Aluminum-based PCB micro-channel heat dissipation control method and system

Abstract

The invention belongs to the technical field of heat dissipation control, and discloses an aluminum-based PCB micro-channel heat dissipation control method and system, and the method comprises the steps: obtaining task scheduling information and the temperature information of each core region; according to the task scheduling information, determining a corresponding relation between each calculation task and each core area and a heat dissipation resource allocation weight of each calculation task; based on the temperature information, the heat dissipation resource allocation weight, the corresponding relation between each calculation task and each core area and the total available cooling liquid flow of the system, the target cooling liquid flow of each core area is solved with the goal of minimizing the weighted average temperature of each core area and minimizing the total cooling liquid flow; according to the target cooling liquid flow, each flow control unit is controlled to adjust the cooling liquid flow of each core area; therefore, prospective heat dissipation control can be carried out in combination with the task scheduling information, the temperature stability of the high-priority task area is guaranteed preferentially, resource allocation is optimized under the total resource constraint, the heat dissipation efficiency is improved, and the energy consumption is reduced.

Description

Technical Field

[0001] This application relates to the technical field of heat dissipation control, and more particularly, to an aluminum-based PCB microchannel heat dissipation control method and system. Background Art

[0002] High-performance computing devices are widely used in fields such as industrial control, data centers, and artificial intelligence. Their core configurations usually include multi-core processors. To ensure the stable operation of the processors under high loads, an efficient heat dissipation system is crucial. In the prior art, some advanced heat dissipation solutions adopt liquid cooling technology, where the processors are installed on aluminum-based printed circuit boards (PCBs) with good thermal conductivity. The aluminum-based PCB internally integrates a precise microchannel network that forms a path for the coolant to circulate. Driven by a pump, the coolant circulates in the microchannels, absorbs the heat generated by the processors during operation, and dissipates the heat through a heat exchanger.

[0003] On high-performance computing devices, a complex multi-task operating system is usually run. The task scheduler of the operating system assigns different computing tasks to different core processors according to a preset scheduling strategy. Due to the differences in the computing intensity and execution status of different tasks, as well as the dynamic migration of tasks between cores, the heat generation power and the corresponding regions of each core processor thus bear different heat flux densities, making it easy to form local hotspots. More challenging is that as the operating system task scheduler continuously switches and reallocates tasks between cores, the positions and intensities of these local hotspots will change rapidly and dynamically with the execution status of the tasks and the core allocation.

[0004] To cope with this dynamic and non-uniform heat generation characteristic, the aluminum-based PCB microchannel heat dissipation system is usually designed for zone control. By configuring independent flow control units (such as microvalves) in different regions of the microchannel network (usually corresponding to the positions of the core processors, so it can be called the core regions), independent adjustment of the coolant flow rate in each region can be achieved. Existing zone heat dissipation control methods mostly adopt temperature feedback-based strategies. The system collects real-time temperature values at the temperature sensors integrated inside the aluminum-based PCB or the processors corresponding to the core processors. When the temperature of a certain region exceeds a preset temperature threshold, the control system will respond by increasing the coolant flow rate in that region to enhance the heat dissipation effect and reduce the temperature.

[0005] However, this control method based purely on hysteresis-based temperature feedback has significant limitations. Core processors executing high-priority or high-security tasks have a low tolerance for temperature fluctuations. Even if the temperature has not yet reached the set feedback threshold, a rapid rise or sustained temperature fluctuation may adversely affect the real-time performance, execution accuracy, or stability of the task, and even increase the risk of calculation errors or system instability. Low-priority tasks, on the other hand, have a relatively high tolerance for temperature fluctuations. Traditional temperature feedback control cannot perceive and distinguish these differences in heat dissipation requirements caused by task characteristics, and cannot prioritize temperature stability for critical tasks.

