Liquid cooling energy consumption optimization method and system for dynamic scheduling of computing power resources

Abstract

The invention discloses a computing power resource dynamic scheduling liquid cooling energy consumption optimization method and system. The method comprises the steps of obtaining target task request information, and allocating a first task priority according to the target task request information; the target task request information at least comprises a task type, required computing power and emergency degree; task scheduling is carried out according to a first scheduling strategy, a first task priority is adjusted, a second task priority is obtained, and the first scheduling strategy is obtained according to task waiting time and available computing power resources; calculating a computing power resource utilization rate after scheduling of the first scheduling strategy, and calculating an optimization effect in combination with the heat exchange efficiency; the heat exchange efficiency is the heat exchange efficiency of the liquid cooling system configured by the target task execution device. By optimizing computing power resource dynamic scheduling and optimizing resource scheduling and cooling strategies at the same time, it is ensured that the computing requirements of low-priority tasks are not ignored while high-priority tasks are processed in time, and the overall resource utilization rate is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of resource energy consumption optimization, and particularly to a method and system for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources. Background Art

[0002] The dynamic scheduling of computing power resources based on task priority means that the system dynamically allocates computing resources according to the priority of tasks. For example, high-priority tasks will be preferentially allocated more computing resources to ensure the timely completion of tasks. Dynamic scheduling means that during the execution of tasks, the system will adjust the resource allocation in real time according to factors such as the usage of resources and the progress of tasks. Liquid cooling is a cooling technology that absorbs the heat generated during the operation of computer hardware through a liquid (such as water or a special coolant). Compared with traditional air cooling technology, liquid cooling can provide more efficient heat dissipation and is suitable for high-density computing tasks. Liquid cooling technology is used to effectively control the temperature of servers or computing devices, avoid overheating, and ensure the stability and performance of hardware operation. Energy consumption optimization refers to reducing energy consumption through reasonable resource scheduling and liquid cooling technology. The energy efficiency of computer hardware and liquid cooling systems directly affects the overall power consumption. Through task priority scheduling, the rationality of computing resource allocation, and the efficiency of liquid cooling technology, excessive energy waste can be reduced, and the energy efficiency of the entire data center or computing platform can be optimized.

[0003] When dynamically scheduling computing power resources, although more resources can be preferentially allocated to high-priority tasks, this priority scheduling may lead to insufficient computing resources for low-priority tasks, thus affecting the completion time of overall tasks. Uneven allocation of computing resources may result in resource idling or waste, leading to low efficiency of the overall computing platform. The goals of liquid cooling systems and energy efficiency optimization are often intertwined. The operation of liquid cooling systems depends on energy, while the goal of energy efficiency optimization is to reduce the overall energy consumption, which requires fine regulation. Simply optimizing the liquid cooling system may lead to increased energy consumption, while simply optimizing energy efficiency may affect the heat dissipation effect. Summary of the Invention

[0004] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the specification of this application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions shall not be used to limit the scope of the present invention.

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the present invention provides a method and system for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources, which can solve the problems mentioned in the background art.

[0007] To solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, the present invention provides a method for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources, including:

[0009] Obtain target task request information, and allocate a first task priority according to the target task request information;

[0010] The target task request information at least includes the task type, the required computing power, and the urgency;

[0011] Perform task scheduling according to a first scheduling policy and adjust the first task priority to obtain a second task priority, where the first scheduling policy is obtained according to the task waiting time and the available computing power resources;

[0012] Calculate the utilization rate of the computing power resources after scheduling by the first scheduling policy, and calculate the optimization effect in combination with the heat exchange efficiency;

[0013] The heat exchange efficiency is the heat exchange efficiency of the liquid cooling system configured for the device executing the target task.

[0014] As a preferred solution of the method for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources according to the present invention, wherein: the allocating the first task priority according to the target task request information includes:

[0015] Allocate a priority value to each task according to the task type, the required computing power, and the urgency;

[0016] Sort the target tasks according to the priority values;

[0017] Add the sorted tasks to the task queue in order of priority and wait for scheduling;

[0018] The task queue at least includes a high-priority task queue, a medium-priority task queue, and a low-priority task queue.

[0019] As a preferred solution of the method for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources according to the present invention, wherein: the first scheduling policy includes:

[0020] Obtain the time when the task enters the task queue, and subtract the time when the task enters the task queue from the current time to obtain the waiting time of the task to be processed;

[0021] Obtain the total computing power resources of the system and the currently used computing power resources, and subtract the currently used computing power resources from the total computing power resources of the system to obtain the available computing power resources;

[0022] Obtain the first scheduling policy according to the task waiting time and the available computing power resources.

