Scheduling system for power load side resources

By designing load aggregation, task scheduling and load regulation modules in the supercomputing center, analyzing the grid load curve and allocating execution time according to the task priority, the grid load imbalance problem of supercomputing center in the grid load scheduling is solved, and the power grid is realized Adaptation and stable operation of load.

CN119051015AActive Publication Date: 2024-11-29SHENZHEN POWER SUPPLY BUREAU

Patent Information

Application Number
CN202411294980.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-14
Publication Date
2024-11-29
Estimated Expiration
2044-09-14

AI Technical Summary

Technical Problem

Supercomputing centers have problems with grid load imbalance in load scheduling management, especially during peak electricity consumption periods, which may lead to power failures. The existing technology lacks effective methods to adjust the execution time of the calculation task based on project priorities and grid load to avoid Impact of peak load period.

Method used

A scheduling system for power load-side resources is designed, including load aggregation module, task scheduling module and load regulation module. By analyzing historical operation data, a load curve is generated, the grid load phase is divided, and task execution is allocated according to task priority and load curve. Time, adjust the operating power of the supercomputer center to match the grid load.

Benefits of technology

It effectively avoids the impact of peak load on the power grid, ensures the stable operation of the power grid, and dynamically adjusts the execution time and power of the calculation task, adapts to the power grid load, and improves the load scheduling and management capabilities of the supercomputing center.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119051015A_ABST
    Figure CN119051015A_ABST
Patent Text Reader

Abstract

The invention provides a scheduling system for power load side resources, and the system comprises a load aggregation module which generates a first load curve according to the historical operation data of a supercomputing center, carries out the stage division of the first load curve to obtain a second load curve, and carries out the operation of the second load curve to obtain a prediction allocation curve; the task scheduling module allocates execution time of the to-be-calculated task according to the priority of the to-be-calculated task and a second load curve to obtain a first scheduling scheme, and sends the first scheduling scheme to a load adjusting module; and the load adjusting module calculates the real-time total power of the super-computing center according to the first scheduling scheme, adjusts the operation power corresponding to different modules of the super-computing center according to the real-time total power and the predicted allocation curve to obtain a second scheduling scheme, and controls the super-computing center to operate. The problem that in the prior art, a scheduling method for adjusting project execution time based on project priorities, power grid loads and computing resource requirements to avoid influences of load peak periods on computing tasks is lacked is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power load dispatching, and in particular to a dispatching system for power load-side resources. Background Art

[0002] In today's era of rapid digital development, supercomputing centers, as major electricity consumers on the load side of the power grid, play a vital role. However, supercomputing centers face complex power management challenges. Typically, supercomputing centers have a large number of computing projects that need to be run, and these projects have significant differences in priority, resource requirements, and execution time. For example, some urgent scientific research computing projects may have the highest priority and need to be given priority to resources; while some routine data processing projects have relatively low priority. At the same time, the resource requirements of different projects at runtime are also different. Some projects may require a lot of computing power, while others are more dependent on storage resources.

[0003] Due to these differences, the load scheduling management of the supercomputing center is particularly critical. If effective load scheduling cannot be carried out, it may cause an imbalance in the load of the power grid and affect the stable operation of the entire power grid. For example, during peak hours of electricity consumption, if the load of the supercomputing center is too high, it may increase the burden on the power grid and even cause power failures. Therefore, how to achieve effective scheduling of power loads and adapt them to the load of the power grid has become an important issue that needs to be solved urgently. Summary of the invention

[0004] The main purpose of this application is to provide a scheduling system for power load-side resources to at least solve the problem in the prior art of lacking a scheduling method for adjusting project execution time based on project priority, grid load, and computing resource requirements to avoid load peaks affecting computing tasks.

[0005] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a scheduling system for power load-side resources is provided, the scheduling system comprising a load aggregation module, a task scheduling module and a load regulation module, the load aggregation module, the task scheduling module and the load regulation module are communicatively connected with each other, wherein: the load aggregation module is used to generate a first load curve according to the historical operation data of the supercomputing center and divide the first load curve into stages according to the load value and the preset threshold to obtain a second load curve, calculate according to the second load curve to obtain a prediction and dispatching curve, send the second load curve to the task scheduling module and send the prediction and dispatching curve to the load regulation module, the first load curve is a curve of the power grid load changing over time; the task scheduling module is used to allocate the execution time of the task to be calculated according to the priority of the task to be calculated and the second load curve, obtain a first scheduling scheme, and send the first scheduling scheme to the load regulation module; the load regulation module is used to calculate the real-time total power of the supercomputing center according to the first scheduling scheme, and adjust the operating power corresponding to different modules of the supercomputing center according to the real-time total power and the prediction and dispatching curve to obtain a second scheduling scheme, and control the operation of the supercomputing center according to the second scheduling scheme.