[0006] At the same time, the total resources of the liquid cooling system are limited. For example, there is an upper limit to the total flow rate of the coolant circulation pump. When multiple core processor areas show a trend of rising temperatures at the same time and need to increase the coolant flow rate, if the system simply responds to the flow increase demand of all areas, it may cause the total flow demand to exceed the maximum capacity of the cooling pump, or force the cooling pump to run at high power or even full load, significantly increasing the overall power consumption of the system, and it may still not be possible to effectively control the temperature of all areas, especially the heat dissipation priority of high-priority areas. How to effectively allocate these resources (such as the total coolant flow rate) under the premise that the total resources of the cooling system are limited, give priority to ensuring the temperature stability of areas with high-priority tasks that are sensitive to temperature, while taking into account the overall heat dissipation effect and system energy consumption, is a key technical challenge that needs to be solved urgently.

[0007] Relying solely on temperature feedback, the system cannot predict which areas will soon execute high-priority tasks and, consequently, generate high heat loads. If proactive task scheduling information can be acquired and effectively utilized in real time, it would be possible to pre-adjust coolant flow to relevant areas before tasks actually start, increase computational intensity, or generate higher heat.

[0008] Implementing dynamic, proactive cooling scheduling based on task priority and scheduling information requires the system to accurately and in real time obtain the processor's task scheduling information (including task priority, assigned cores, etc.) and the task execution status of each core. A model or strategy must be established to correlate task priority with the priority or weight of cooling resource allocation. Based on this information, the cooling control system must be able to rapidly calculate and dynamically adjust the coolant flow rates across multiple microchannel zones. Furthermore, this scheduling strategy must balance the cooling requirements of different zones within the constraints of limited total cooling resources, while ensuring the temperature stability of high-priority tasks and optimizing overall system energy consumption.

[0009] In view of the above problems, the existing technology is in urgent need of improvement. Summary of the Invention

[0010] The purpose of this application is to provide an aluminum-based PCB microchannel heat dissipation control method and system, which can perform proactive heat dissipation control in combination with task scheduling information, prioritize ensuring the temperature stability of high-priority task areas, and optimize resource allocation under the total resource constraint, improve the heat dissipation efficiency and reduce energy consumption.

[0011] In the first aspect, this application provides an aluminum-based PCB microchannel heat dissipation control method for controlling the aluminum-based PCB microchannel heat dissipation system of a computing device with multiple core processors. The aluminum-based PCB microchannel heat dissipation system has multiple core areas and multiple flow control units. Each of the core processors is respectively arranged at each of the core areas, and each of the flow control units is respectively used to adjust the coolant flow of each of the core areas. The steps of this method include: A1. Obtain task scheduling information and temperature information of each core area; the task scheduling information includes the task priority of each computing task and the corresponding task-assigned core processor. A2. Determine the correspondence between each computing task and each core area according to the task-assigned core processor of each computing task. A3. Obtain the heat dissipation resource allocation weight of each computing task according to the task priority of each computing task. A4. Based on the temperature information of each core area, the heat dissipation resource allocation weight of each computing task, the correspondence between each computing task and each core area, and the total available coolant flow of the system, with the goal of minimizing the weighted average temperature of each core area and minimizing the total coolant flow, use the resource allocation algorithm to solve the target coolant flow of each core area. A5. Control each flow control unit to adjust the coolant flow of each core area according to the target coolant flow of each core area.

[0012] Preferably, step A3 includes: A301. Query the task priority and heat dissipation resource allocation weight mapping table according to the task priority of each computing task to obtain the heat dissipation resource allocation weight of each computing task.

[0013] Preferably, the task scheduling information further includes the task type and task volume information of each computing task; After step A301, it includes: A302. Determine the energy consumption level of each computing task according to the task type and task volume information of each computing task. A303. Adjust the heat dissipation resource allocation weight of each computing task according to the energy consumption level of each computing task to obtain the final heat dissipation resource allocation weight.

[0014] Preferably, step A302 includes: Query the mapping table of task type and processor frequency impact factor according to the task type of each computing task to obtain the corresponding frequency impact factor; Estimate the processor frequency increase value of each computing task during execution according to the task volume information and frequency impact factor of each computing task; Calculate the energy consumption level of each computing task according to the processor frequency increase value and task volume information.