[0023] As a preferred solution of the liquid cooling energy consumption optimization method for dynamic scheduling of computing power resources according to the present invention, wherein: the task scheduling according to the first scheduling strategy and adjusting the first task priority includes:

[0024] When there is a task to be scheduled, the task is preferentially taken out from the high-priority task queue;

[0025] For the high-priority task queue, if the available computing power resources meet the computing power requirements of the task, the task is allocated to the corresponding computing node for execution;

[0026] If the available computing power resources do not meet the computing power requirements of the task, wait until there are sufficient resources available.

[0027] As a preferred solution of the liquid cooling energy consumption optimization method for dynamic scheduling of computing power resources according to the present invention, wherein: the task scheduling according to the first scheduling strategy and adjusting the first task priority further includes:

[0028] For the tasks in the medium-priority task queue and the low-priority task queue, if the available computing power resources meet the computing power requirements of the task and the waiting time of the task in the queue is greater than the expected task waiting time, the priority of the task is adjusted;

[0029] Schedule according to the adjusted priority. If the adjusted task priorities are equal, preferentially process the task with the higher initial priority.

[0030] As a preferred solution of the liquid cooling energy consumption optimization method for dynamic scheduling of computing power resources according to the present invention, wherein: calculating the utilization rate of computing power resources after scheduling by the first scheduling strategy includes:

[0031] Obtain the completion time and start time of the task, and subtract the start time from the completion time of the task to obtain the processing time of the task;

[0032] Obtain the computing power resource usage of the task according to the real-time monitoring of the usage of computing power resources;

[0033] Calculate the utilization rate of computing power resources according to the total computing power resources of the system and the total cycle time, combined with the processing time of the task and the computing power resource usage.

[0034] As a preferred solution of the liquid cooling energy consumption optimization method for dynamic scheduling of computing power resources according to the present invention, wherein: the second task priority is used as the first task priority for the next round of optimization.

[0035] In a second aspect, the present invention provides a liquid cooling energy consumption optimization system for dynamic scheduling of computing power resources, including:

[0036] A data acquisition and processing module, configured to acquire target task request information and allocate a first task priority according to the target task request information;

[0037] The target task request information at least includes a task type, required computing power, and urgency;

[0038] A scheduling module, configured to perform task scheduling according to a first scheduling policy and adjust the first task priority to obtain a second task priority, where the first scheduling policy is obtained according to the task waiting time and available computing power resources;

[0039] An optimization effect judgment module, configured to calculate the utilization rate of computing power resources after scheduling by the first scheduling policy and calculate the optimization effect in combination with the heat exchange efficiency;

[0040] The heat exchange efficiency is the heat exchange efficiency of the coolant cooling system configured for the device executing the target task.

[0041] In a third aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the steps of the method described above are implemented.

[0042] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method described above are implemented.

[0043] Compared with the prior art, the beneficial effects of the present invention: The present invention proposes a method and system for optimizing liquid cooling energy consumption in dynamic scheduling of computing power resources, acquires target task request information, and allocates a first task priority according to the target task request information; the target task request information at least includes a task type, required computing power, and urgency; performs task scheduling according to a first scheduling policy and adjusts the first task priority to obtain a second task priority, where the first scheduling policy is obtained according to the task waiting time and available computing power resources; calculates the utilization rate of computing power resources after scheduling by the first scheduling policy and calculates the optimization effect in combination with the heat exchange efficiency; the heat exchange efficiency is the heat exchange efficiency of the coolant cooling system configured for the device executing the target task. By optimizing the dynamic scheduling of computing power resources and simultaneously optimizing the resource scheduling and cooling strategy, it is ensured that while high-priority tasks are processed in a timely manner, the computing requirements of low-priority tasks are not ignored, and the overall resource utilization rate is improved. Description of the Drawings

[0044] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. Among them:

[0045] Figure 1 It is a flowchart of a method for optimizing the liquid cooling energy consumption in the dynamic scheduling of computing power resources provided by an embodiment of the present invention;

[0046] Figure 2 It is an internal structure diagram of a computer device for a method for optimizing the liquid cooling energy consumption in the dynamic scheduling of computing power resources provided by an embodiment of the present invention. Detailed implementation manners

[0047] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will make a detailed description of the specific implementation manners of the present invention with reference to the drawings of the specification. Obviously, the described embodiments are some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0048] Embodiment 1

[0049] Referring to Figure 1 - Figure 2 , which is the first embodiment of the present invention. This embodiment provides a method and system for optimizing the liquid cooling energy consumption in the dynamic scheduling of computing power resources, including:

[0050] In the existing related technologies, there are some problems. For example, although high-priority tasks can obtain computing resources first, low-priority tasks may wait for a long time due to insufficient resources, resulting in low overall task completion efficiency. In addition, the unreasonable allocation of computing resources may also lead to resource idleness or waste, increasing energy consumption. Moreover, the operation of the liquid cooling system also consumes energy. How to optimize energy consumption while optimizing liquid cooling heat dissipation is an urgent problem to be solved.