[0006] Optionally, the load aggregation module includes a load curve generation module, a load curve division module and a dispatch curve generation module, wherein: the load curve generation module is used to make predictions based on the historical operation data of the power grid, generate a curve of the power grid load changing with time within a first preset time period, obtain the first load curve, and send the first load curve to the load curve division module, the length corresponding to the first preset time period is a preset duration, and the starting time of the first preset time period is the current time; the load curve division module is used to traverse the first load curve, determine the curve in the first load curve corresponding to the power grid load less than or equal to the first threshold as the trough stage, determine the curve in the first load curve corresponding to the power grid load greater than the first threshold and less than the second threshold as the balance stage, determine the curve in the first load curve corresponding to the power grid load greater than or equal to the second threshold as the peak stage, obtain the second load curve, and send the second load curve to the dispatch curve generation module and the task scheduling module; the dispatch curve generation module is used to fit the first preset coefficient, the second preset coefficient and the second load curve to obtain a predicted dispatch curve, and send the predicted dispatch curve to the load regulation module.

[0007] Optionally, the task scheduling module includes a priority determination module and a task matching module, wherein: the priority determination module is used to traverse the required computing speed and safety score of the task to be calculated to divide the task to be calculated into a first priority, a second priority and a third priority, the first priority being higher than the third priority, and the third priority being higher than the second priority; the task matching module is used to assign the task to be calculated of the first priority to the trough stage for execution, assign the task to be calculated of the second priority to the peak stage for execution, and assign the task to be calculated of the third priority to the balance stage for execution, so as to obtain the first scheduling scheme.

[0008] Optionally, the load regulation module includes a power consumption calculation module, a safety power consumption adjustment module and a computing power power consumption adjustment module, wherein: the power consumption calculation module is used to determine the safety power consumption of the safety components in the supercomputing center, the computing power consumption of the computing components and other power consumption of other power-consuming components according to the first scheduling scheme, calculate the heat dissipation power consumption of the heat dissipation components of the supercomputing center according to the safety power consumption, the computing power consumption and the other power consumption, and calculate the real-time total power according to the safety power consumption, the computing power consumption, the heat dissipation power consumption and the other power consumption; the safety power consumption adjustment module is used to adjust the safety power consumption according to the real-time total power and the predicted allocation curve when the task to be calculated is of the first priority; the computing power power consumption adjustment module is used to adjust the computing power consumption according to the real-time total power and the predicted allocation curve.

[0009] Optionally, a prediction dispatch curve is obtained by fitting the first preset coefficient, the second preset coefficient and the second load curve, including: calculating the product of the second load curve and the first preset coefficient to obtain a third load curve; and summing the third load curve and the second preset coefficient to obtain the prediction dispatch curve.

[0010] Optionally, the required computing speeds and security scores of the tasks to be calculated are traversed to divide the tasks to be calculated into a first priority, a second priority and a third priority, including: determining the tasks to be calculated whose corresponding computing speed is greater than a third threshold and whose required security score is greater than a fourth threshold as the first priority; determining the tasks to be calculated whose corresponding computing speed is less than a fifth threshold and whose required security score is less than a sixth threshold as the second priority, the third threshold is greater than the fifth threshold, and the fourth threshold is greater than the sixth threshold; determining the tasks to be calculated other than the first priority and the second priority as the third priority.

[0011] Optionally, the heat dissipation power consumption of the heat dissipation component of the supercomputing center is calculated according to the safety power consumption, the computing power power consumption and the other power consumptions, and the real-time total power is calculated according to the safety power consumption, the computing power power consumption, the heat dissipation power consumption and the other power consumptions, including: calculating the sum of the safety power consumption, the computing power power consumption and the other power consumptions to obtain the component power consumption; calculating the product of the component power consumption and a third preset coefficient to obtain the heat dissipation power consumption; summing the safety power consumption, the computing power power consumption, the heat dissipation power consumption and the other power consumption to obtain the real-time total power.

[0012] Optionally, when the task to be calculated is of the first priority level, the safety power consumption is adjusted according to the real-time total power and the predicted scheduling curve, including: determining the current scheduling value according to the predicted scheduling curve; and when the current scheduling value is greater than the real-time total power, increasing the safety power consumption.

[0013] Optionally, when the task to be calculated is of the first priority level, the safe power consumption is adjusted according to the real-time total power and the predicted scheduling curve, including: determining the current scheduling value according to the predicted scheduling curve; when the current scheduling value is greater than the real-time total power, increasing the computing power consumption.