[0015] Preferably, step A303 includes: Take each core processor as the target core in turn, and obtain the set of all core processors in the neighborhood centered on the target core as the neighborhood core set; Calculate the thermal impact factor of each neighborhood core on the target core according to the preset thermal conductivity coefficient and the distance between each neighborhood core and the target core in the neighborhood core set; among them, the thermal impact factor of the target core on itself is 1, and the thermal impact factor of other neighborhood cores on the target core is proportional to the thermal conductivity coefficient and inversely proportional to the distance; Calculate the sum of the products of the thermal impact factors of all neighborhood cores and the energy consumption levels of the corresponding computing tasks to adjust the heat dissipation resource allocation weight of the target core and obtain the final heat dissipation resource allocation weight.

[0016] Preferably, the resource allocation algorithm is a linear programming algorithm, a quadratic programming algorithm or a model predictive control algorithm.

[0017] Preferably, step A4 includes: A401. Construct an optimization objective function, which includes minimizing the weighted average temperature of each core area and minimizing the total coolant flow; among them, the weight of each core area is determined by the heat dissipation resource allocation weight of each computing task and the corresponding relationship between each computing task and each core area, and the weighted average temperature is obtained by multiplying the temperature information of each core area by the corresponding weight, summing and then dividing by the total weight; A402. Use an optimization algorithm based on sequential quadratic programming, with the coolant flow of each core area as the optimization variable, and solve the optimization objective function under the constraints to obtain the target coolant flow of each core area; the constraints include the upper and lower limits of the coolant flow of each core area and the total available coolant flow of the system.

[0018] Preferably, step A402 includes: B1. Initialize the parameters of the sequential quadratic programming algorithm. The initialized parameters include the initial coolant flow values of each core area, the Hessian matrix, the Lagrange multiplier estimate value, and the convergence tolerance; B2. Calculate the value and gradient of the optimization objective function based on the coolant flow rate values in each core region at the current iteration step, and calculate the constraint conditions and their Jacobian matrices. B3. Construct a quadratic programming sub-problem, which takes the coolant flow rate values in each core region at the current iteration step as the center, and based on the calculated value and gradient of the optimization objective function and the estimated Lagrange multiplier, performs a quadratic approximation of the optimization objective function and a linear approximation of the constraint conditions; among them, the quadratic approximation is approximated by the Hessian matrix, and the linear approximation is approximated by the Jacobian matrix. B4. Use the active set method to solve the quadratic programming sub-problem to obtain the search direction and search step size of the coolant flow rate values in each core region. B5. Update the coolant flow rate values in each core region according to the search direction and search step size of the coolant flow rate values in each core region, and adjust the search step size according to the rmijo line search strategy to ensure that the value of the objective function decreases and the constraint conditions are satisfied. B6. Determine whether the convergence condition determined by the convergence tolerance is satisfied. If satisfied, use the latest coolant flow rate values in each core region as the target coolant flow rate in each core region; otherwise, update the Hessian matrix and the estimated Lagrange multiplier using the BFGS formula, and return to step B2.

[0019] Preferably, after step A402, it further includes: A403. Obtain the task content categories of each computing task, query the preset mapping table of task content categories and temperature sensitivity levels, and obtain the temperature sensitivity levels corresponding to each computing task. A404. Adjust the target coolant flow rate in each core region according to the temperature sensitivity levels of each computing task using the preset adjustment rules.