[0051] This application provides a method that can effectively solve the above-mentioned problems. Next, multiple embodiments will be combined to elaborate in detail how to implement the method for optimizing the liquid cooling energy consumption in the dynamic scheduling of computing power resources;

[0052] Figure 1 A flowchart of a method for optimizing the liquid cooling energy consumption in the dynamic scheduling of computing power resources is shown, including:

[0053] S101, obtaining target task request information and assigning a first task priority according to the target task request information;

[0054] In the embodiments of the present application, the target task request information at least includes the task type, the required computing power, and the urgency level;

[0055] In an alternative embodiment, allocating the first task priority according to the target task request information may specifically include: parsing the target task request information to identify the specific requirements of the task, such as key information like the task type, the required computing power, and the urgency level. These information are the basis for subsequent task priority allocation and task scheduling. The system will, according to the preset priority allocation rules, comprehensively consider the task type, the required computing power, and the urgency level, and assign a reasonable priority value to each task. This value reflects the urgency and importance of the task and is an important basis for task scheduling.

[0056] In an alternative embodiment, after allocating the first task priority, the system will sort these tasks according to the priority values to ensure that high-priority tasks can be processed first. The sorted tasks will be sequentially added to the task queue for waiting to be scheduled. The task queue is an ordered set of tasks, which is sorted according to the priority of the tasks to ensure that the system can process tasks in the order of priority.

[0057] In an alternative embodiment, in the task queue, in order to further refine the management and scheduling of tasks, the system can also divide the task queue into a high-priority task queue, a medium-priority task queue, and a low-priority task queue. In this way, the system can perform more targeted scheduling processing according to tasks of different priorities, improving the efficiency and accuracy of task processing.

[0058] Meanwhile, the system will also monitor the status of the task queue in real time, including information such as the number of tasks, the priority, and the waiting time. These information are crucial for subsequent task scheduling and priority adjustment. By monitoring the status of the task queue in real time, the system can timely discover and solve problems in task scheduling to ensure that tasks can be processed in a timely manner according to the priority order.

[0059] In the embodiments of the present application, allocating the first task priority according to the target task request information includes:

[0060] Assigning a priority value to each task according to the task type, the required computing power, and the urgency level;

[0061] Sorting the target tasks according to the priority values;

[0062] Sequentially adding the sorted tasks to the task queue according to the priority for waiting to be scheduled;

[0063] The task queue at least includes a high-priority task queue, a medium-priority task queue, and a low-priority task queue.

[0064] Exemplarily, a priority value is assigned to each task based on the type of task, required computing power, and urgency.

[0065] For example, in a video rendering farm, high-resolution, urgent video rendering tasks are set to high priority, while ordinary video rendering tasks are set to low priority.

[0066] The task queues include high-priority task queues, medium-priority task queues, and low-priority task queues. When a task arrives at the system, it is placed in the corresponding queue according to its priority.

[0067] For example, set the value range of high priority tasks to 8-10, medium priority tasks to 4-7, and low priority tasks to 1-3.

[0068] The sorted tasks are added to the task queue in order of priority. The task queue is a buffer area that temporarily stores pending tasks. The task queue is organized in the form of a priority queue.

[0069] It should be noted that obtaining the target task request information and assigning the first task priority according to the target task request information can ensure that the system can process multiple task requests in an orderly and efficient manner. By assigning a reasonable priority to each task, the system can give priority to urgent and important tasks, avoiding tasks being shelved for a long time due to waiting for resources, thereby improving the overall task processing efficiency. At the same time, the setting of the task queue also makes the management and scheduling of tasks clearer and more orderly, which helps to reasonably allocate and utilize system resources. This priority allocation and task queue management method provides a solid foundation for the subsequent dynamic scheduling of computing resources and optimization of liquid cooling energy consumption.

[0070] S102, performing task scheduling and adjusting the first task priority according to a first scheduling strategy to obtain a second task priority, wherein the first scheduling strategy is obtained according to the task waiting time and available computing resources;

[0071] In an optional embodiment, the first scheduling strategy is used to dynamically adjust according to the task waiting time and available computing resources to ensure that the tasks can be processed efficiently and orderly. Specifically, the first scheduling strategy will comprehensively consider the waiting time of the task in the queue and the available computing resources of the current system. When the task waiting time is long and the system computing resources are sufficient, the system will give priority to scheduling these tasks to reduce the waiting time of the task and improve the efficiency of task processing. When the system computing resources are tight, the system will give priority to scheduling high-priority tasks to ensure that critical tasks can be processed in a timely manner.

[0072] In an alternative embodiment, during the scheduling process, the system also dynamically adjusts the priorities of tasks based on the execution status of tasks and the real-time changes in computing power resources. For example, when a low-priority task has been waiting in the queue for too long and the system's computing power resources gradually become sufficient, the system will increase the priority of this task so that it can be processed more quickly. This way of dynamically adjusting priorities helps the system better balance the processing requirements of tasks with different priorities and improve the overall efficiency and satisfaction of task processing.