[0014] According to another aspect of the present application, a power grid is provided, comprising any one of the above-mentioned systems.

[0015] By applying the technical solution of the present application, a scheduling system for resources on the power load side is provided, the scheduling system comprising a load aggregation module, a task scheduling module and a load regulation module, wherein the load aggregation module, the task scheduling module and the load regulation module are communicatively connected with each other, wherein: the load aggregation module is used to generate a first load curve according to the historical operation data of the supercomputing center and divide the first load curve into stages according to the load value and the preset threshold value to obtain a second load curve, calculate according to the second load curve to obtain a prediction and dispatching curve, send the second load curve to the task scheduling module and send the prediction and dispatching curve to the load regulation module, the first load curve is a curve of the power grid load changing over time; the task scheduling module is used to allocate the execution time of the task to be calculated according to the priority of the task to be calculated and the second load curve, obtain a first scheduling scheme, and send the first scheduling scheme to the load regulation module; the load regulation module is used to calculate the real-time total power of the supercomputing center according to the first scheduling scheme, and adjust the operating power corresponding to different modules of the supercomputing center according to the real-time total power and the prediction and dispatching curve to obtain a second scheduling scheme, and control the operation of the supercomputing center according to the second scheduling scheme. The present application divides time according to the grid load of the power grid, and classifies computing tasks based on project priority, and matches the classified computing tasks with the divided time periods to achieve project calculations adapted to the grid load, thereby solving the problem in the prior art of lacking a scheduling method for adjusting the project execution time based on project priority, grid load, and computing resource requirements to avoid load peaks affecting computing tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A structural block diagram of a power load side resource dispatching system provided according to an embodiment of the present application is shown;

[0017] Figure 2 A structural block diagram of a load aggregation module provided according to an embodiment of the present application is shown;

[0018] Figure 3 A power grid load period division diagram of a load curve division module provided according to an embodiment of the present application is shown;

[0019] Figure 4 A structural block diagram of a task scheduling module provided according to an embodiment of the present application is shown;

[0020] Figure 5 A structural block diagram of a load regulation module provided according to an embodiment of the present application is shown;

[0021] Figure 6 It shows that the scheduling curve generation module provided in the embodiment of the present application generates a predicted scheduling curve. DETAILED DESCRIPTION

[0022] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0023] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0024] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0025] As introduced in the background technology, in the prior art, there are usually a large number of computing projects that need to be run, and these projects may have different priorities, resource requirements, execution time and other characteristics. Therefore, effective load scheduling is needed to manage these projects to adapt to the grid load. In order to solve the problem that the prior art lacks a scheduling method that adjusts the project execution time based on project priority, grid load, and computing resource requirements to avoid the impact of load peak periods on computing tasks, an embodiment of the present application provides a scheduling system for power load-side resources.

[0026] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0027] Figure 1 is a structural block diagram of a dispatching system for power load side resources according to an embodiment of the present application. Figure 2 As shown, the scheduling system includes: a load aggregation module, a task scheduling module and a load regulation module, and the load aggregation module, the task scheduling module and the load regulation module are communicatively connected, wherein:

[0028] The load aggregation module is used to generate a first load curve according to the historical operation data of the supercomputing center and divide the first load curve into stages according to the load value and the preset threshold to obtain a second load curve, calculate according to the second load curve to obtain a prediction and dispatching curve, send the second load curve to the task scheduling module and send the prediction and dispatching curve to the load regulation module, and the first load curve is a curve of the power grid load changing over time;

[0029] Specifically, the load aggregation module obtains power data of the grid load changing over time within a preset time period from the supercomputing center, and plots it into a curve of the grid load changing over time, i.e., the first load curve. A preset threshold is set, and the first load curve is divided into different grid load stages according to the power values ​​of the grid load at different times in the first load curve to obtain the second load curve, and send it to the task scheduling module. The negative value of the second load curve is taken, and the allocation coefficient is added to perform calculations to obtain the allocation curves of different grid load stages, and send them to the load regulation module.

[0030] The task scheduling module is used to allocate the execution time of the task to be calculated according to the priority of the task to be calculated and the second load curve, obtain a first scheduling scheme, and send the first scheduling scheme to the load adjustment module;

[0031] Specifically, in the task scheduling module, execution priorities are assigned to tasks according to the types of different tasks to be calculated, and execution scheduling plans for different tasks to be calculated are comprehensively assigned according to the operation priorities of different tasks to be calculated and the power grid load periods corresponding to the second load curve.

