Computing power management method and system and storage medium

Abstract

The invention provides a computing power management method and system and a storage medium, and the method comprises the steps: obtaining the distribution of all computing nodes in a network management range, and obtaining the computing power resource data and work logs of each computing node; based on the computing power resource data and the work log of each computing node, determining a computing power coefficient and a time adaptation coefficient of each computing node; when a burst calculation task occurs in the data network management range, acquiring a time point and a task level of the burst calculation task; and determining a target computing node for computing the burst computing task based on the time point and the task level of the burst computing task and the computing power coefficient and the time adaptation coefficient of each computing node. Therefore, under the condition that the computing nodes work in turn and the burst computing task occurs, the target computing node is quickly determined through the computing power coefficient and the time adaptation coefficient of each computing node, so that the burst computing task is processed by utilizing the target computing node, and the computing efficiency is improved.

Description

Technical Field

[0001] The present disclosure relates to the fields of cloud computing edge and big data edge technology, and in particular to a computing power management method, system and storage medium. Background Art

[0002] Computing power management in the data network is to allocate and manage the computing resources of all computing nodes within the management scope of the data network. When computing tasks are assigned to computing nodes within the management scope of the data network, the computing nodes work in turns to ensure energy conservation. However, when faced with sudden computing tasks, some computing nodes suspend their work and cannot quickly find suitable computing nodes, thereby reducing computing efficiency. Summary of the invention

[0003] In order to overcome the problems existing in the related art, the present disclosure provides a computing power management method, system and storage medium.

[0004] According to a first aspect of an embodiment of the present disclosure, a computing power management method is provided, the method comprising:

[0005] Obtain the distribution of all computing nodes within the management scope of the data network, and obtain the computing power resource data and work log of each computing node;

[0006] Determine the computing power coefficient and time adaptation coefficient of each computing node based on the computing power resource data and work log of each computing node;

[0007] When a burst computing task occurs within the management scope of the data network, obtaining the time point and task level of the burst computing task;

[0008] Based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node, a target computing node for computing the burst computing task is determined.

[0009] Optionally, in the embodiment of the present disclosure, determining the computing power coefficient and the time adaptation coefficient of each computing node based on the computing power resource data and the work log of each computing node includes:

[0010] Determine the computing power coefficient of each computing node based on the computing power resource data of each computing node;

[0011] Based on the work log of each computing node, a time adaptation coefficient of each computing node is determined.

[0012] Optionally, in the embodiment of the present disclosure, determining the computing power coefficient of each computing node based on the computing power resource data of each computing node includes:

[0013] Determining a CPU resource impact index of each computing node based on the CPU resource impact factor of each computing node;

[0014] Determining a memory resource impact index of each computing node based on the memory resource impact factor of each computing node;

[0015] Determine the GPU resource impact index of each computing node based on the GPU resource impact factor of each computing node;

[0016] Determining a storage device resource impact index of each computing node based on the storage device type of each computing node;

[0017] Based on the CPU resource impact index, the memory resource impact index, the GPU resource impact index and the storage device resource impact index of each computing node, the computing power coefficient of each computing node is determined.

[0018] Optionally, in the embodiment of the present disclosure, determining the time adaptation coefficient of each computing node based on the work log of each computing node includes:

[0019] Dividing the time period into multiple time intervals based on preset intervals;

[0020] Based on the work log of each of the computing nodes, fitting a first variation curve of the amount of computing tasks completed in each of the time intervals over time and a second variation curve of the average amount of computing tasks completed over time;

[0021] Based on the first change curve and the second change curve, a time adaptation coefficient of each time interval of each computing node is determined.

[0022] Optionally, in an embodiment of the present disclosure, determining a target computing node for computing the burst computing task based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node includes:

[0023] Determining a first computing node set for computing the burst computing task based on a time point of the burst computing task and a time adaptation coefficient of each of the computing nodes;

[0024] Based on the task level of the burst computing task and the computing power coefficient of each computing node in the first computing node set, the first computing node set is screened to determine a target computing node for computing the burst computing task.

