Computing power network node allocation method, device, electronic device and storage medium

Abstract

The embodiments of the present application provide a method, device, electronic device and storage medium for allocating computing power network nodes. The method includes: obtaining the device parameters of the device corresponding to the task to be allocated, and the node parameters of the computing power network nodes in the area where the task to be allocated is located; determining the task delay corresponding to the task to be allocated based on the device parameters and the node parameters; processing the task data of the task to be allocated based on a pre-built computing power scheduling model to obtain the computing power requirement value of the task to be allocated; determining the target computing power network node in the computing power network node according to the task delay and the computing power requirement value; and allocating the task to be allocated to the target computing power network node according to the task weight corresponding to the task to be allocated. The embodiments of the present application can achieve the purpose of selecting the best computing power network node for the task to be allocated.

Description

Technical Field

[0001] The present application relates to the field of Internet technology, and in particular to a computing power network node allocation method, device, electronic device and storage medium. Background Art

[0002] With the continuous development of the big data field, more and more concepts have been proposed and applied to production, and the concept of data lake is one of them. Its concept refers to the introduction of Alibaba Cloud: Data Lake is a centralized repository that can store structured and unstructured data of any scale and support big data and AI computing. As a core component of the cloud-native data lake architecture, Data Lake Formation (DLF) helps users to build cloud-native data lake solutions simply and quickly. Data Lake Formation provides unified management of metadata on the lake, enterprise-level permission control, and seamless connection with multiple computing engines, breaking data silos and gaining insights into business value.

[0003] In actual applications, the computing power network needs to first select the network and then select the computing power pool (cloud computing computing power network node or edge computing computing power network node). As the business volume increases, how to select the best computing power pool for the tasks to be processed is an urgent problem to be solved. Summary of the invention

[0004] The technical problem to be solved by the embodiments of the present application is to provide a computing power network node allocation method, device, electronic device and storage medium to select the most suitable computing power network node for the task to be allocated.

[0005] In a first aspect, an embodiment of the present application provides a method for allocating computing power network nodes, the method comprising:

[0006] Obtaining device parameters of the device corresponding to the task to be assigned, and node parameters of the computing power network nodes in the area where the task to be assigned is located;

[0007] Determining a task delay corresponding to the task to be assigned based on the device parameters and the node parameters;

[0008] Processing the task data of the task to be assigned based on a pre-built computing power scheduling model to obtain a computing power requirement value of the task to be assigned;

[0009] Determine a target computing power network node among the computing power network nodes according to the task delay and the computing power requirement value;

[0010] According to the task weight corresponding to the task to be assigned, the task to be assigned is assigned to the target computing power network node.

[0011] Optionally, the determining, based on the device parameters and the node parameters, a task delay corresponding to the task to be assigned includes:

[0012] Calculate a first transmission delay between the device and the computing power network node according to the device transmission power, the channel gain parameter between the device and the computing power network node, and the data transmission bandwidth of the communication link between the device and the computing power network node;

[0013] Calculate a first computing delay of the task to be assigned according to the computing power resources allocated to the device by the computing power network node and the ratio of the task to be assigned to the computing power network node;

[0014] Calculate a second transmission delay between the computing power network node and the cloud data center according to the network bandwidth provided by the computing power network node for the task to be assigned;

[0015] Calculating a second computing delay of the cloud data center according to computing resources allocated by the cloud data center to the task to be assigned;

[0016] Based on the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay, a task delay corresponding to the task to be assigned is determined.

[0017] Optionally, determining the task delay corresponding to the to-be-assigned task based on the first transmission delay, the first calculation delay, the second transmission delay, and the second calculation delay includes:

[0018] Calculate the sum of the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay, and use the sum of the delays as the task delay corresponding to the task to be assigned; or

[0019] A weighted average is performed on the first transmission delay, the first calculation delay, the second transmission delay, and the second calculation delay to obtain an average delay, and the average delay is used as the task delay corresponding to the task to be assigned.

[0020] Optionally, the processing of the task data of the task to be assigned based on the pre-built computing power scheduling model to obtain the computing power requirement value of the task to be assigned includes:

[0021] Obtaining the amount of task data corresponding to the task to be assigned;

[0022] The task data volume, computing chip resources and chip computing capability value are processed based on the computing power scheduling model to obtain the computing power requirement value of the task to be assigned.

[0023] Optionally, determining a target computing power network node among the computing power network nodes according to the task delay and the computing power requirement value includes:

[0024] Obtaining the idle computing power of the computing power network node;

[0025] According to the idle computing power, select candidate computing power network nodes whose idle computing power is greater than the computing power requirement value from the computing power network nodes;

[0026] The target computing power network node corresponding to the task to be assigned is screened out from the candidate computing power network nodes according to the task delay and the business demand index.

[0027] Optionally, the selecting a target computing power network node corresponding to the task to be assigned from the candidate computing power network nodes according to the task delay and the service demand index includes:

[0028] When the service demand index indicates that the task is processed with optimal resources, the computing power network node with the lowest task latency is selected from the candidate computing power network nodes as the target computing power network node; or

[0029] When the business demand index indicates processing the task at the lowest cost, the computing power network node with the highest task latency is selected from the candidate computing power network nodes as the target computing power network node.

[0030] Optionally, allocating the task to be allocated to the target computing power network node according to the task weight corresponding to the task to be allocated includes:

[0031] The task to be assigned and the target computing network node corresponding to the task to be assigned are stored in a task queue in the form of a key-value pair;

[0032] Call Flink stream data to transfer the tasks to be assigned in the task queue to the target computing power network node corresponding to the tasks to be assigned in the order of task weight from high to low.