[0032] The load regulation module is used to calculate the real-time total power of the supercomputing center according to the first scheduling plan, and adjust the operating power corresponding to different modules of the supercomputing center according to the real-time total power and the predicted allocation curve to obtain a second scheduling plan, and control the operation of the supercomputing center according to the second scheduling plan.

[0033] Specifically, the real-time total power and the predicted allocation curve are compared in the load regulation module, and are adjusted in real time. When the scheduling value is greater than the total power before allocation, the load regulation module sends a computing power increase instruction to the supercomputing center.

[0034] Applying the technical solution of the present application, in the above-mentioned scheduling system for power load-side resources, the above-mentioned scheduling system includes a load aggregation module, a task scheduling module and a load regulation module, and the above-mentioned load aggregation module, the above-mentioned task scheduling module and the above-mentioned load regulation module are communicatively connected, wherein: the above-mentioned load aggregation module is used to generate a first load curve according to the historical operation data of the supercomputing center and divide the above-mentioned first load curve into stages according to the load value and the preset threshold to obtain a second load curve, calculate according to the above-mentioned second load curve to obtain a predicted dispatching curve, send the above-mentioned second load curve to the above-mentioned task scheduling module and send the above-mentioned predicted dispatching curve to the above-mentioned load regulation module, the above-mentioned first load curve is a curve of the power grid load changing over time; the above-mentioned task scheduling module is used to allocate the execution time of the above-mentioned task to be calculated according to the priority of the task to be calculated and the above-mentioned second load curve, obtain a first scheduling plan, and send the above-mentioned first scheduling plan to the above-mentioned load regulation module; the above-mentioned load regulation module is used to calculate the real-time total power of the above-mentioned supercomputing center according to the above-mentioned first scheduling plan, and adjust the operating power corresponding to different modules of the supercomputing center according to the above-mentioned real-time total power and the above-mentioned predicted dispatching curve to obtain a second scheduling plan, and control the operation of the above-mentioned supercomputing center according to the above-mentioned second scheduling plan. The present application divides time according to the grid load of the power grid, and classifies computing tasks based on project priority, and matches the classified computing tasks with the divided time periods to achieve project calculations adapted to the grid load, thereby solving the problem in the prior art of lacking a scheduling method for adjusting the project execution time based on project priority, grid load, and computing resource requirements to avoid load peaks affecting computing tasks.

[0035] In one embodiment of the present application, Figure 3 As shown, the load aggregation module includes a load curve generation module, a load curve division module and a dispatch curve generation module, wherein:

[0036] The load curve generation module is used to make predictions based on the historical operation data of the power grid, generate a curve of the load of the power grid changing with time within a first preset time period, obtain the first load curve, and send the first load curve to the load curve division module. The duration corresponding to the first preset time period is the preset duration, and the starting time of the first preset time period is the current time.

[0037] Specifically, the load curve generation module first obtains the historical operation data of the grid load changing with time within a preset time period, generates a load curve according to the data of the grid load changing with time within the first preset time period, i.e., the first load curve, and then sends the generated first load curve to the load curve division module.

[0038] The load curve division module is used to traverse the first load curve, determine the curve in the first load curve corresponding to the power grid load less than or equal to the first threshold as the valley stage, determine the curve in the first load curve corresponding to the power grid load greater than the first threshold and less than the second threshold as the balance stage, determine the curve in the first load curve corresponding to the power grid load greater than or equal to the second threshold as the peak stage, obtain the second load curve, and send the second load curve to the dispatch curve generation module and the task scheduling module;

[0039] Specifically, Figure 4 As shown, the load curve division module divides the power grid load period into a valley stage less than or equal to a first threshold, a balance stage greater than the first threshold and less than the second threshold, and a peak stage greater than or equal to the second threshold according to the power grid load power threshold at different times within a preset time period, and obtains the second load curve, the first threshold is 800, the second threshold is 1500, and then sends the second load curve to the scheduling curve generation module and the task scheduling module.

[0040] The dispatch curve generation module is used to obtain the predicted dispatch curve by fitting the first preset coefficient, the second preset coefficient and the second load curve, and send the predicted dispatch curve to the load regulation module.

[0041] Specifically, the above-mentioned dispatch curve generation module first takes the negative value of the above-mentioned second load curve, that is, the first preset coefficient, and then adds the deployment coefficient, that is, the second preset coefficient, to obtain the above-mentioned predicted dispatch curve.