[0073] In an alternative embodiment, by implementing the first scheduling policy, the system can achieve dynamic scheduling and optimization of computing power resources, ensuring that high-priority tasks are processed in a timely manner while not neglecting the computing requirements of low-priority tasks. This scheduling policy helps improve the resource utilization rate of the system, reduce resource idleness and waste, and thus lower the energy consumption of the system.

[0074] In the embodiment of the present application, the first scheduling policy includes:

[0075] Obtain the time when the task enters the task queue, and subtract the time when the task enters the task queue from the current time to obtain the waiting time of the task to be processed;

[0076] Obtain the total computing power resources of the system and the currently used computing power resources, and subtract the currently used computing power resources from the total computing power resources of the system to obtain the available computing power resources;

[0077] Obtain the first scheduling policy based on the task waiting time and the available computing power resources.

[0078] In the embodiment of the present application, performing task scheduling according to the first scheduling policy and adjusting the first task priority includes:

[0079] When there is a task to be scheduled, preferentially take out tasks from the high-priority task queue;

[0080] For the high-priority task queue, if the available computing power resources meet the computing power requirements of the task, allocate the task to the corresponding computing node for execution;

[0081] If the available computing power resources do not meet the computing power requirements of the task, wait until there are sufficient resources available.

[0082] In the embodiment of the present application, performing task scheduling according to the first scheduling policy and adjusting the first task priority further includes:

[0083] For tasks in the medium-priority task queue and the low-priority task queue, if the available computing power resources meet the computing power requirements of the task and the waiting time of the task in the queue is greater than the expected task waiting time, adjust the priority of the task;

[0084] Schedule according to the adjusted priority. If the adjusted task priorities are equal, give priority to the task with the higher initial priority.

[0085] Exemplarily, obtain the time when the task enters the queue, and subtract the time when the task enters the queue from the current time to obtain the waiting time of the task to be processed.

[0086] In an alternative embodiment, the process of obtaining the available computing power resources by monitoring the usage of computing power resources in real time is as follows:

[0087] Obtain the total computing power resources of the system and the currently used computing power resources, and subtract the currently used computing power resources from the total computing power resources of the system to obtain the available computing power resources.

[0088] In an alternative embodiment, the process of task scheduling and adjusting task priorities according to the waiting time of the task and the available computing power resources is as follows:

[0089] When a task needs to be scheduled, give priority to retrieving tasks from the high-priority task queue.

[0090] If the available computing power resources meet the computing power requirements of the task, allocate the task to the corresponding computing node for execution.

[0091] If the available computing power resources do not meet the computing power requirements of the task, wait until sufficient resources are available.

[0092] In an alternative embodiment, for tasks in the medium-priority task queue and the low-priority task queue,

[0093] If the available computing power resources meet the computing power requirements of the task and the waiting time of the task in the queue is greater than the expected waiting time of the task, adjust the priority of the task.

[0094] Schedule according to the adjusted priority. If the adjusted task priorities are equal, give priority to the task with the higher initial priority.

[0095] In an alternative embodiment, the calculation formula for adjusting the task priority is as follows: P t = P0 + α * t w ; where P t is the adjusted priority, P0 is the initial priority, α is the priority adjustment coefficient, and t w is the waiting time of the task in the queue.

[0096] In the embodiments of the present application, the second task priority is used as the first task priority for the next round of optimization.

[0097] It should be noted that task scheduling is performed according to the first scheduling policy and the first task priority is adjusted to obtain the second task priority. The first scheduling policy can more flexibly respond to different task requirements by obtaining the task waiting time and available computing power resources, ensuring the rationality and efficiency of task scheduling. By comprehensively considering the task waiting time and available computing power resources, the system can dynamically adjust the processing order of tasks, avoid resource idleness or waste, and ensure that high-priority tasks are processed in a timely manner. This scheduling policy helps to improve the overall task processing efficiency, reduce the task waiting time, and thus enhance the user experience and system performance. In addition, by dynamically adjusting the task priority, the system can better balance the processing requirements of different priority tasks and achieve reasonable allocation and utilization of resources.

[0098] S103. Calculate the utilization rate of computing power resources after scheduling by the first scheduling policy, and calculate the optimization effect in combination with the heat exchange efficiency;

[0099] In the embodiment of the present application, the heat exchange efficiency is the heat exchange efficiency of the coolant cooling system configured for the device executing the target task.

[0100] In the embodiment of the present application, calculating the utilization rate of computing power resources after scheduling by the first scheduling policy includes:

[0101] Obtain the completion time and start time of the task, and subtract the start time of the task from the completion time to obtain the processing time of the task;

[0102] Obtain the amount of computing power resources used by the task according to the real-time monitoring of the usage of computing power resources;

[0103] Calculate the utilization rate of computing power resources based on the total amount of computing power resources and total cycle time of the system, in combination with the processing time and amount of computing power resources used by the task.