[0025] Optionally, in the embodiment of the present disclosure, determining the first set of computing nodes for computing the burst computing task based on the time point of the burst computing task and the time adaptation coefficient of each of the computing nodes includes:

[0026] Determine the target time interval to which the time point at which the burst computing task occurs belongs;

[0027] Determining the most suitable time interval for each computing node based on the time adaptation coefficient of each computing node;

[0028] The computing node whose most suitable time interval is the target time interval is added to the first computing node set used to calculate the burst computing task.

[0029] Optionally, in an embodiment of the present disclosure, screening the first computing node set to determine a target computing node for computing the burst computing task based on the task level of the burst computing task and the computing power coefficient of each computing node in the first computing node set includes:

[0030] Determining a target computing power coefficient required for the burst computing task based on the task level of the burst computing task;

[0031] Adding computing nodes in the first computing node set that meet the target computing power coefficient to a second computing node set used to calculate the burst computing task;

[0032] The computing node with the smallest computing power coefficient in the second computing node set is determined as the target computing node for calculating the burst computing task.

[0033] According to a second aspect of an embodiment of the present disclosure, a computing power management system is provided, the system comprising:

[0034] A first acquisition unit is used to acquire the distribution of all computing nodes within the management scope of the data network, and acquire the computing resource data and work log of each computing node;

[0035] A first determining unit, configured to determine a computing power coefficient and a time adaptation coefficient of each computing node based on computing power resource data and a work log of each computing node;

[0036] A second acquisition unit is used to acquire the time point and task level of the burst computing task when a burst computing task occurs within the management scope of the data network;

[0037] The second determination unit is used to determine the target computing node for calculating the burst computing task based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node.

[0038] According to a third aspect of an embodiment of the present disclosure, there is provided an electronic device, including:

[0039] processor;

[0040] a memory for storing processor-executable instructions;

[0041] The processor is configured to implement the method described in the first aspect above.

[0042] According to a fourth aspect of an embodiment of the present disclosure, a non-temporary computer-readable storage medium is provided, wherein the computer storage medium stores computer-executable instructions; after the computer-executable instructions are executed by a processor, the method described in the first aspect above can be implemented.

[0043] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0044] The present disclosure proposes a computing power management method, system and storage medium, the method includes obtaining the distribution of all computing nodes within the management scope of the data network, and obtaining the computing power resource data and work log of each computing node; determining the computing power coefficient and time adaptation coefficient of each computing node based on the computing power resource data and work log of each computing node; when a burst computing task occurs within the management scope of the data network, obtaining the time point and task level of the burst computing task; based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node, determining the target computing node for calculating the burst computing task. Therefore, the present disclosure can quickly determine the target computing node through the computing power coefficient and time adaptation coefficient of each computing node when the computing nodes work in turn and a burst computing task occurs, so that the target computing node can be used to process the burst computing task, thereby improving computing efficiency.

[0045] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0047] Figure 1 is a flowchart of a computing power management method according to some embodiments of the present disclosure;

[0048] Figure 2 is a flowchart of a computing power management method according to some embodiments of the present disclosure;

[0049] Figure 3 is a schematic diagram of the structure of a computing power management system according to some embodiments of the present disclosure;

[0050] Figure 4It is a block diagram of an electronic device suitable for executing a computing power management method according to some embodiments of the present disclosure. DETAILED DESCRIPTION

[0051] Some embodiments of the present disclosure will be described in detail here, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. Various changes, modifications and equivalents of the methods, systems and / or systems described herein will become apparent after understanding the present disclosure. For example, the order of operations described herein is merely an example and is not limited to those orders set forth herein, but can be changed as becomes apparent after understanding the present disclosure, except for operations that must be performed in a specific order. In addition, for clarity and brevity, descriptions of features known in the art may be omitted.