[0042] In one embodiment of the present application, Figure 5 As shown, the above task scheduling module includes a priority determination module and a task matching module, wherein:

[0043] The priority determination module is used to traverse the required computing speed and safety score of the tasks to be calculated to divide the tasks to be calculated into a first priority, a second priority and a third priority, wherein the first priority is higher than the third priority, and the third priority is higher than the second priority;

[0044] Specifically, the above-mentioned computing tasks can be divided into security tasks, general tasks and restriction tasks. Among them, security tasks need to be processed first and require high security and fast computing speed. They are classified as the first priority. General tasks have a medium processing priority and the computing speed can be appropriately reduced. They are classified as the third priority. Restriction tasks have a low processing priority and the calculation can be terminated when necessary. They are classified as the second priority.

[0045] The task matching module is used to assign the tasks to be calculated of the first priority to the trough phase for execution, assign the tasks to be calculated of the second priority to the peak phase for execution, and assign the tasks to be calculated of the third priority to the balance phase for execution, so as to obtain the first scheduling scheme.

[0046] Specifically, the above-mentioned different task types are assigned to different power grid load stages according to their priorities, among which, the security tasks are suitable for execution in the low-power consumption stage, that is, the above-mentioned tasks to be calculated with the first priority are assigned to the above-mentioned low-power consumption stage, and the ordinary tasks are suitable for execution in the balancing time period, that is, the above-mentioned tasks to be calculated with the third priority are assigned to the above-mentioned balancing stage, and the restriction tasks are suitable for execution in the low-power consumption stage, that is, the above-mentioned tasks to be calculated with the second priority are assigned to the above-mentioned low-power consumption stage.

[0047] In one embodiment of the present application, Figure 6 As shown, the load adjustment module includes a power consumption calculation module, a security power consumption adjustment module and a computing power power consumption adjustment module, wherein:

[0048] The power consumption calculation module is used to determine the safety power consumption of the safety components in the supercomputing center, the computing power consumption of the computing components and other power consumption of other power-consuming components according to the first scheduling scheme, calculate the heat dissipation power consumption of the heat dissipation components of the supercomputing center according to the safety power consumption, the computing power consumption and the other power consumption, and calculate the real-time total power according to the safety power consumption, the computing power consumption, the heat dissipation power consumption and the other power consumption;

[0049] Specifically, the power consumption calculation module calculates the real-time total power using the formula: total power = safety power consumption + computing power consumption + other power consumption + heat dissipation power consumption, heat dissipation power consumption = (safety power consumption + computing power consumption + other power consumption) * heat dissipation coefficient, where the heat dissipation coefficient is 0.1.

[0050] The safety power consumption adjustment module is used to adjust the safety power consumption according to the real-time total power and the predicted allocation curve when the task to be calculated is of the first priority;

[0051] Specifically, the above-mentioned safety power consumption adjustment module is used to perform real-time regulation on the safety power consumption of security tasks. When the power consumption during the low-power phase is greater than the total power before adjustment, the above-mentioned safety power consumption adjustment module sends instructions to the above-mentioned supercomputing center to increase the above-mentioned safety power consumption.

[0052] The computing power consumption adjustment module is used to adjust the computing power consumption according to the real-time total power and the predicted allocation curve.

[0053] Specifically, the computing power power consumption adjustment module is used to perform real-time regulation of computing power power consumption. When the power consumption of the predicted allocation curve is greater than the total power before adjustment, the computing power power consumption adjustment module sends instructions to the supercomputing center to increase computing power.

[0054] In one embodiment of the present application, the prediction dispatch curve is obtained by fitting the first preset coefficient, the second preset coefficient and the second load curve, including:

[0055] Calculating the product of the second load curve and the first preset coefficient to obtain a third load curve;

[0056] Specifically, the first preset coefficient is a negative value, and the second load curve is multiplied by the first preset coefficient to obtain the third load curve with a negative power value.

[0057] The third load curve and the second preset coefficient are summed to obtain the forecast dispatch curve.

[0058] Specifically, the second preset coefficient is a positive constant, and the third load curve and the second preset coefficient are summed to obtain the predicted dispatch curve with a positive power value.

[0059] The above computing tasks can be divided into security tasks, general tasks and restriction tasks. Among them, security tasks need to be processed first and require high security and fast computing speed. They are classified as the first priority. General tasks have a medium processing priority and the computing speed can be appropriately reduced. They are classified as the third priority. Restriction tasks have a low processing priority and the calculation can be terminated when necessary. They are classified as the second priority.

[0060] In one embodiment of the present application, the required computing speed and safety score of the above-mentioned tasks to be calculated are traversed to divide the above-mentioned tasks to be calculated into a first priority, a second priority and a third priority, including:

[0061] Determine the above-mentioned to-be-calculated tasks whose corresponding calculation speed is greater than the third threshold and whose required safety score is greater than the fourth threshold as the above-mentioned first priority;

[0062] Specifically, the above-mentioned tasks to be calculated that require fast calculation speed and high security are determined as the above-mentioned first priority.