[0104] In an optional embodiment, obtain the completion time and start time of the task, and subtract the start time of the task from the completion time to obtain the processing time of the task;

[0105] Obtain the amount of computing power resources used by the task according to the real-time monitoring of the usage of computing power resources;

[0106] Obtain the total amount of computing power resources and total cycle time of the system, and calculate the utilization rate of the system's computing power resources in combination with the processing time and amount of computing power resources used by the task. The specific calculation formula is as follows:

[0107]

[0108] In the formula, U is the utilization rate of the system's computing power resources, R z is the amount of computing power resources used, t z is the total cycle time, t i is the processing time of the task, Ri is the computing power resource usage of the task;

[0109] It should be noted that there are many benefits of the computing power resource utilization rate of the computing system, which directly affects the system performance, efficiency, and cost. The following are the main benefits of the computing power resource utilization rate of the computing system:

[0110] 1. Improve resource utilization efficiency

[0111] Optimize resource allocation: By monitoring and calculating the resource utilization rate, it can be ensured that computing resources (such as CPU, GPU, memory, storage, etc.) are fully utilized instead of being wasted. This can effectively avoid the idleness of computing resources, thereby improving the overall system utilization efficiency.

[0112] Load balancing: Reasonably allocate the load to avoid overloading some computing resources while other resources are idle. In this way, the resource usage of the system will be more balanced.

[0113] 2. Reduce costs

[0114] Cost savings: If computing tasks can be run while efficiently utilizing resources, unnecessary hardware expansion can be avoided. For example, in a cloud computing environment, by dynamically adjusting resource allocation, enterprises can flexibly use computing resources according to demand, avoid purchasing excessive computing power, and reduce waste.

[0115] Energy savings: Improving resource utilization rate can also reduce energy consumption. Especially in large-scale data centers, reducing the energy consumption of the system is an important cost-saving measure.

[0116] 3. Improve system performance

[0117] Reduce response time: When efficiently utilizing system resources, user requests or computing tasks can be responded to more quickly. This is especially important for applications that require real-time processing (such as online services, AI inference, etc.).

[0118] Improve task completion speed: The system can process more tasks or more complex calculations simultaneously, enhancing the overall computing power.

[0119] 4. Extend the hardware lifespan

[0120] Reduce resource overload and waste: By avoiding long-term resource idleness or overloaded operation, the risk of hardware failure or damage can be reduced, thereby extending the service life of computing resources (such as servers, storage, etc.).

[0121] 5. Help predict future resource requirements

[0122] Dynamic adjustment: By calculating the changing trend of resource utilization rate, it is possible to better predict future resource requirements, make advance plans for resource expansion, and avoid performance bottlenecks or resource shortages during peak periods.

[0123] Optimized large-scale expansion: Dynamically adjust resources according to the changes in utilization rate, reasonably plan the expansion of hardware, and avoid over-purchasing or resource surplus.

[0124] 6. Improve business decision-making

[0125] Provide data support: Accurate resource utilization rate indicators can provide decision-making support for management, helping them make more reasonable business decisions. For example, selecting the most suitable resource type, deciding whether to upgrade hardware or adjust the business architecture, etc.

[0126] 7. Enhance system scalability

[0127] Expand on demand: By understanding the utilization of various resources, the system can be expanded more flexibly. When demand increases, computing resources can be quickly expanded, and when demand decreases, resources can be appropriately reduced, thus achieving elastic expansion.

[0128] 8. Improve service quality

[0129] Higher availability: By dynamically adjusting system resources, it is possible to ensure the stable operation of the service under different loads, avoid system crashes or service interruptions caused by resource shortages, and thus improve the overall availability of the service.

[0130] 9. Support automated management

[0131] Intelligent scheduling: Through real-time monitoring of resource utilization rate, automated resource scheduling and load balancing can be achieved, improving the intelligent level of system management and reducing the need for manual intervention.

[0132] It should be noted that thermal conductivity refers to the ability of a material to conduct heat. When calculating the heat exchange efficiency, it is necessary to first determine the thermal conductivity value of the cooling medium used.

[0133] In an optional embodiment, increasing the thermal conductivity helps reduce energy waste caused by insufficient heat dissipation or low efficiency. When the thermal conductivity is not increased, due to the limited heat transfer ability of the coolant, the liquid cooling system may need to operate at a high power for a long time, resulting in energy waste. After increasing the thermal conductivity, the liquid cooling system can more accurately match the heat dissipation requirements, reduce unnecessary power consumption, lower the energy waste index, and thus reduce energy consumption.

[0134] Furthermore, measure the thermal conductivity of the cooling medium used in the liquid cooling technology through experimental measurement, referring to material manuals, or according to the Maxwell-Garnett formula for the hybrid coolant.