[0033] In a second aspect, an embodiment of the present application provides a computing power network node allocation device, the device comprising:

[0034] A parameter acquisition module, used to acquire device parameters of the device corresponding to the task to be assigned, and node parameters of the computing power network nodes in the area where the task to be assigned is located;

[0035] A delay determination module, used to determine the task delay corresponding to the task to be assigned based on the device parameters and the node parameters;

[0036] A computing power requirement acquisition module is used to process the task data of the task to be assigned based on a pre-built computing power scheduling model to obtain the computing power requirement value of the task to be assigned;

[0037] A target node determination module, used to determine a target computing power network node among the computing power network nodes according to the task delay and the computing power requirement value;

[0038] The task allocation module is used to allocate the task to be allocated to the target computing power network node according to the task weight corresponding to the task to be allocated.

[0039] Optionally, the delay determination module includes:

[0040] A first transmission delay calculation unit, configured to calculate a first transmission delay between the device and the computing power network node according to a device transmission power, a channel gain parameter between the device and the computing power network node, and a data transmission bandwidth of a communication link between the device and the computing power network node;

[0041] A first computing delay calculation unit, configured to calculate a first computing delay of the task to be assigned according to the computing power resources allocated to the device by the computing power network node and the ratio of the task to be assigned to the computing power network node;

[0042] A second transmission delay calculation unit, configured to calculate a second transmission delay between the computing power network node and the cloud data center according to a network bandwidth provided by the computing power network node for the task to be assigned;

[0043] A second computing delay calculation unit, configured to calculate a second computing delay of the cloud data center according to computing resources allocated by the cloud data center to the task to be allocated;

[0044] A task delay determining unit is used to determine the task delay corresponding to the to-be-allocated task based on the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay.

[0045] Optionally, the task delay determination unit includes:

[0046] A first task delay acquisition subunit is used to calculate the sum of the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay, and use the sum of the delays as the task delay corresponding to the task to be assigned;

[0047] The second task delay acquisition subunit is used to perform weighted averaging on the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay to obtain an average delay, and use the average delay as the task delay corresponding to the task to be assigned.

[0048] Optionally, the computing power requirement acquisition module includes:

[0049] A task data volume acquisition unit, used to acquire the task data volume corresponding to the task to be assigned;

[0050] A computing power requirement value acquisition unit is used to process the task data volume, computing chip resources and chip computing capacity value based on the computing power scheduling model to obtain the computing power requirement value of the task to be assigned.

[0051] Optionally, the target node determination module includes:

[0052] An idle computing power acquisition unit, used to acquire the idle computing power of the computing power network node;

[0053] A candidate node screening unit, configured to screen, according to the idle computing power, candidate computing power network nodes whose idle computing power is greater than the computing power requirement value from the computing power network nodes;

[0054] The target node screening unit is used to screen out the target computing power network node corresponding to the task to be assigned from the candidate computing power network nodes according to the task delay and the business demand index.

[0055] Optionally, the target node screening unit includes:

[0056] A first target node acquisition subunit is used to select a computing power network node with the lowest task latency from the candidate computing power network nodes as the target computing power network node when the business demand index indicates that the task is processed with optimal resources;

[0057] The second target node acquisition subunit is used to select the computing power network node with the highest task latency from the candidate computing power network nodes as the target computing power network node when the business demand index indicates processing the task at the lowest cost.

[0058] Optionally, the task allocation module includes:

[0059] A task storage unit, used to store the to-be-assigned task and the target computing network node corresponding to the to-be-assigned task in a task queue in the form of a key-value pair;

[0060] The task allocation unit is used to call the Flink stream data to transfer the to-be-allocated tasks in the task queue to the target computing power network node corresponding to the to-be-allocated tasks in the order of task weight from high to low.

[0061] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0062] A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, any of the above-mentioned methods for allocating computing power network nodes is implemented.

[0063] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium. When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute any of the computing power network node allocation methods described above.

[0064] Compared with the prior art, the embodiments of the present application have the following advantages:

[0065] In an embodiment of the present application, the device parameters of the device corresponding to the task to be assigned and the node parameters of the computing power network nodes in the area where the task to be assigned is located are obtained. Based on the device parameters and the node parameters, the task delay corresponding to the task to be assigned is determined. The task data of the task to be assigned is processed based on the pre-built computing power scheduling model to obtain the computing power requirement value of the task to be assigned. According to the task delay and the computing power requirement value, the target computing power network node in the computing power network node is determined. According to the task weight corresponding to the task to be assigned, the task to be assigned is assigned to the target computing power network node. The embodiment of the present application screens out the target computing power network node by combining the task to be assigned with the task delay of the computing power node network and the computing power required, so as to achieve the purpose of selecting the best computing power network node for the task to be assigned. At the same time, the task to be assigned is assigned to the target computing power network node according to the task weight corresponding to the task to be assigned, which can improve the processing efficiency of high-priority tasks and improve the user experience.

[0066] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 A flowchart of a method for allocating computing power network nodes provided in an embodiment of the present application;

[0068] Figure 2 A flowchart of a method for determining task delay provided in an embodiment of the present application;

[0069] Figure 3 A flowchart of a method for obtaining task delay provided in an embodiment of the present application;

[0070] Figure 4 A flowchart of a method for obtaining a computing power requirement value provided in an embodiment of the present application;

[0071] Figure 5 A flowchart of a method for selecting target computing power network nodes provided in an embodiment of the present application;

[0072] Figure 6 A flowchart of another method for selecting target computing power network nodes provided in an embodiment of the present application;

[0073] Figure 7 A flowchart of a method for allocating tasks to be allocated provided in an embodiment of the present application;

[0074] Figure 8 A schematic diagram of a real-time data warehouse architecture provided in an embodiment of the present application;

[0075] Fig. 9 A schematic diagram of the workflow of a Flink SQL engine provided in an embodiment of the present application;

[0076] Fig.10 A schematic diagram of the structure of a computing power network node allocation device provided in an embodiment of the present application;

[0077] Fig.11 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0078] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0079] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0080] Reference Figure 1 , shows a flowchart of the steps of a computing power network node allocation method provided by an embodiment of the present application, such as Figure 1 As shown, the computing power network node allocation method may include: step 101, step 102, step 103, step 104 and step 105.