[0063] The to-be-calculated tasks corresponding to the calculation speed being less than the fifth threshold and the required safety score being less than the sixth threshold are determined as the second priority, the third threshold is greater than the fifth threshold, and the fourth threshold is greater than the sixth threshold;

[0064] Specifically, the calculation speed can be significantly reduced, the calculation can be terminated when necessary, and the above-mentioned tasks to be calculated with low security requirements are determined as the above-mentioned second priority.

[0065] The tasks to be calculated other than the first priority and the second priority are determined as the third priority.

[0066] Specifically, the above-mentioned to-be-calculated tasks that require a medium computing speed and have a medium security and can be reduced in computing speed are determined as the above-mentioned third priority.

[0067] In one embodiment of the present application, the heat dissipation power consumption of the heat dissipation component of the supercomputing center is calculated according to the safety power consumption, the computing power consumption and the other power consumption, and the real-time total power is calculated according to the safety power consumption, the computing power consumption, the heat dissipation power consumption and the other power consumption, including:

[0068] Calculate the sum of the above security power consumption, the above computing power consumption and the above other power consumption to obtain the component power consumption;

[0069] Specifically, the above-mentioned component power consumption is the power consumption generated when the system components of the scheduling system are running, and component power consumption = security power consumption + computing power consumption + other power consumption.

[0070] Calculate the product of the power consumption of the component and the third preset coefficient to obtain the heat dissipation power consumption;

[0071] Specifically, the third preset coefficient is 0.1, and heat dissipation power consumption = (safety power consumption + computing power consumption + other power consumption) * 0.1.

[0072] The above-mentioned security power consumption, the above-mentioned computing power consumption, the above-mentioned heat dissipation power consumption and the above-mentioned other power consumption are summed to obtain the above-mentioned real-time total power.

[0073] Specifically, real-time total power = security power consumption + computing power consumption + other power consumption + heat dissipation power consumption.

[0074] In an embodiment of the present application, when the task to be calculated is of the first priority, adjusting the safe power consumption according to the real-time total power and the predicted allocation curve includes:

[0075] Determine the current dispatch value according to the above-mentioned forecast dispatch curve;

[0076] Specifically, in the load aggregation module, a prediction dispatch curve is obtained by calculation according to the second load curve, and the prediction dispatch curve is sent to the load regulation module, and the current dispatch value is determined in the load regulation module according to the prediction dispatch curve.

[0077] When the current scheduling value is greater than the real-time total power, the safe power consumption is increased.

[0078] Specifically, the safety power consumption of the above-mentioned tasks to be calculated with the above-mentioned first priority is regulated, and when the above-mentioned current scheduling value is greater than the above-mentioned real-time total power, an instruction to increase the power consumption of the safety system is sent to the supercomputing center.

[0079] In an embodiment of the present application, when the task to be calculated is of the first priority, adjusting the safe power consumption according to the real-time total power and the predicted allocation curve includes:

[0080] Determine the current dispatch value according to the predicted dispatch curve;

[0081] Specifically, in the load aggregation module, a prediction dispatch curve is obtained by calculation according to the second load curve, and the prediction dispatch curve is sent to the load regulation module, and the current dispatch value is determined in the load regulation module according to the prediction dispatch curve.

[0082] When the current scheduling value is greater than the real-time total power, the computing power consumption is increased.

[0083] Specifically, the computing power consumption of the above-mentioned tasks to be calculated with the above-mentioned first priority is regulated, and when the above-mentioned current scheduling value is greater than the above-mentioned real-time total power, an instruction to increase the computing power is sent to the supercomputing center.

[0084] The present application provides a power grid, comprising a system that executes any one of the above-mentioned scheduling systems for power load-side resources.

[0085] Specifically, a dispatching system for power load-side resources includes: a load aggregation module, a task scheduling module and a load regulation module, wherein the load aggregation module, the task scheduling module and the load regulation module are communicatively connected, wherein:

[0086] The load aggregation module is used to generate a first load curve according to the historical operation data of the supercomputing center and divide the first load curve into stages according to the load value and the preset threshold to obtain a second load curve, calculate according to the second load curve to obtain a prediction and dispatching curve, send the second load curve to the task scheduling module and send the prediction and dispatching curve to the load regulation module, and the first load curve is a curve of the power grid load changing over time;

[0087] Specifically, the load aggregation module obtains power data of the grid load changing over time within a preset time period from the supercomputing center, and plots it into a curve of the grid load changing over time, i.e., the first load curve. A preset threshold is set, and the first load curve is divided into different grid load stages according to the power values ​​of the grid load at different times in the first load curve to obtain the second load curve, and send it to the task scheduling module. The negative value of the second load curve is taken, and the allocation coefficient is added to perform calculations to obtain the allocation curves of different grid load stages, and send them to the load regulation module.