[0135] Furthermore, the state information in the heat transfer process includes the heat exchange area, temperature difference, and thermal resistance.

[0136] Furthermore, the heat exchange area refers to the effective contact area for heat exchange between the cooling medium and the heat-generating component.

[0137] Furthermore, design the heat exchange area according to the heat dissipation requirements and space limitations of the system.

[0138] Furthermore, the temperature difference refers to the difference between the temperature of the heat-generating component and the temperature of the cooling medium, which is the driving force for heat transfer. Obtain the temperatures of the system's heat-generating component and the cooling medium respectively, and then calculate the difference between the two to get the temperature difference.

[0139] Furthermore, thermal resistance is a measure that impedes heat transfer and is related to factors such as the heat exchange path and material properties; the value of thermal resistance can be obtained through theoretical calculation or experimental testing.

[0140] Furthermore, the thermal conductivity of the cooling medium, combined with the heat exchange area, temperature difference, and thermal resistance obtained during the heat transfer process, is used to calculate the heat exchange efficiency of the liquid cooling system using the heat exchange efficiency calculation formula. The specific calculation formula for the heat exchange efficiency is as follows:

[0141]

[0142] In the formula, J is the heat exchange efficiency, D is the thermal conductivity of the cooling medium, A is the heat exchange area, ΔT is the temperature difference, and r is the thermal resistance.

[0143] In an alternative embodiment, the optimization effect of the computing power resource dynamic scheduling liquid cooling energy consumption system can be judged based on fuzzy inference according to the system computing power resource utilization rate and the heat exchange efficiency of the liquid cooling system.

[0144] In an alternative embodiment, calculating the computing power resource utilization rate after the first scheduling strategy is executed and combining it with the heat exchange efficiency to calculate the optimization effect can also be achieved by constructing an optimization effect evaluation model. Specifically, this evaluation model can comprehensively analyze and predict the computing power resource utilization rate and the heat exchange efficiency based on historical data and real-time data of the current system. By comparing the data changes before and after implementing the first scheduling strategy, the system optimization effect can be quantitatively evaluated, including aspects such as the improvement of the computing power resource utilization rate, the reduction of energy consumption, and the improvement of the overall system performance. In addition, the evaluation model can also dynamically adjust the optimization strategy according to real-time data to ensure that the system always remains in the optimal state.

[0145] This application selects the computing power resource utilization rate after scheduling by the first scheduling strategy based on fuzzy inference, and calculates the optimization effect in combination with the heat exchange efficiency. The reason is that the fuzzy inference method can handle uncertainty and ambiguity problems, which is particularly important in complex system optimization. In practical applications, the computing power resource utilization rate and heat exchange efficiency of the system are often affected by various factors, and these factors may be difficult to accurately quantify or predict. The fuzzy inference method can make reasonable decisions and judgments under incomplete or uncertain information by introducing fuzzy sets and fuzzy rules.

[0146] In the fuzzy inference process, first, key indicators such as the computing power resource utilization rate and heat exchange efficiency of the system need to be fuzzified and mapped into the corresponding fuzzy sets. Then, based on the experience and knowledge of experts, a fuzzy rule base is constructed, and these rules describe the correlation and influence degree between different indicators. When the system is running, data such as the computing power resource utilization rate and heat exchange efficiency are collected in real time, and after fuzzification, they are input into the fuzzy inference system. The fuzzy inference system performs inference calculations according to the preset rule base and outputs the evaluation result of the optimization effect.

[0147] Compared with traditional optimization effect evaluation methods, the method based on fuzzy inference has the following advantages:

[0148] 1. Strong adaptability: The fuzzy inference method can handle various uncertainty and ambiguity problems and is applicable to complex and changeable system environments.

[0149] 2. High flexibility: By adjusting the fuzzy rules and fuzzy sets, it can easily adapt to different optimization goals and constraint conditions.

[0150] 3. Good interpretability: The fuzzy inference process is easy to understand and explain, which helps managers better understand the system optimization effect and improvement direction.

[0151] In summary, this application selects the computing power resource utilization rate after scheduling by the first scheduling strategy based on fuzzy inference and calculates the optimization effect in combination with the heat exchange efficiency, which is a reasonable and effective method. It can accurately evaluate and improve the optimization effect of the liquid cooling energy consumption system for dynamic scheduling of computing power resources in a complex and changeable system environment.

[0152] In summary, the present invention proposes a method for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources, obtaining target task request information, and assigning a first task priority according to the target task request information; the target task request information includes at least task type, required computing power, and urgency; performing task scheduling according to the first scheduling strategy and adjusting the first task priority to obtain a second task priority, where the first scheduling strategy is obtained based on the task waiting time and available computing power resources; calculating the utilization rate of computing power resources after scheduling by the first scheduling strategy, and calculating the optimization effect in combination with the heat exchange efficiency; the heat exchange efficiency is the heat exchange efficiency of the liquid cooling system configured for the device executing the target task. By optimizing the dynamic scheduling of computing power resources and simultaneously optimizing the resource scheduling and cooling strategy, it is ensured that while high-priority tasks are processed in a timely manner, the computing requirements of low-priority tasks are not ignored, and the overall resource utilization rate is improved.