[0081] Step 101: Obtain device parameters of a device corresponding to a task to be assigned, and node parameters of a computing network node in an area where the task to be assigned is located.

[0082] The embodiments of the present application can be applied to the scenario of allocating the best computing power network node to the task to be assigned.

[0083] In the specific implementation, a real-time data warehouse based on iceberg data lake technology can be deployed on the central server. In terms of form, iceberg is a table format. That is, it is an intermediate layer based on the computing layer (Flink, Spark, Hive, Presto) and the storage layer (ORC, Parquet, AVRO). It provides a basic environment for performance adaptation and optimization of big data computing. The architecture of the deployed real-time data warehouse can be as follows: Figure 8 As shown. Figure 8 As shown in the figure, the computing layer includes Spark, Flink, Hive and Presto, and the storage layer includes ORC, Parquet and AVRO. The table format layer is Table Format, and the key-value mapping relationship is saved through Redis to indicate the corresponding relationship between tasks and computing network nodes.

[0084] In practical applications, distributed computing task nodes (i.e. computing network nodes) can be created. The total delay of a computing task is calculated through the model algorithm, including the transmission delay from the user to the edge computing node, the processing delay of the edge node, the transmission delay between the edge node and the computing node of the cloud data center, and the computing delay of the cloud computing node, that is, the total delay of the computing task in the system to reach the edge computing node. And the optimized average delay. The average delay from the network buried point to the table format is obtained, and the network computing node is selected according to the weight of the buried point to complete the computing task.

[0085] The device parameters refer to the parameters of the device that issues the task to be assigned. In this example, the device parameters may include parameters such as the transmission power of the device.

[0086] Node parameters refer to the parameters of the computing network nodes in the area where the task to be assigned is located. In this example, the node parameters may include: computing resources allocated to the task, channel gain and other parameters.

[0087] After obtaining the task to be assigned, the device parameters of the device corresponding to the task to be assigned and the node parameters of the computing network nodes in the area where the task to be assigned is located can be obtained. Specifically, when allocating computing network nodes to the task to be assigned, it is necessary to allocate them in combination with the computing network nodes in the area where the task to be assigned is located. For example, taking two provinces as an example, the tasks to be assigned in Province A can only be allocated by selecting computing network nodes from Province A, and the tasks to be assigned in Province B can only be allocated by selecting computing network nodes from Province B, etc., and computing network nodes cannot be selected across provinces.

[0088] It can be understood that the above examples are merely examples listed for a better understanding of the technical solutions of the embodiments of the present application, and are not intended to be the sole limitation to the embodiments.

[0089] After obtaining the device parameters of the device corresponding to the task to be assigned and the node parameters of the computing network nodes in the area where the task to be assigned is located, step 102 is executed.

[0090] Step 102: Determine the task delay corresponding to the task to be assigned based on the device parameters and the node parameters.

[0091] After obtaining the device parameters of the device corresponding to the task to be assigned and the node parameters of the computing network nodes in the area where the task to be assigned is located, the task delay corresponding to the task to be assigned can be determined based on the device parameters and node parameters. In this example, the task delay can be calculated by the transmission delay from the user (i.e., user equipment (UE)) to the edge computing node, the processing delay of the edge node, the transmission delay between the edge node and the computing node of the cloud data center, and the computing delay of the computing node of the cloud data center. The task delay acquisition process can be combined with Figure 2 This is described in detail as follows.

[0092] Reference Figure 2 , shows a flowchart of the steps of a method for determining task delay provided by an embodiment of the present application, such as Figure 2 As shown, the task delay determination method may include: step 201, step 202, step 203, step 204 and step 205.

[0093] Step 201: Calculate the first transmission delay between the device and the computing power network node based on the device transmission power, the channel gain parameter between the device and the computing power network node, and the data transmission bandwidth of the communication link between the device and the computing power network node.

[0094] In this embodiment, the first transmission delay refers to the delay of transmitting tasks between the device and the computing power network node.

[0095] The acquired device parameters include the device transmission power, and the node parameters may include the channel gain parameters between the device and the computing power network node and the data transmission bandwidth of the communication link between the device and the computing power network node.

[0096] After obtaining the device parameters and node parameters, the first transmission delay between the device and the computing network node can be calculated based on the device transmission power contained in the device parameters, the channel gain parameter between the device and the computing network node contained in the node parameters, and the data transmission bandwidth of the communication link between the device and the computing network node. The calculation of the first transmission delay can be shown in the following formula (1):

[0097]

[0098] In the above formula (1), is the first transmission delay, p i is the transmit power of the ith device; h k,i is the channel gain from the i-th user terminal to the k-th edge node, which is a random independent and identically distributed variable; σ 2 is the additive white Gaussian noise power, and B is the data transmission bandwidth of the wireless communication link.

[0099] Step 202: Calculate the first computing delay of the task to be assigned based on the computing power resources allocated to the device by the computing power network node and the ratio of the task to be assigned to the computing power network node.

[0100] The first computing delay refers to the delay of the computing network node in calculating the task to be assigned.

[0101] The obtained node parameters include the computing power resources allocated to the device by the computing power network node, and the proportion of the tasks to be assigned to the computing power network node.

[0102] After obtaining the node parameters, the first computing delay of the task to be assigned can be calculated based on the computing resources allocated to the device by the computing network node contained in the node parameters and the ratio of the task to be assigned to the computing network node. The calculation method for the first computing delay can be shown in the following formula (2):

[0103]

[0104] In the above formula (2), is the first calculation delay, λ i is the proportion of the computing task of the i-th user device assigned to its corresponding edge computing node (i.e., computing power network node), λ i The interval is [0,1], then 1-λ i , a proportion of computing tasks are offloaded to cloud data centers, using Represents the computing resources allocated by the k-th edge computing node to the i-th user device.