[0052] The embodiments described in some embodiments of the present disclosure below do not represent all embodiments consistent with the present disclosure. Instead, they are only examples of systems and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0053] Figure 1 FIG. 1 is a flow chart of a computing power management method according to some embodiments of the present disclosure. As shown in the figure, the method may include the following steps:

[0054] Step 101: Obtain the distribution of all computing nodes within the management scope of the data network, and obtain the computing power resource data and work log of each computing node.

[0055] It should be noted that the executor of the above-mentioned computing power management method is a computing power management system, which can be implemented by software and / or hardware. The computing power management system in this embodiment can be configured in an electronic device.

[0056] In this example embodiment, the electronic device may include devices such as terminal devices and servers, and this embodiment does not limit the electronic device.

[0057] In the embodiment of the present disclosure, the distribution of all computing nodes within the management scope of the data network is obtained, and the m computing nodes within the management scope of the data network are numbered as: n1, n2, n3, ..., n m , in order to facilitate subsequent search. Among them, n i is the i-th computing node, i and m are both positive integers, i belongs to [1, m], m>1.

[0058] Furthermore, in the embodiments of the present disclosure, the computing power resource data of each computing node can be obtained through the configuration information of each computing node.

[0059] Step 102: Determine the computing power coefficient and time adaptation coefficient of each computing node based on the computing power resource data and work log of each computing node.

[0060] In the embodiment of the present disclosure, after obtaining the computing power resource data and work log of each computing node through the above steps, the computing power coefficient and time adaptation coefficient of each computing node can be determined based on the computing power resource data and work log of each computing node.

[0061] Specifically, in the embodiments of the present disclosure, the method for determining the computing power coefficient and the time adaptation coefficient of each computing node based on the computing power resource data and work log of each computing node may include: determining the computing power coefficient of each computing node based on the computing power resource data of each computing node, and determining the time adaptation coefficient of each computing node based on the work log of each computing node.

[0062] Among them, in the embodiment of the present disclosure, the method for determining the computing power coefficient of each computing node based on the computing power resource data of each computing node may include the following steps:

[0063] Step 1: Based on the CPU (Central Processing Unit) resource impact factor of each computing node, determine the CPU resource impact index of each computing node;

[0064] Step 2: Based on the memory resource impact factor of each computing node, determine the memory resource impact index of each computing node;

[0065] Step 3: Determine the GPU resource impact index of each computing node based on the GPU (Graphics Processing Unit) resource impact factor of each computing node;

[0066] Step 4: Based on the storage device type of each computing node, determine the storage device resource impact index of each computing node;

[0067] Step 5: Determine the computing power coefficient of each computing node based on the CPU resource impact index, memory resource impact index, GPU resource impact index, and storage device resource impact index of each computing node.

[0068] In the embodiment of the present disclosure, the above-mentioned CPU resource impact factor may include: the number of CPU cores, CPU clock frequency and CPU cache. And, in the embodiment of the present disclosure, the above-mentioned method for determining the CPU resource impact index of each computing node based on the CPU resource impact factor of each computing node may include: based on the number of CPU cores, CPU clock frequency and CPU cache of each computing node, determining the CPU resource impact index of each computing node by a first formula, wherein the first formula is:

[0069]

[0070] in, is the CPU resource impact index of the i-th computing node, is the number of CPU cores of the i-th computing node, is the CPU clock frequency of the ith computing node, is the CPU cache of the i-th computing node, A is the conversion coefficient, and They are the preset standard impact factor data respectively.

[0071] And, in the embodiment of the present disclosure, the above-mentioned memory resource impact factor may include memory capacity and memory bandwidth size. In the embodiment of the present disclosure, the above-mentioned method for determining the memory resource impact index of each computing node based on the memory resource impact factor of each computing node may include: based on the memory capacity and memory bandwidth size of each computing node, determining the memory resource impact index of each computing node by a second formula, wherein the second formula is:

[0072]

[0073] in, is the memory resource impact index of the i-th computing node, is the memory capacity of the i-th computing node, is the memory bandwidth size of the i-th computing node, B is the conversion coefficient, and They are the preset standard impact factor data respectively.