[0088] The task scheduling module is used to allocate the execution time of the task to be calculated according to the priority of the task to be calculated and the second load curve, obtain a first scheduling scheme, and send the first scheduling scheme to the load adjustment module;

[0089] Specifically, in the task scheduling module, execution priorities are assigned to tasks according to the types of different tasks to be calculated, and execution scheduling plans for different tasks to be calculated are comprehensively assigned according to the operation priorities of different tasks to be calculated and the power grid load periods corresponding to the second load curve.

[0090] The load regulation module is used to calculate the real-time total power of the supercomputing center according to the first scheduling plan, and adjust the operating power corresponding to different modules of the supercomputing center according to the real-time total power and the predicted allocation curve to obtain a second scheduling plan, and control the operation of the supercomputing center according to the second scheduling plan.

[0091] Specifically, the real-time total power and the predicted allocation curve are compared in the load regulation module, and are adjusted in real time. When the scheduling value is greater than the total power before allocation, the load regulation module sends a computing power increase instruction to the supercomputing center.

[0092] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device 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, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0093] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0094] 1) In a dispatching system for power load-side resources of the present application, the dispatching system includes a load aggregation module, a task scheduling module and a load regulation module, and the load aggregation module, the task scheduling module and the load regulation module are communicatively connected with each other, wherein: the load aggregation module is used to generate a first load curve according to the historical operation data of the supercomputing center and divide the first load curve into stages according to the load value and the preset threshold to obtain a second load curve, calculate according to the second load curve to obtain a predicted dispatching curve, send the second load curve to the task scheduling module and send the predicted dispatching curve to the load regulation module, the first load curve is a curve of the power grid load changing over time; the task scheduling module is used to allocate the execution time of the task to be calculated according to the priority of the task to be calculated and the second load curve, obtain a first dispatching scheme, and send the first dispatching scheme to the load regulation module; the load regulation module is used to calculate the real-time total power of the supercomputing center according to the first dispatching scheme, and adjust the operating power corresponding to different modules of the supercomputing center according to the real-time total power and the predicted dispatching curve to obtain a second dispatching scheme, and control the operation of the supercomputing center according to the second dispatching scheme. The present application divides time according to the grid load of the power grid, and classifies computing tasks based on project priority, and matches the classified computing tasks with the divided time periods to achieve project calculations adapted to the grid load, thereby solving the problem in the prior art of lacking a scheduling method for adjusting the project execution time based on project priority, grid load, and computing resource requirements to avoid load peaks affecting computing tasks.

[0095] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A dispatching system for power load side resources, characterized in that: The scheduling system includes a load aggregation module, a task scheduling module and a load regulation module, wherein the load aggregation module, the task scheduling module and the load regulation module are communicatively connected to each other, wherein: The load aggregation module is used to generate a first load curve according to the historical operation data of the supercomputing center, divide the first load curve into stages according to the load value and the preset threshold value to obtain a second load curve, calculate according to the second load curve to obtain a prediction and dispatching curve, send the second load curve to the task scheduling module and send the prediction and dispatching curve to the load regulation module, and the first load curve is a curve of the grid load changing over time; The task scheduling module is used to allocate the execution time of the task to be calculated according to the priority of the task to be calculated and the second load curve, obtain a first scheduling scheme, and send the first scheduling scheme to the load adjustment module; The load regulation module is used to calculate the real-time total power of the supercomputing center according to the first scheduling plan, and adjust the operating power corresponding to different modules of the supercomputing center according to the real-time total power and the predicted allocation curve to obtain a second scheduling plan, and control the operation of the supercomputing center according to the second scheduling plan.

2. The system according to claim 1, characterized in that The load aggregation module includes a load curve generation module, a load curve division module and a dispatch curve generation module, wherein: The load curve generation module is used to make predictions based on the historical operation data of the power grid, generate a curve of the load of the power grid changing with time within a first preset time period, obtain the first load curve, and send the first load curve to the load curve division module, the length corresponding to the first preset time period is a preset time length, and the starting time of the first preset time period is the current time; The load curve division module is used to traverse the first load curve, determine the curve in the first load curve corresponding to the power grid load less than or equal to the first threshold as the valley stage, determine the curve in the first load curve corresponding to the power grid load greater than the first threshold and less than the second threshold as the balance stage, determine the curve in the first load curve corresponding to the power grid load greater than or equal to the second threshold as the peak stage, obtain the second load curve, and send the second load curve to the dispatch curve generation module and the task scheduling module; The dispatch curve generation module is used to obtain a predicted dispatch curve by fitting according to the first preset coefficient, the second preset coefficient and the second load curve, and send the predicted dispatch curve to the load adjustment module.