[0105] Step 203: Calculate the second transmission delay between the computing power network node and the cloud data center based on the network bandwidth provided by the computing power network node for the task to be assigned.

[0106] The second transmission delay refers to the delay in transmitting the tasks to be assigned between the computing network nodes and the cloud data center.

[0107] The acquired node parameters include the network bandwidth provided by the computing power network node for the task to be assigned. After acquiring the node parameters, the second transmission delay between the computing power network node and the cloud data center can be calculated based on the network bandwidth provided by the computing power network node for the task to be assigned contained in the node parameters. The calculation method for the second transmission delay can be shown in the following formula (3):

[0108]

[0109] In the above formula (3), is the second transmission delay, W k,i is the network bandwidth provided by the computing power network node for the tasks to be assigned, i The ratio of the computing task of the i-th user device allocated to its corresponding edge computing node (i.e., computing power network node).

[0110] Step 204: Calculate a second computing delay of the cloud data center according to the computing resources allocated by the cloud data center to the task to be assigned.

[0111] The second computing delay refers to the delay of computing the task to be assigned using the computing resources allocated by the operation data center for the task to be assigned.

[0112] The obtained node parameters include the computing resources allocated by the cloud data center for the task to be assigned. After obtaining the node parameters, the second computing delay of the cloud data center can be calculated based on the computing resources allocated by the cloud data center for the task to be assigned contained in the node parameters. The calculation method for the second computing delay can be combined with the following formula (4):

[0113]

[0114] In the above formula (4), is the second computation delay, f i c is the computing resources allocated by the cloud data center for the tasks to be assigned, i The ratio of the computing task of the i-th user device allocated to its corresponding edge computing node (i.e., computing power network node).

[0115] After the first calculation delay, the second calculation delay, the first transmission delay and the second transmission delay are obtained through the above steps, step 205 is performed.

[0116] Step 205: Determine a task delay corresponding to the task to be assigned based on the first transmission delay, the first calculation delay, the second transmission delay, and the second calculation delay.

[0117] After obtaining the first calculation delay, the second calculation delay, the first transmission delay and the second transmission delay, the task delay corresponding to the task to be assigned may be determined based on the first calculation delay, the second calculation delay, the first transmission delay and the second transmission delay.

[0118] In a specific implementation, the total delay of the first calculation delay, the second calculation delay, the first transmission delay, and the second transmission delay can be calculated as the task delay corresponding to the task to be assigned. The first calculation delay, the second calculation delay, the first transmission delay, and the second transmission delay can also be weighted averaged to obtain an average delay as the task delay corresponding to the task to be assigned. Figure 3 This is described in detail as follows.

[0119] Reference Figure 3 , shows a flowchart of the steps of a method for obtaining task delay provided by an embodiment of the present application, such as Figure 3 As shown, the task delay acquisition method may include: step 301 and step 302.

[0120] Step 301: Calculate and obtain the sum of the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay, and use the sum of the delays as the task delay corresponding to the task to be assigned.

[0121] In this embodiment, after obtaining the first calculation delay, the second calculation delay, the first transmission delay, and the second transmission delay, the sum of the first calculation delay, the second calculation delay, the first transmission delay, and the second transmission delay can be calculated, and the sum of the delays can be used as the task delay corresponding to the task to be assigned. The calculation formula is shown in the following formula (5):

[0122]

[0123] In the above formula (5), t i is the total delay, is the first transmission delay, is the first calculation delay, is the second transmission delay, Calculate the delay for the second.

[0124] Step 302: performing weighted averaging on the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay to obtain an average delay, and using the average delay as the task delay corresponding to the task to be assigned.

[0125] In the specific implementation, in the wireless time domain, latency is one of the important characteristics for measuring system performance. The latency characteristics of the system can be measured by the sum of the task queue lengths in the cloud, network, edge, and end. Considering the dynamic queue characteristics of edge nodes and cloud data center nodes, the average latency can be used as the task latency of the task to be assigned.

[0126] After obtaining the first calculation delay, the second calculation delay, the first transmission delay, and the second transmission delay, a weighted average may be performed on the first calculation delay, the second calculation delay, the first transmission delay, and the second transmission delay to obtain an average delay, and the average delay is used as the task delay corresponding to the task to be assigned. The calculation method of the average delay may be as shown in the following formula (6):

[0127]

[0128] In the above formula (6), S k (t) is the computing task queue unloaded to the server side of the cloud data center at time t, Q k (t) is the task queue existing on the computing network node, and t represents the tth decision time.

[0129] After the task delay corresponding to the task to be assigned is determined based on the device parameters and the node parameters, step 104 is executed.

[0130] Step 103: Process the task data of the task to be assigned based on the pre-built computing power scheduling model to obtain the computing power requirement value of the task to be assigned.

[0131] The computing power scheduling model refers to a pre-built model used to calculate the required computing power of tasks to be assigned.

[0132] The computing power requirement value refers to the computing power value required to process the task to be assigned. Computing power (also known as hash rate) is a unit of measurement for the processing power of the Bitcoin network. It is the speed at which the computer calculates the output of the hash function. The Bitcoin network must perform intensive mathematical and cryptographic related operations for security purposes. For example, when the network reaches a hash rate of 10Th / s, it means that it can perform 10 trillion calculations per second.

[0133] After obtaining the tasks to be assigned, the task data of the tasks to be assigned can be processed based on the pre-built computing power scheduling model to obtain the computing power requirement value of the tasks to be assigned. Figure 4This is described in detail as follows.

[0134] Reference Figure 4 , which shows a flowchart of a method for obtaining a computing power requirement value provided by an embodiment of the present application, such as Figure 4 As shown, the method for obtaining the computing power requirement value may include: step 401 and step 402.

[0135] Step 401: Obtain the task data volume corresponding to the task to be assigned.

[0136] In this embodiment, after the tasks to be assigned are acquired, the task data volume of the tasks to be assigned may be acquired.

[0137] After the task data volume of the task to be assigned is obtained, step 402 is executed.