[0074] Further, in the embodiment of the present disclosure, the above-mentioned GPU resource impact factor may include the number of GPU stream processors, the size of GPU video memory, and the size of GPU video memory bandwidth. Among them, in the embodiment of the present disclosure, the above-mentioned method for determining the GPU resource impact index of each computing node based on the GPU resource impact factor of each computing node may include: based on the number of GPU stream processors, the size of GPU video memory, and the size of GPU video memory bandwidth of each computing node, determining the GPU resource impact index of each computing node by a third formula, wherein the third formula is:

[0075]

[0076] in, is the GPU resource impact index of the i-th computing node, is the number of GPU stream processors of the i-th computing node, is the GPU memory size of the i-th computing node, is the GPU memory bandwidth size of the i-th computing node, C is the conversion coefficient, and They are the preset standard impact factor data respectively.

[0077] Further, in the embodiment of the present disclosure, the method for determining the storage device resource impact index of each computing node based on the storage device type of each computing node may include: if the storage device type of the i-th computing node is a solid state drive, then the storage device resource impact index of the i-th computing node is If the storage device type of the i-th computing node is a mechanical keyboard, then the storage device resource impact index of the i-th computing node is Where D is the conversion factor.

[0078] And, in the embodiment of the present disclosure, the method for determining the computing power coefficient of each computing node based on the CPU resource impact index, memory resource impact index, GPU resource impact index and storage device resource impact index of each computing node may include: determining the computing power coefficient of each computing node by a fourth formula based on the CPU resource impact index, memory resource impact index, GPU resource impact index and storage device resource impact index of each computing node, wherein the fourth formula is:

[0079]

[0080] in, is the computing power coefficient of the i-th computing node, α1, α2, α3 and α4 are the CPU resource impact indicators Memory resource impact indicators GPU resource impact indicators and storage device resource impact indicators The weight coefficient of .

[0081] In the embodiment of the present disclosure, the above α1, α2, α3 and α4 are related to the type of task to be calculated. For example, in the embodiment of the present disclosure, if the type of task to be calculated is a computationally intensive task, α1>α3>α2=α4>0; if the type of task to be calculated is a complex graphics processing task, α3>α1>α2=α4>0.

[0082] Further, in the embodiment of the present disclosure, the method for determining the time adaptation coefficient of each computing node based on the work log of each computing node may include the following steps:

[0083] Step 1021, dividing the time period into multiple time intervals based on a preset interval;

[0084] Step 1022: based on the work log of each computing node, fit a first variation curve of the amount of computing tasks completed in each time interval over time and a second variation curve of the average amount of computing tasks completed over time;

[0085] Step 1023: Determine the adaptation value of each time interval of each computing node based on the first change curve and the second change curve;

[0086] Step 1024: sort the adaptation values ​​of each time interval of each computing node, and determine the time adaptation coefficient of each computing node based on the sorting result.

[0087] In the embodiment of the present disclosure, each day can be divided into q time intervals based on a preset interval. Among them, in the embodiment of the present disclosure, the above-mentioned preset interval can be set as needed. For example, the preset interval is 4 hours.

[0088] Furthermore, in the embodiment of the present disclosure, a first variation curve W of the amount of computing tasks completed in each time interval of each day over time can be fitted according to the work log of the i-th computing node. ni (t) and the second curve of the average amount of completed calculation tasks changing with time Among them, the ordinate in the first change curve is the amount of computing tasks completed in each time interval, and the horizontal axis is each time interval; the ordinate in the second change curve is the value of the amount of computing tasks completed in each time interval divided by the amount of computing tasks completed in all time intervals, and the horizontal axis is each time interval.

[0089] Further, in the embodiment of the present disclosure, the first change curve and the second change curve are obtained through the above steps, and the time adaptation coefficient of each time interval of each computing node can be determined by the fifth formula based on the first change curve and the second change curve, wherein the fifth formula is:

[0090]

[0091] in, is the time adaptation coefficient of the jth time interval of the ith computing node, Δt is the specific interval size, θ and μ are the preset weight coefficients, t j-1 is the starting point of the jth time interval.