3. The system according to claim 2, characterized in that The task scheduling module includes a priority determination module and a task matching module, wherein: The priority determination module is used to traverse the required computing speed and safety score of the task to be calculated to divide the task to be calculated into a first priority, a second priority and a third priority, wherein the first priority is higher than the third priority, and the third priority is higher than the second priority; The task matching module is used to assign the tasks to be calculated of the first priority to the trough stage for execution, assign the tasks to be calculated of the second priority to the peak stage for execution, and assign the tasks to be calculated of the third priority to the balance stage for execution, so as to obtain the first scheduling plan.

4. The system according to claim 3, characterized in that The load adjustment module includes a power consumption calculation module, a safety power consumption adjustment module and a computing power power consumption adjustment module, wherein: The power consumption calculation module is used to determine the safety power consumption of the safety components in the supercomputing center, the computing power consumption of the computing components and other power consumption of other power-consuming components according to the first scheduling scheme, calculate the heat dissipation power consumption of the heat dissipation components of the supercomputing center according to the safety power consumption, the computing power consumption and the other power consumption, and calculate the real-time total power according to the safety power consumption, the computing power consumption, the heat dissipation power consumption and the other power consumption; The safety power consumption adjustment module is used for adjusting the safety power consumption according to the real-time total power and the predicted allocation curve when the task to be calculated is of the first priority; The computing power consumption adjustment module is used to adjust the computing power consumption according to the real-time total power and the predicted allocation curve.

5. The system according to claim 2, characterized in that The prediction dispatch curve is obtained by fitting the first preset coefficient, the second preset coefficient and the second load curve, including: Calculating the product of the second load curve and the first preset coefficient to obtain a third load curve; The third load curve and the second preset coefficient are summed to obtain the predicted dispatch curve.

6. The system according to claim 3, characterized in that Traversing the required computing speed and safety score of the task to be calculated to divide the task to be calculated into a first priority, a second priority and a third priority, including: Determine the to-be-calculated task whose corresponding calculation speed is greater than a third threshold and whose required safety score is greater than a fourth threshold as the first priority; Determine the to-be-calculated task corresponding to the calculation speed being less than the fifth threshold and the required safety score being less than the sixth threshold as the second priority, the third threshold being greater than the fifth threshold, and the fourth threshold being greater than the sixth threshold; The to-be-calculated tasks other than the first priority and the second priority are determined as a third priority.

7. The system according to claim 4, characterized in that Calculating the heat dissipation power consumption of the heat dissipation component of the supercomputing center according to the safety power consumption, the computing power consumption, and the other power consumptions, and calculating the real-time total power according to the safety power consumption, the computing power consumption, the heat dissipation power consumption, and the other power consumptions, including: Calculating the sum of the security power consumption, the computing power consumption and the other power consumption to obtain component power consumption; Calculating the product of the power consumption of the component and a third preset coefficient to obtain the heat dissipation power consumption; The security power consumption, the computing power consumption, the heat dissipation power consumption and the other power consumption are summed to obtain the real-time total power.

8. The system according to claim 4, characterized in that When the task to be calculated is of the first priority, adjusting the safe power consumption according to the real-time total power and the predicted allocation curve includes: Determine the current dispatch value according to the predicted dispatch curve; When the current scheduling value is greater than the real-time total power, the safe power consumption is increased.

9. The system according to claim 4, characterized in that When the task to be calculated is of the first priority, adjusting the safe power consumption according to the real-time total power and the predicted allocation curve includes: Determine the current dispatch value according to the predicted dispatch curve; When the current scheduling value is greater than the real-time total power, the computing power consumption is increased.

10. A power grid, characterized in that: A system comprising any one of claims 1 to 7.

Citation Information

Patent Citations

  • Method and device for determining scheduling scheme of computing task

    CN108733464A

  • Comprehensive scheduling method and device for multiple scheduling units of source network load storage, and electronic equipment

    CN115549203A

  • Method and device for fitting power consumer 24-point load into 96-point curve, and storage medium

    CN117633431A

Cited By

  • Calculation and power collaboration system and method

    CN120691373A