[0138] Step 402: Based on the computing power scheduling model, the task data volume, computing chip resources and chip computing capacity value are processed to obtain the computing power requirement value of the task to be assigned.

[0139] After obtaining the task data volume of the task to be assigned, the computing power scheduling model can be called to process the task data volume, computing chip resources and chip computing power value to obtain the computing power requirement value of the task to be assigned.

[0140] In this example, the computing power scheduling model is shown in the following formula (7):

[0141]

[0142] In the above formula (7), C br is the total computing power requirement; f(x) is the mapping function; q is the redundant computing power. Taking parallel computing power as an example, assuming there are three different types of parallel computing chip resources: b1, b2, and b3, then f(β j ) represents the mapping function of the parallel computing capability that the j-th parallel computing chip can provide, and q2 represents the redundant computing power of parallel computing.

[0143] After processing the task data of the task to be assigned based on the pre-built computing power scheduling model to obtain the computing power requirement value of the task to be assigned, step 104 is executed.

[0144] Step 104: Determine a target computing power network node among the computing power network nodes according to the task delay and the computing power requirement value.

[0145] The target computing power network node refers to the computing power network node used to process the task to be assigned in the computing power network node.

[0146] After obtaining the task delay and computing power requirement value corresponding to the task to be assigned, the target computing power network node in the computing power network node can be determined according to the task delay and computing power requirement value.

[0147] In the specific implementation, the candidate computing network nodes with a computing power requirement greater than the task to be assigned can be selected from the computing network nodes according to the remaining computing power of the computing network nodes, and then the target computing network nodes corresponding to the task to be assigned can be selected from the candidate computing network nodes according to the task delay and business demand index. Figure 5 This is described in detail as follows.

[0148] Reference Figure 5 , shows a flowchart of the steps of a target computing power network node screening method provided by an embodiment of the present application, such as Figure 5 As shown, the target computing power network node screening method may include: step 501, step 502 and step 503.

[0149] Step 501: Obtain the idle computing power of the computing power network node.

[0150] In this embodiment, the idle computing power refers to the remaining available computing power of the computing network nodes in the area where the task to be assigned is located.

[0151] After obtaining the computing power network nodes in the area where the task to be assigned is located, the idle computing power of the computing power network nodes can be obtained. In a specific implementation, the computing power network nodes can report the remaining available idle computing power, or the idle computing power of the computing power network nodes can be calculated based on the computing power currently used by the computing power network nodes and the total computing power of the computing power network nodes. Specifically, the specific method for obtaining the computing power network nodes can be determined according to business needs, and this embodiment does not limit this.

[0152] After obtaining the idle computing power of the computing power network node, execute step 502.

[0153] Step 502: Based on the idle computing power, select candidate computing power network nodes whose idle computing power is greater than the computing power requirement value from the computing power network nodes.

[0154] A candidate computing power network node refers to a network node whose idle computing power is greater than the computing power requirement value required by the task to be assigned among the computing power network nodes in the area where the task to be assigned is located.

[0155] After obtaining the idle computing power of the computing power network nodes, candidate computing power network nodes whose idle computing power is greater than the computing power requirement value can be screened out from the computing power network nodes according to the idle computing power of the computing power network nodes. For example, the computing power network nodes in the area where the task to be assigned is located include: node 1, node 2, node 3, node 4 and node 5, the computing power requirement value of the task to be assigned is 15, the idle computing power of node 1 is 12, the idle computing power of node 2 is 18, the idle computing power of node 3 is 19, the idle computing power of node 4 is 29, and the idle computing power of node 5 is 44. At this time, node 2, node 3, node 4 and node 5 can be used as candidate computing power network nodes, etc.

[0156] It can be understood that the above examples are merely examples listed for a better understanding of the technical solutions of the embodiments of the present application, and are not intended to be the sole limitation to the embodiments.

[0157] After selecting candidate computing power network nodes whose idle computing power is greater than the computing power requirement value from the computing power network nodes according to the idle computing power, execute step 503.

[0158] Step 503: Filter out the target computing power network node corresponding to the task to be assigned from the candidate computing power network nodes according to the task delay and the business demand index.

[0159] The business demand index may be used to indicate a demand index of a device to which a task is to be assigned. In this example, the business demand index may include an index for processing a task with optimal resources and an index for processing a task with the lowest cost.

[0160] After selecting candidate computing power network nodes whose idle computing power is greater than the computing power demand value from the computing power network nodes according to the idle computing power, the target computing power network nodes corresponding to the to-be-assigned tasks can be selected from the candidate computing power network nodes according to the task delay and the business demand index. Specifically, the target computing power network nodes can be selected according to different business demand indexes. Figure 6 This is described in detail as follows.

[0161] Reference Figure 6 , shows a flowchart of the steps of a method for obtaining a target computing power network node provided by an embodiment of the present application, such as Figure 6 As shown, the method for acquiring the target computing power network node may include: step 601 and step 602.

[0162] Step 601: When the business demand index indicates that the task is processed with optimal resources, the computing power network node with the lowest task latency is selected from the candidate computing power network nodes as the target computing power network node.

[0163] In this embodiment, when the business demand index indicates that the task should be processed with the best resources, the computing network node with the lowest task delay can be selected from the candidate computing network nodes as the target computing network node, that is, the task is processed at the fastest speed.

[0164] Step 602: When the business demand index indicates that the task can be processed at the lowest cost, the computing power network node with the highest task latency is selected from the candidate computing power network nodes as the target computing power network node.

[0165] When the business demand index indicates that the task should be processed at the lowest cost, the computing network node with the highest task latency can be selected from the candidate computing network nodes as the target computing network node, that is, the task should be processed in the most cost-effective way.

[0166] After selecting the target computing power network node corresponding to the task to be assigned from the candidate computing power network nodes according to the task delay and the business demand index, execute step 105.

[0167] Step 105: Allocate the task to be assigned to the target computing power network node according to the task weight corresponding to the task to be assigned.