[0092] And, in the embodiment of the present disclosure, after obtaining the adaptation value of each time interval of each computing node through the above steps, the adaptation value of each time interval of each computing node can be sorted, and the time adaptation coefficient of each computing node can be determined based on the sorting result. Among them, in the embodiment of the present disclosure, the method of sorting the adaptation value of each time interval of each computing node and determining the time adaptation coefficient of each computing node based on the sorting result can include: arranging the adaptation value of each time interval of each computing node in descending order, and determining the time interval ranked first as the time adaptation coefficient of each computing node.

[0093] Step 103: When a burst computing task occurs within the management scope of the data network, the time point and task level of the burst computing task are obtained.

[0094] In the embodiment of the present disclosure, the task level of the burst computing task can be determined based on the time required to complete the burst computing task and the task type to which the burst computing task belongs in combination with historical data.

[0095] Step 104: Based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node, determine the target computing node for computing the burst computing task.

[0096] In the disclosed embodiment, after the computing power coefficient and time adaptation coefficient of each computing node are determined through the above steps, the target computing node for calculating the burst computing task can be determined based on the time point and task level of the burst computing task.

[0097] Specifically, in the embodiment of the present disclosure, the method for determining the target computing node for computing the burst computing task based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node may include the following steps:

[0098] Step 1041: Determine a first computing node set for computing the burst computing task based on the time point of the burst computing task and the time adaptation coefficient of each computing node;

[0099] Step 1042: Based on the task level of the burst computing task and the computing power coefficient of each computing node in the first computing node set, the first computing node set is screened to determine a target computing node for computing the burst computing task.

[0100] In an embodiment of the present disclosure, a method for determining a first set of computing nodes for computing a burst computing task based on the time point of the burst computing task and the time adaptation coefficient of each computing node may include: determining a target time interval to which the time point at which the burst computing task occurs belongs, and adding computing nodes whose time adaptation coefficient is the target time interval to the first set of computing nodes for computing the burst computing task.

[0101] In the embodiment of the present disclosure, assuming that the burst computing task occurs at time t0, then in q time intervals A1, A2, ..., A q Determine the target time interval A to which the time point t0 belongs K , q>k>0. And, in the embodiment of the present disclosure, the time adaptation coefficient is set to the target time interval A K The computing nodes are added to the first computing node set for computing burst computing tasks, completing the preliminary screening of computing nodes.

[0102] Furthermore, in the embodiment of the present disclosure, after obtaining the first computing node set through the above steps, the first computing node set can be screened to determine the target computing node for computing the burst computing task based on the task level of the burst computing task and the computing power coefficient of each computing node in the first computing node set.

[0103] Specifically, in an embodiment of the present disclosure, based on the task level of the burst computing task and the computing power coefficient of each computing node in the first computing node set, the method for screening the first computing node set to determine the target computing node for computing the burst computing task may include the following steps:

[0104] Step 10421: Determine the target computing power coefficient required for the burst computing task based on the task level of the burst computing task;

[0105] Step 10422: Add computing nodes in the first computing node set that meet the target computing power coefficient to the second computing node set used to calculate the burst computing task;

[0106] Step 10423: Determine the computing node with the smallest computing power coefficient in the second computing node set as the target computing node for computing the burst computing task.

[0107] In the disclosed embodiment, the target computing power coefficient required for the burst computing task can be determined by the correspondence between the task level of the burst computing task and the computing power coefficient that is preset.

[0108] Furthermore, in an embodiment of the present disclosure, computing nodes in the first computing node set that meet the target computing power coefficient are added to the second computing node set used to calculate burst computing tasks, and the computing power coefficients of the computing nodes in the second computing node set are sorted, and the computing node with the smallest computing power coefficient is determined as the target computing node for calculating the burst computing task.

[0109] Furthermore, in the disclosed embodiment, the burst computing task is computed using the target computing node.