[0153] Embodiment 2

[0154] In a preferred embodiment, the specific steps for judging the optimization effect of the liquid cooling energy consumption system for dynamic scheduling of computing power resources based on the system computing power resource utilization rate and the heat exchange efficiency of the liquid cooling system are as follows:

[0155] Step C1, define the computing power resource utilization rate and the heat exchange efficiency as input variables, and divide them into different fuzzy sets respectively.

[0156] For example, "Low", "Medium", "High" for the computing power resource utilization rate, and "Low", "Medium", "High" for the heat exchange efficiency.

[0157] Step C2, define the optimization effect of the liquid cooling energy consumption system for dynamic scheduling of computing power resources as the output variable, and divide it into a fuzzy set. For example, "Low", "High" for the optimization effect of the liquid cooling energy consumption system for dynamic scheduling of computing power resources.

[0158] Step C3, formulate a set of fuzzy rules to describe the influence of different input variables on the output variable. The definition of the rules can be based on professional knowledge or obtained through data analysis and experiments. For example:

[0159] Mark the computing power resource utilization rate as U, the heat exchange efficiency as J, and the optimization effect of the liquid cooling energy consumption system for dynamic scheduling of computing power resources as P, then it can be defined

[0160] Rule 1: If (U is High) and (J is High), then (P is High)

[0161] Rule 2: If (U is Low) and (J is Low), then (P is Low) ...

[0163] Step C4: Perform fuzzy inference according to fuzzy rules to determine the optimization effect of the liquid cooling energy consumption system for dynamic scheduling of computing power resources.

[0164] It should be noted that the division of fuzzy sets can be adjusted according to actual situations. For example, although three fuzzy sets are used as an example in this embodiment, in fact, the computing power resource utilization rate, heat exchange efficiency, and the optimization effect of the liquid cooling energy consumption system for dynamic scheduling of computing power resources can be divided into more than three sets to facilitate better accurate identification.

[0165] Furthermore, for the judgment of the high or low of the computing power resource utilization rate and heat exchange efficiency, thresholds can be set according to actual situations for judgment; when the computing power resource utilization rate is higher than 80%, it is labeled as "High", and when the heat exchange efficiency is higher than 85%, it is labeled as "High", etc., which will not be elaborated here.

[0166] When P is High, it indicates that the optimization effect of the liquid cooling energy consumption system for dynamic scheduling of computing power resources is good.

[0167] When P is Low, it indicates the optimization effect of the liquid cooling energy consumption system for dynamic scheduling of computing power resources.

[0168] It should be noted that when P is Low, it indicates that the optimization effect of the liquid cooling energy consumption system for dynamic scheduling of computing power resources needs to be improved. At this time, the system can automatically trigger a series of preset adjustment measures, such as increasing the cooling efficiency of the liquid cooling system, optimizing the allocation strategy of computing power resources, etc., in order to achieve a better energy consumption optimization effect.

[0169] In summary, by supporting automated management, the method and system for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources not only achieve the efficient utilization of resources and the significant reduction of energy consumption, but also provide a strong guarantee for the continuous and stable operation of the business through intelligent adjustment and optimization.

[0170] Embodiment 3

[0171] This embodiment also provides a system for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources, including:

[0172] A data acquisition and processing module, configured to acquire target task request information and allocate a first task priority according to the target task request information;

[0173] The target task request information includes at least the task type, required computing power, and urgency;

[0174] A scheduling module, configured to perform task scheduling according to a first scheduling strategy and adjust the first task priority to obtain a second task priority, where the first scheduling strategy is obtained according to the task waiting time and available computing power resources;

[0175] An optimization effect judgment module is used to calculate the utilization rate of computing power resources after scheduling by the first scheduling policy, and calculate the optimization effect in combination with the heat exchange efficiency;

[0176] The heat exchange efficiency is the heat exchange efficiency of the coolant cooling system configured for the device executing the target task.

[0177] Each of the above unit modules can be embedded in the processor of the computer device in hardware form or independent of it, or stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above modules.

[0178] This embodiment also provides a computer device, which can be a terminal, and its internal structure diagram can be as Figure 2 shown. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a carrier network, NFC (Near Field Communication) or other technologies. When the computer program is executed by the processor, it realizes a method for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, a trackball or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0179] This embodiment also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by the processor, the following steps are realized:

[0180] Obtain target task request information, and allocate a first task priority according to the target task request information;

[0181] The target task request information includes at least the task type, the required computing power, and the urgency;

[0182] Perform task scheduling according to the first scheduling policy and adjust the first task priority to obtain a second task priority. The first scheduling policy is obtained according to the task waiting time and the available computing power resources;

[0183] Calculate the utilization rate of computing power resources after scheduling by the first scheduling policy, and calculate the optimization effect in combination with the heat exchange efficiency;

[0184] The heat exchange efficiency is the heat exchange efficiency of the coolant cooling system configured for the device performing the target task.