[0168] The task weight can be used to indicate the importance of the task to be assigned. In this example, the higher the task weight is, the more important the task to be assigned is. Conversely, the lower the task weight is, the less important the task to be assigned is.

[0169] After selecting the target computing network node corresponding to the task to be assigned from the candidate computing network nodes according to the task delay and business demand index, the task to be assigned can be assigned to the target computing network node according to the task weight corresponding to the task to be assigned. In this embodiment, the Flink data flow can be called to transfer the task to be assigned according to the task weight. Figure 7 Describe in detail.

[0170] Reference Figure 7 , shows a flowchart of a method for allocating tasks to be allocated provided by an embodiment of the present application, such as Figure 7 As shown, the method for allocating tasks to be allocated may include: step 701 and step 702.

[0171] Step 701: The task to be assigned and the target computing network node corresponding to the task to be assigned are stored in a task queue in the form of a key-value pair.

[0172] After determining the target computing network node corresponding to the task to be assigned, the task to be assigned and the target computing network node corresponding to the task to be assigned can be stored in the task queue in the form of a key-value pair, that is, key-value.

[0173] After the tasks to be assigned and the target computing network nodes corresponding to the tasks to be assigned are stored in the task queue in the form of key-value pairs, step 702 is executed.

[0174] Step 702: Call Flink stream data to transfer the tasks to be assigned in the task queue to the target computing power network node corresponding to the tasks to be assigned in order of task weight from high to low.

[0175] After storing the tasks to be assigned and the target computing network nodes corresponding to the tasks to be assigned in the task queue in the form of key-value pairs, you can call Flink streaming data to transfer the tasks to be assigned in the task queue to the target computing network nodes corresponding to the tasks to be assigned in the order of task weight from high to low.

[0176] In the specific implementation, the key-value storage form of redis can be used to connect with Flink to form Flink-connection-redis. By establishing a key-value mapping relationship, a session table is formed for FlinkSQL to call. The workflow of the Flink SQL engine can be as follows Fig. 9 shown.

[0177] like Fig. 9 As shown in the figure, the SQL / TableAPI has the following steps from input to compilation into an executable JobGraph:

[0178] 1. Convert SQL text / TableAPI code into a logical execution plan (Logical Plan);

[0179] 2. The logical execution plan is converted into a physical execution plan (Physical Plan) through the optimizer;

[0180] 3. After the Transformation is generated through code generation technology, it is further compiled into an executable JobGraph and submitted for execution.

[0181] JobGraph optimizes StreamGraph, such as setting which operators can be chained to reduce network overhead. In the graph structure of Flink tasks, some operators are chained together (reducing serialization and network overhead and improving efficiency).

[0182] SQL implementation details:

[0183] S1. Convert SQL text / TableAPI code into a logical execution plan;

[0184] S2, SQL / TableAPI converts SQL parsing into AST abstract syntax tree through the calcite framework;

[0185] S3, SQL Validator obtains metadata from the Catalog to verify expressions, table information, etc., and converts them into relational algebra expressions (RelNode);

[0186] S4. The Optimizer then converts the relational algebra expression into a logical execution plan in the initial state.

[0187] The embodiment of the present application uses the redis key-value storage format and combines it with FlinkSQL to complete data statistics. A key link in solving the data lake solution is the adaptation problem between data storage and computing engines. At the same time, when the business demand for AI analysis of network point information is received, the network point information and weights can be combined with flink streaming data in the form of redis to pass to the most suitable computing power network node.

[0188] The computing power network node allocation method provided in the embodiment of the present application obtains the device parameters of the device corresponding to the task to be allocated, and the node parameters of the computing power network nodes in the area where the task to be allocated is located. Based on the device parameters and the node parameters, the task delay corresponding to the task to be allocated is determined. The task data of the task to be allocated is processed based on the pre-built computing power scheduling model to obtain the computing power requirement value of the task to be allocated. According to the task delay and the computing power requirement value, the target computing power network node in the computing power network node is determined. According to the task weight corresponding to the task to be allocated, the task to be allocated is allocated to the target computing power network node. The embodiment of the present application screens out the target computing power network node by combining the task to be allocated with the task delay of the computing power node network and the computing power required, so as to achieve the purpose of selecting the best computing power network node for the task to be allocated. At the same time, the task to be allocated is allocated to the target computing power network node according to the task weight corresponding to the task to be allocated, which can improve the processing efficiency of high-priority tasks and improve the user experience.

[0189] Reference Fig.10 , shows a schematic diagram of the structure of a computing power network node allocation device provided in an embodiment of the present application, such as Fig.10 As shown, the computing power network node allocation device 1000 may include the following modules:

[0190] The parameter acquisition module 1001 is used to acquire the device parameters of the device corresponding to the task to be assigned, and the node parameters of the computing power network nodes in the area where the task to be assigned is located;

[0191] A delay determination module 1002, configured to determine a task delay corresponding to the task to be assigned based on the device parameters and the node parameters;

[0192] The computing power requirement acquisition module 1003 is used to process the task data of the task to be assigned based on the pre-built computing power scheduling model to obtain the computing power requirement value of the task to be assigned;

[0193] A target node determination module 1004 is used to determine a target computing power network node among the computing power network nodes according to the task delay and the computing power requirement value;

[0194] The task allocation module 1005 is used to allocate the task to be allocated to the target computing power network node according to the task weight corresponding to the task to be allocated.

[0195] Optionally, the delay determination module 1002 includes:

[0196] A first transmission delay calculation unit, configured to calculate a first transmission delay between the device and the computing power network node according to a device transmission power, a channel gain parameter between the device and the computing power network node, and a data transmission bandwidth of a communication link between the device and the computing power network node;

[0197] A first computing delay calculation unit, configured to calculate a first computing delay of the task to be assigned according to the computing power resources allocated to the device by the computing power network node and the ratio of the task to be assigned to the computing power network node;

[0198] A second transmission delay calculation unit, configured to calculate a second transmission delay between the computing power network node and the cloud data center according to a network bandwidth provided by the computing power network node for the task to be assigned;

[0199] A second computing delay calculation unit, configured to calculate a second computing delay of the cloud data center according to computing resources allocated by the cloud data center to the task to be allocated;

[0200] A task delay determining unit is used to determine the task delay corresponding to the to-be-allocated task based on the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay.