[0110] Among them, in the embodiments of the present disclosure, the above method can be applied to the scheduling of cloud computing. For example, social media platforms use cloud computing to process a large amount of user data and real-time requests. The user data and real-time requests need to be processed by computing nodes and then forwarded. The above method can help the media platform to achieve fast message push, online chat, dynamic sharing, etc.

[0111] The present disclosure proposes a computing power management method, which obtains the distribution of all computing nodes within the management scope of the data network, and obtains the computing power resource data and work log of each computing node; based on the computing power resource data and work log of each computing node, the computing power coefficient and time adaptation coefficient of each computing node are determined; when a burst computing task occurs within the management scope of the data network, the time point and task level of the burst computing task are obtained; based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node, the target computing node for calculating the burst computing task is determined. Therefore, the present disclosure can quickly determine the target computing node through the computing power coefficient and time adaptation coefficient of each computing node when the computing nodes work in turn and a burst computing task occurs, so that the target computing node can be used to process the burst computing task, thereby improving the computing efficiency.

[0112] Based on the above description, Figure 2 A flowchart of a computing power management method proposed in an embodiment of the present disclosure is provided. The method may include the following steps:

[0113] Step 201: Obtain the distribution of all computing nodes within the management scope of the data network;

[0114] Step 202: Obtain computing resource data of each computing node and daily work log of each computing node (the above-mentioned work log);

[0115] Step 203: Determine the computing power coefficient of each computing node based on the computing power resource data of each computing node;

[0116] Step 204: Obtain the time adaptation coefficient of each computing node based on the daily work log of each computing node;

[0117] Step 205: Obtain the time point when the burst computing task occurs;

[0118] Step 206: Obtain the task level of the burst computing task;

[0119] Step 207: Preliminarily screen the computing nodes used to calculate the outstanding computing tasks according to the time point when the burst computing tasks occur and the time adaptation coefficient of each computing node to obtain a first computing node set;

[0120] Step 208: finally screen the computing nodes in the first computing node set according to the task level of the burst computing task and the computing power coefficient of each computing node to obtain a target computing node.

[0121] In the embodiment of the present disclosure, for other contents of the above steps 201 to 208, reference may be made to the detailed description in the above embodiment, and the embodiment of the present disclosure will not be elaborated here.

[0122] Figure 3 A computing power management system is shown according to some embodiments of the present disclosure, and the system may include:

[0123] The first acquisition unit 301 is used to obtain the distribution of all computing nodes within the management scope of the data network, and obtain the computing resource data and work log of each computing node;

[0124] A first determining unit 302 is used to determine a computing power coefficient and a time adaptation coefficient of each computing node based on computing power resource data and work logs of each computing node;

[0125] The second acquisition unit 303 is used to acquire the time point and task level of the burst computing task when a burst computing task occurs within the management scope of the data network;

[0126] The second determining unit 304 is used to determine the target computing node for computing the burst computing task based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node.

[0127] In the embodiment of the present disclosure, the first determining unit 302 is specifically configured to:

[0128] Determine the computing power coefficient of each computing node based on the computing power resource data of each computing node;

[0129] Based on the work log of each computing node, the time adaptation coefficient of each computing node is determined.

[0130] Optionally, in the embodiment of the present disclosure, the first determining unit 302 is further configured to:

[0131] Based on the CPU resource impact factor of each computing node, determine the CPU resource impact index of each computing node;

[0132] Based on the memory resource impact factor of each computing node, determine the memory resource impact index of each computing node;

[0133] Based on the GPU resource impact factor of each computing node, determine the GPU resource impact index of each computing node;

[0134] Determine the storage device resource impact index of each computing node based on the storage device type of each computing node;

[0135] The computing power coefficient of each computing node is determined based on the CPU resource impact index, memory resource impact index, GPU resource impact index, and storage device resource impact index of each computing node.

[0136] Optionally, in the embodiment of the present disclosure, the above system is also used for:

[0137] Dividing the time period into multiple time intervals based on preset intervals;

[0138] Based on the work log of each computing node, fit the first change curve of the amount of computing tasks completed in each time interval over time and the second change curve of the average amount of computing tasks completed over time;

[0139] Determining an adaptation value of each time interval of each computing node based on the first change curve and the second change curve;

[0140] The adaptation values ​​of each time interval of each computing node are sorted, and the time adaptation coefficient of each computing node is determined based on the sorting result.