[0185] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

[0186] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages. For example, object-oriented programming languages such as Java and interpreted scripting languages such as JavaScript.

[0187] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0188] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device realizes the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0189] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus, causing a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, so that the instructions executed on the computer or other programmable apparatus provide steps for implementing the functions specified in one process or a plurality of processes and / or blocks Figure 1 one process or a plurality of processes and / or blocks Figure 1 steps for implementing the functions specified in one block or a plurality of blocks.

[0190] Although the preferred embodiments of the present application have been described, additional changes and modifications can be made to these embodiments by those skilled in the art once they learn of the basic creative concept. Therefore, the appended claims are intended to be construed to cover the preferred embodiments as well as all changes and modifications falling within the scope of the present application.

[0191] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. A method for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources, characterized in that Including: Obtain target task request information and allocate a first task priority according to the target task request information; The target task request information at least includes task type, required computing power, and urgency; Perform task scheduling according to a first scheduling policy and adjust the first task priority to obtain a second task priority, where the first scheduling policy is obtained based on task waiting time and available computing power resources; Calculate the utilization rate of computing power resources after scheduling by the first scheduling policy, and calculate the optimization effect in combination with the heat exchange efficiency; The heat exchange efficiency is the heat exchange efficiency of the liquid cooling system configured for the device executing the target task.

2. The liquid cooling energy consumption optimization method for dynamic scheduling of computing power resources according to claim 1, wherein, The allocating the first task priority according to the target task request information includes: Allocate a priority value to each task according to the task type, required computing power, and urgency; Sort the target tasks according to the priority value; Add the sorted tasks to the task queue in order of priority to wait for scheduling; The task queue at least includes a high-priority task queue, a medium-priority task queue, and a low-priority task queue.

3. The liquid cooling energy consumption optimization method for dynamic scheduling of computing power resources according to claim 2, wherein, The first scheduling policy includes: Obtain the time when the task enters the task queue, and subtract the time when the task enters the task queue from the current time to obtain the waiting time of the task to be processed; Obtain the total computing power resources of the system and the currently used computing power resources, and subtract the currently used computing power resources from the total computing power resources of the system to obtain the available computing power resources; Obtain the first scheduling policy according to the task waiting time and the available computing power resources.

4. The method for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources according to claim 3, wherein The performing task scheduling according to the first scheduling policy and adjusting the first task priority includes: When there is a task to be scheduled, preferentially take out the task from the high-priority task queue; For the high-priority task queue, if the available computing power resources meet the computing power requirements of the task, allocate the task to the corresponding computing node for execution; If the available computing power resources do not meet the computing power requirements of the task, wait until there are enough resources available.

5. The method for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources according to claim 4, wherein The performing task scheduling according to the first scheduling policy and adjusting the first task priority further includes: For tasks in the medium-priority task queue and the low-priority task queue, if the available computing power resources meet the computing power requirements of the task, and the waiting time of the task in the queue is greater than the expected task waiting time, adjust the priority of the task; Perform scheduling according to the adjusted priority. If the adjusted task priorities are equal, preferentially process the task with the higher initial priority.

6. The method for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources according to claim 5, wherein, The calculating the utilization rate of computing power resources after scheduling by the first scheduling policy includes: Obtain the completion time and start time of the task, and subtract the start time of the task from the completion time of the task to obtain the processing time of the task; Obtain the computing power resource usage of the task according to the real-time monitoring of the usage of computing power resources; Calculate the utilization rate of computing power resources based on the total computing power resource amount and total cycle time of the system, in combination with the processing time of the task and the computing power resource usage.

7. The method for optimizing the liquid cooling energy consumption of dynamic scheduling of computing power resources according to claim 6, wherein The second task priority is used as the first task priority for the next round of optimization.

8. A liquid cooling energy consumption optimization system for dynamic scheduling of computing power resources, characterized in that, Including: A data acquisition and processing module, configured to obtain target task request information and allocate a first task priority according to the target task request information; The target task request information at least includes task type, required computing power, and urgency; A scheduling module, configured to perform task scheduling according to a first scheduling policy and adjust a first task priority to obtain a second task priority, where the first scheduling policy is obtained based on task waiting time and available computing power resources; An optimization effect judgment module, configured to calculate the utilization rate of computing power resources after scheduling by the first scheduling policy and calculate the optimization effect in combination with the heat exchange efficiency; The heat exchange efficiency is the heat exchange efficiency of the liquid cooling system configured for the device executing the target task.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images