[0201] Optionally, the task delay determination unit includes:

[0202] A first task delay acquisition subunit is used to calculate the sum of the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay, and use the sum of the delays as the task delay corresponding to the task to be assigned;

[0203] The second task delay acquisition subunit is used to perform weighted averaging on the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay to obtain an average delay, and use the average delay as the task delay corresponding to the task to be assigned.

[0204] Optionally, the computing power requirement acquisition module 1003 includes:

[0205] A task data volume acquisition unit, used to acquire the task data volume corresponding to the task to be assigned;

[0206] The computing power requirement value acquisition unit is used to process the task data volume, computing chip resources and chip computing capacity value based on the computing power scheduling model to obtain the computing power requirement value of the task to be assigned.

[0207] Optionally, the target node determination module 1004 includes:

[0208] An idle computing power acquisition unit, used to acquire the idle computing power of the computing power network node;

[0209] A candidate node screening unit, configured to screen, according to the idle computing power, candidate computing power network nodes whose idle computing power is greater than the computing power requirement value from the computing power network nodes;

[0210] The target node screening unit is used to screen out the target computing power network node corresponding to the task to be assigned from the candidate computing power network nodes according to the task delay and the business demand index.

[0211] Optionally, the target node screening unit includes:

[0212] A first target node acquisition subunit is used to select a computing power network node with the lowest task latency from the candidate computing power network nodes as the target computing power network node when the business demand index indicates that the task is processed with optimal resources;

[0213] The second target node acquisition subunit is used to select the computing power network node with the highest task latency from the candidate computing power network nodes as the target computing power network node when the business demand index indicates processing the task at the lowest cost.

[0214] Optionally, the task assignment module 1005 includes:

[0215] A task storage unit, used to store the to-be-assigned task and the target computing network node corresponding to the to-be-assigned task in a task queue in the form of a key-value pair;

[0216] The task allocation unit is used to call the Flink stream data to transfer the to-be-allocated tasks in the task queue to the target computing power network node corresponding to the to-be-allocated tasks in the order of task weight from high to low.

[0217] The computing power network node allocation device provided in the embodiment of the present application obtains the device parameters of the device corresponding to the task to be allocated and the node parameters of the computing power network nodes in the area where the task to be allocated is located. Based on the device parameters and the node parameters, the task delay corresponding to the task to be allocated is determined. The task data of the task to be allocated is processed based on the pre-built computing power scheduling model to obtain the computing power requirement value of the task to be allocated. According to the task delay and the computing power requirement value, the target computing power network node in the computing power network node is determined. According to the task weight corresponding to the task to be allocated, the task to be allocated is allocated to the target computing power network node. The embodiment of the present application screens out the target computing power network node by combining the task to be allocated with the task delay of the computing power node network and the computing power required, so as to achieve the purpose of selecting the best computing power network node for the task to be allocated. At the same time, the task to be allocated is allocated to the target computing power network node according to the task weight corresponding to the task to be allocated, which can improve the processing efficiency of high-priority tasks and improve the user experience.

[0218] An embodiment of the present application also provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program implements the above-mentioned computing power network node allocation method when executed by the processor.

[0219] Fig.11 FIG. 1 is a schematic diagram showing the structure of an electronic device 1100 according to an embodiment of the present invention. Fig.11 As shown, the electronic device 1100 includes a central processing unit (CPU) 1101, which can perform various appropriate actions and processes according to computer program instructions stored in a read-only memory (ROM) 1102 or computer program instructions loaded from a storage unit 1108 to a random access memory (RAM) 1103. In the RAM 1103, various programs and data required for the operation of the electronic device 1100 can also be stored. The CPU 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0220] Multiple components in the electronic device 1100 are connected to the I / O interface 1105, including: an input unit 1106, such as a keyboard, a mouse, a microphone, etc.; an output unit 1107, such as various types of displays, speakers, etc.; a storage unit 1108, such as a disk, an optical disk, etc.; and a communication unit 1109, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 1109 allows the electronic device 1100 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0221] The various processes and processing described above may be performed by the processing unit 1101. For example, the method of any of the above embodiments may be implemented as a computer software program, which is tangibly contained in a computer-readable medium, such as the storage unit 1108. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 1100 via the ROM 1102 and / or the communication unit 1109. When the computer program is loaded into the RAM 1103 and executed by the CPU 1101, one or more actions in the method described above may be performed.

[0222] In addition, an embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the above-mentioned computing power network node allocation method is implemented.

[0223] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0224] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, devices, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0225] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, terminals (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0226] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0227] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal so that a series of operating steps are executed on the computer or other programmable terminal to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable terminal. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0228] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0229] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal. In the absence of further restrictions, the elements defined by the statement "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or terminal including the elements.