[0141] Optionally, in the embodiment of the present disclosure, the second determining unit 304 is specifically configured to:

[0142] Determining a first set of computing nodes for computing the burst computing task based on a time point of the burst computing task and a time adaptation coefficient of each computing node;

[0143] Based on the task level of the burst computing task and the computing power coefficient of each computing node in the first computing node set, the first computing node set is screened to determine a target computing node for computing the burst computing task.

[0144] Optionally, in the embodiment of the present disclosure, the second determining unit 304 is further configured to:

[0145] Determine the target time interval to which the time point of the burst computing task occurs;

[0146] The computing nodes whose time adaptation coefficients are the target time intervals are added to the first computing node set used for computing the burst computing tasks.

[0147] Optionally, in the embodiment of the present disclosure, the second determining unit 304 is further configured to:

[0148] Based on the task level of the burst computing task, determine the target computing power coefficient required for the burst computing task;

[0149] Adding computing nodes in the first computing node set that meet the target computing power coefficient to the second computing node set used to calculate the burst computing task;

[0150] The computing node with the smallest computing power coefficient in the second computing node set is determined as the target computing node for computing the burst computing task.

[0151] In one or more embodiments of the present disclosure, it includes obtaining the distribution of all computing nodes within the management scope of the data network, and obtaining the computing power resource data and work log of each computing node; determining the computing power coefficient and time adaptation coefficient of each computing node based on the computing power resource data and work log of each computing node; when a burst computing task occurs within the management scope of the data network, obtaining the time point and task level of the burst computing task; based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node, determining the target computing node for calculating the burst computing task. Thus, the present disclosure can quickly determine the target computing node through the computing power coefficient and time adaptation coefficient of each computing node when the computing nodes work in turns and a burst computing task occurs, so that the target computing node can be used to process the burst computing task, thereby improving computing efficiency.

[0152] like Figure 4 As shown, the electronic device 400 includes a computing unit 401, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 402 or a computer program loaded from a storage unit 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the electronic device 400 can also be stored. The computing unit 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0153] A number of components in the electronic device 400 are connected to the I / O interface 405, including: an input unit 408, such as a keyboard, a mouse, etc.; an output unit 407, such as various types of displays, speakers, etc.; a storage unit 408, such as a disk, an optical disk, etc.; and a communication unit 409, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 409 allows the electronic device 400 to exchange information / data with other electronic devices through a computer network such as the Internet and / or various telecommunication networks.

[0154] The computing unit 401 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 401 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 401 performs the various methods and processes described above, such as the computing power management method. For example, in some embodiments, the computing power management method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 408. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 400 via the ROM 402 and / or the communication unit 409. When the computer program is loaded into the RAM 403 and executed by the computing unit 401, one or more steps of the computing power management method described above may be performed. Alternatively, in other embodiments, the computing unit 401 may be configured to perform the computing power management method in any other appropriate manner (e.g., by means of firmware).

[0155] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations may include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input system, and at least one output system, and transmit data and instructions to the storage system, the at least one input system, and the at least one output system.

[0156] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing system, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine as a stand-alone software package and partially on a remote machine, or entirely on a remote machine or server.

[0157] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, system, or electronic device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or electronic device, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage electronic device, a magnetic storage electronic device, or any suitable combination of the foregoing.

[0158] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display system (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing system (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of systems can also be used to provide interaction with a user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0159] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.

[0160] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs running on the corresponding computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services ("Virtual Private Server", or "VPS" for short). The server may also be a server of a distributed system, or a server combined with a blockchain.

[0161] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0162] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. A computing power management method, characterized in that: The method comprises: Obtain the distribution of all computing nodes within the management scope of the data network, and obtain the computing power resource data and work log of each computing node; Determine the computing power coefficient and time adaptation coefficient of each computing node based on the computing power resource data and work log of each computing node; When a burst computing task occurs within the management scope of the data network, obtaining the time point and task level of the burst computing task; Based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node, a target computing node for computing the burst computing task is determined.

