first type of nodes is aggregated at a first preset time interval and a first preset location area; For the first entry node, the node information of the second type of nodes is aggregated at a second preset time interval and a second preset location area; Wherein, the second preset time interval is greater than the first preset time interval, and the second preset location area is greater than the first preset location area.

9. The method of claim 5, wherein, The method further comprises: If the second entry node does not have routing decision capability, a routing decision scheme is formulated for the access service of the second entry node according to the node information, and the formulated routing decision scheme is taken as the delivery information for the second entry node, wherein the second entry node is an entry node without routing decision capability. The method comprises:

10. The method of claim 9, wherein, ​ In the formula, S 2 is the variation range of a certain network link network node performance information parameter, n is the number of network nodes of the certain network link, Zi=|Xi-Yi| is the deviation of the current value Xi of the certain network node performance information parameter of the i∈(1, n)th network node from the last aggregated historical value Yi, and μ is the expected value of the certain network node performance information parameter.

11. The method of claim 5, wherein, ​ ​ 12. A computing power network information notification and routing decision method, characterized in that, ​ The first entrance node receives the distributed information announced by the orchestration and control system of the computing power network, and the distributed information is the information specially prepared for the first entrance node with routing decision-making capability by the orchestration and control system after real-time acquisition and storage of the node information in the computing power network. The first entrance node makes routing decisions for the services accessing the computing power network through itself according to the received distributed information.

13. The method of claim 12, wherein, The first entrance node makes routing decisions for the services accessing the computing power network through itself according to the received distributed information, specifically including: The first entrance node sorts the services accessing the computing power network through itself according to the service categories and user priorities. The first entrance node makes routing decisions for each of the sorted services in turn according to the received distributed information.

14. The method of claim 13, wherein, The first entrance node makes routing decisions for each of the sorted services in turn according to the received distributed information, specifically including: The first entrance node selects the computing power nodes for each of the sorted services in turn according to the computing power node information and network node information in the received distributed information, and selects the network links composed of network nodes to the selected computing power nodes.

15. A computing power network orchestration management and control system, characterized in that, including: The resource acquisition module is configured to acquire and store the node information in the computing power network in real time. The resource management module is connected with the resource acquisition module and configured to prepare the distributed information for each entrance node in the computing power network according to the node information, including the distributed information specially prepared for the first entrance node with routing decision-making capability. The resource configuration module is connected with the resource management module and configured to announce the distributed information to each entrance node, so that each entrance node receives the distributed information announced to itself, and the first entrance node makes routing decisions for the services accessing the computing power network through itself according to the received distributed information.

16. A hashpower network entry node, comprising: including: The resource receiving module is configured to receive the distributed information announced by the orchestration and control system of the computing power network, and the distributed information is the information specially prepared for the first entrance node with routing decision-making capability by the orchestration and control system after real-time acquisition and storage of the node information in the computing power network. The service scheduling module is connected with the resource receiving module and configured to make routing decisions for the services accessing the computing power network through the first entrance node according to the received distributed information.

17. A computer apparatus, comprising: The memory and the processor, the memory stores a computer program, when the processor runs the computer program stored in the memory, the processor executes the computing power network information announcement and routing decision method in any one of claims 1-11 or 12-14.

18. A computer-readable storage medium, characterized in that, The computer program is stored on the memory and executed by the processor to realize the computing power network information announcement and routing decision method in any one of claims 1-11 or 12-14.

Images