[0230] The above is a detailed introduction to a computing power network node allocation method, a computing power network node allocation device, an electronic device and a computer-readable storage medium provided by the present application. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A method for allocating computing power network nodes, characterized in that: The method comprises: Obtaining device parameters of the device corresponding to the task to be assigned, and node parameters of the computing power network nodes in the area where the task to be assigned is located; Determining a task delay corresponding to the task to be assigned based on the device parameters and the node parameters; Processing the task data of the task to be assigned based on a pre-built computing power scheduling model to obtain a computing power requirement value of the task to be assigned; Determine a target computing power network node among the computing power network nodes according to the task delay and the computing power requirement value; Allocate the task to be assigned to the target computing power network node according to the task weight corresponding to the task to be assigned; The determining, based on the device parameters and the node parameters, the task delay corresponding to the task to be assigned includes: Calculate a first transmission delay between the device and the computing power network node according to the device transmission power, the channel gain parameter between the device and the computing power network node, and the data transmission bandwidth of the communication link between the device and the computing power network node; Calculate a first computing delay of the task to be assigned according to the computing power resources allocated to the device by the computing power network node and the ratio of the task to be assigned to the computing power network node; Calculate a second transmission delay between the computing power network node and the cloud data center according to the network bandwidth provided by the computing power network node for the task to be assigned; Calculating a second computing delay of the cloud data center according to computing resources allocated by the cloud data center to the task to be assigned; Determine a task delay corresponding to the task to be assigned based on the first transmission delay, the first calculation delay, the second transmission delay, and the second calculation delay; The determining, according to the task delay and the computing power requirement value, a target computing power network node among the computing power network nodes comprises: Obtaining the idle computing power of the computing power network node; According to the idle computing power, select candidate computing power network nodes whose idle computing power is greater than the computing power requirement value from the computing power network nodes; Filtering the target computing power network node corresponding to the task to be assigned from the candidate computing power network nodes according to the task delay and the business demand index; The step of selecting a target computing power network node corresponding to the task to be assigned from the candidate computing power network nodes according to the task delay and the business demand index includes: When the service demand index indicates that the task is processed with optimal resources, the computing power network node with the lowest task latency is selected from the candidate computing power network nodes as the target computing power network node; or When the business demand index indicates processing the task at the lowest cost, selecting the computing power network node with the highest task latency from the candidate computing power network nodes as the target computing power network node; The allocating the task to be allocated to the target computing power network node according to the task weight corresponding to the task to be allocated includes: The task to be assigned and the target computing network node corresponding to the task to be assigned are stored in a task queue in the form of a key-value pair; Call Flink stream data to transfer the tasks to be assigned in the task queue to the target computing power network node corresponding to the tasks to be assigned in the order of task weight from high to low.

2. The method according to claim 1, characterized in that The determining, based on the first transmission delay, the first calculation delay, the second transmission delay, and the second calculation delay, the task delay corresponding to the task to be assigned includes: Calculate the sum of the first transmission delay, the first calculation delay, the second transmission delay and the second calculation delay, and use the sum of the delays as the task delay corresponding to the task to be assigned; or A weighted average is performed on the first transmission delay, the first calculation delay, the second transmission delay, and the second calculation delay to obtain an average delay, and the average delay is used as the task delay corresponding to the task to be assigned.

3. The method according to claim 1, characterized in that The processing of the task data of the task to be assigned based on the pre-built computing power scheduling model to obtain the computing power requirement value of the task to be assigned includes: Obtaining the amount of task data corresponding to the task to be assigned; The task data volume, computing chip resources and chip computing capability value are processed based on the computing power scheduling model to obtain the computing power requirement value of the task to be assigned.

4. A computing power network node allocation device, characterized in that: The device comprises: A parameter acquisition module, used to acquire device parameters of the device corresponding to the task to be assigned, and node parameters of the computing power network nodes in the area where the task to be assigned is located; A delay determination module, used to determine the task delay corresponding to the task to be assigned based on the device parameters and the node parameters; A computing power requirement acquisition module is used to process the task data of the task to be assigned based on a pre-built computing power scheduling model to obtain the computing power requirement value of the task to be assigned; A target node determination module, used to determine a target computing power network node among the computing power network nodes according to the task delay and the computing power requirement value; A task allocation module, used to allocate the task to be allocated to the target computing power network node according to the task weight corresponding to the task to be allocated; The delay determination module comprises: A first transmission delay calculation unit, configured to calculate a first transmission delay between the device and the computing power network node according to a device transmission power, a channel gain parameter between the device and the computing power network node, and a data transmission bandwidth of a communication link between the device and the computing power network node; A first computing delay calculation unit, configured to calculate a first computing delay of the task to be assigned according to the computing power resources allocated to the device by the computing power network node and the ratio of the task to be assigned to the computing power network node; A second transmission delay calculation unit, configured to calculate a second transmission delay between the computing power network node and the cloud data center according to a network bandwidth provided by the computing power network node for the task to be assigned; A second computing delay calculation unit, configured to calculate a second computing delay of the cloud data center according to computing resources allocated by the cloud data center to the task to be allocated; a task delay determining unit, configured to determine a task delay corresponding to the to-be-assigned task based on the first transmission delay, the first calculation delay, the second transmission delay, and the second calculation delay; The target node determination module includes: An idle computing power acquisition unit, used to acquire the idle computing power of the computing power network node; A candidate node screening unit, configured to screen, according to the idle computing power, candidate computing power network nodes whose idle computing power is greater than the computing power requirement value from the computing power network nodes; A target node screening unit, used to screen out a target computing power network node corresponding to the task to be assigned from the candidate computing power network nodes according to the task delay and the business demand index; The target node screening unit comprises: A first target node acquisition subunit is used to select a computing power network node with the lowest task latency from the candidate computing power network nodes as the target computing power network node when the business demand index indicates that the task is processed with optimal resources; A second target node acquisition subunit is configured to select, when the business demand index indicates processing the task at the lowest cost, a computing power network node with the highest task latency from the candidate computing power network nodes as the target computing power network node; The task allocation module includes: A task storage unit, used to store the to-be-assigned task and the target computing network node corresponding to the to-be-assigned task in a task queue in the form of a key-value pair; The task allocation unit is used to call Flink stream data to transfer the to-be-allocated tasks in the task queue to the target computing power network node corresponding to the to-be-allocated tasks in the order of task weight from high to low.

5. An electronic device, characterized in that: include: A processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the computing power network node allocation method described in any one of claims 1 to 3 when executing the program.

6. A computer-readable storage medium, characterized in that: When the instructions in the storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute the computing power network node allocation method described in any one of claims 1 to 3.

Images