2. The method according to claim 1, characterized in that The determining of the computing power coefficient and the time adaptation coefficient of each computing node based on the computing power resource data and the work log of each computing node includes: Determine the computing power coefficient of each computing node based on the computing power resource data of each computing node; Based on the work log of each computing node, a time adaptation coefficient of each computing node is determined.

3. The method according to claim 2, characterized in that The determining, based on the computing power resource data of each computing node, a computing power coefficient of each computing node, comprises: Determining a CPU resource impact index of each computing node based on a central processing unit (CPU) resource impact factor of each computing node; Determining a memory resource impact index of each computing node based on the memory resource impact factor of each computing node; Determine the GPU resource impact index of each computing node based on the GPU resource impact factor of each computing node; Determining a storage device resource impact index of each computing node based on the storage device type of each computing node; Based on the CPU resource impact index, the memory resource impact index, the GPU resource impact index and the storage device resource impact index of each computing node, the computing power coefficient of each computing node is determined.

4. The method according to claim 2, characterized in that The determining the time adaptation coefficient of each computing node based on the work log of each computing node includes: Dividing the time period into multiple time intervals based on preset intervals; Based on the work log of each of the computing nodes, fitting a first variation curve of the amount of computing tasks completed in each of the time intervals over time and a second variation curve of the average amount of computing tasks completed over time; Determining an adaptation value of each time interval of each computing node based on the first change curve and the second change curve; The adaptation values ​​of each time interval of each computing node are sorted, and the time adaptation coefficient of each computing node is determined based on the sorting result.

5. The method according to claim 1, characterized in that The determining of a target computing node for computing the burst computing task based on the time point and task level of the burst computing task and the computing power coefficient and time adaptation coefficient of each computing node includes: Determining a first computing node set for computing the burst computing task based on a time point of the burst computing task and a time adaptation coefficient of each of the computing nodes; Based on the task level of the burst computing task and the computing power coefficient of each computing node in the first computing node set, the first computing node set is screened to determine a target computing node for computing the burst computing task.

6. The method according to claim 5, characterized in that The determining, based on the time point of the burst computing task and the time adaptation coefficient of each of the computing nodes, a first computing node set for computing the burst computing task comprises: Determine the target time interval to which the time point at which the burst computing task occurs belongs; The computing nodes whose time adaptation coefficient is a target time interval are added to a first computing node set used to calculate the burst computing task.

7. The method according to claim 5, characterized in that The step of screening the first computing node set to determine a target computing node for computing the burst computing task based on the task level of the burst computing task and the computing power coefficient of each computing node in the first computing node set includes: Determining a target computing power coefficient required for the burst computing task based on the task level of the burst computing task; Adding computing nodes in the first computing node set that meet the target computing power coefficient to a second computing node set used to calculate the burst computing task; The computing node with the smallest computing power coefficient in the second computing node set is determined as the target computing node for calculating the burst computing task.

8. A computing power management system, characterized in that: The system comprises: A first acquisition unit is used to acquire the distribution of all computing nodes within the management scope of the data network, and acquire the computing resource data and work log of each computing node; A first determining unit, configured to determine a computing power coefficient and a time adaptation coefficient of each computing node based on computing power resource data and a work log of each computing node; A second acquisition unit is used to acquire the time point and task level of the burst computing task when a burst computing task occurs within the management scope of the data network; The second determination unit is used to determine the target computing node for calculating the burst computing task based on the time point and task level of the burst computing task, and the computing power coefficient and time adaptation coefficient of each computing node.

9. An electronic device, characterized in that: include: at least one processor; as well as a memory in communication with the at least one processor, wherein: The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, wherein the computer storage medium stores computer-executable instructions; after the computer-executable instructions are executed by a processor, the method according to any one of claims 1 to 7 can be implemented.

Images