[0020] In a second aspect, the present application provides an aluminum-based PCB microchannel heat dissipation control system for controlling the aluminum-based PCB microchannel heat dissipation system of a computing device with multiple core processors. The aluminum-based PCB microchannel heat dissipation system has multiple core regions and multiple flow control units. Each of the core processors is respectively arranged at each of the core regions, and each of the flow control units is respectively used to adjust the coolant flow rate in each of the core regions; the system includes: An information acquisition module, configured to acquire task scheduling information and temperature information in each core region; the task scheduling information includes the task priority of each computing task and the corresponding task-assigned core processor. A mapping module, configured to determine the correspondence between each computing task and each core region according to the task-assigned core processor of each computing task. A weight determination module, configured to obtain the heat dissipation resource allocation weights for each computing task according to the task priorities of each computing task; An optimization calculation module, configured to, based on the temperature information of each core area, the heat dissipation resource allocation weights for each computing task, the correspondence between each computing task and each core area, and the total available coolant flow rate of the system, with the goal of minimizing the weighted average temperature of each core area and minimizing the total coolant flow rate, use a resource allocation algorithm to solve the target coolant flow rate for each core area; A control execution module, configured to control each flow control unit to adjust the coolant flow rate of each core area according to the target coolant flow rate of each core area.

[0021] Beneficial effects: An aluminum-based PCB microchannel heat dissipation control method and system provided by the present application, by combining task scheduling information and temperature information, determining the heat dissipation resource allocation weights based on task priorities, and using an optimization algorithm to solve the target coolant flow rate, can perform forward-looking heat dissipation control in combination with task scheduling information, prioritize ensuring the temperature stability of high-priority task areas, and optimize resource allocation under the total resource constraint, improve the heat dissipation efficiency and reduce energy consumption. Description of the Drawings

[0022] Figure 1 It is a flowchart of the aluminum-based PCB microchannel heat dissipation control method provided by an embodiment of the present application.

[0023] Figure 2 It is a structural schematic diagram of the aluminum-based PCB microchannel heat dissipation control system provided by an embodiment of the present application.

[0024] Label description: 1. Information acquisition module; 2. Mapping module; 3. Weight determination module; 4. Optimization calculation module; 5. Control execution module. Detailed Embodiments

[0025] Next, the technical solutions in the present application will be clearly and completely described in conjunction with the accompanying drawings in the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the present application usually described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application to be protected, but only represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present application.

[0026] It should be noted that similar reference numerals and letters denote similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. At the same time, in the description of the present application, the terms "first", "second", etc. are only used for distinguishing descriptions and cannot be construed as indicating or implying relative importance.

[0027] Referring Figure 1 , the present application proposes an aluminum-based PCB microchannel heat dissipation control method for controlling the aluminum-based PCB microchannel heat dissipation system of a computing device with multiple core processors. The aluminum-based PCB microchannel heat dissipation system has multiple core regions and multiple flow control units. Each of the core processors is respectively arranged at each of the core regions, and each of the flow control units is respectively used to adjust the coolant flow rate of each of the core regions. The steps of the method include: A1. Obtain task scheduling information and temperature information of each core region; the task scheduling information includes the task priorities of various computing tasks and the corresponding core processors to which the tasks are assigned; A2. Determine the correspondence between various computing tasks and each core region according to the core processors to which the various computing tasks are assigned; A3. Obtain the heat dissipation resource allocation weights of various computing tasks according to the task priorities of various computing tasks; A4. Based on the temperature information of each core region, the heat dissipation resource allocation weights of various computing tasks, the correspondence between various computing tasks and each core region, and the total available coolant flow rate of the system, with the goal of minimizing the weighted average temperature of each core region and minimizing the total coolant flow rate, use a resource allocation algorithm to solve the target coolant flow rate of each core region; A5. Control each flow control unit to adjust the coolant flow rate of each core region according to the target coolant flow rate of each core region.

[0028] Among them, the task scheduling information refers to the data provided by the operating system about the computing tasks that are running or about to run, including the importance of various computing tasks and the information about which core processor the corresponding task is assigned to execute. Its main purpose is to provide forward-looking input about the task characteristics, so that the system can predict the differences in heat dissipation requirements in different regions.

[0029] Among them, the heat dissipation resource allocation weight is a value determined according to factors such as the task priority of each computing task, and is used to quantify the importance or priority level of this task or its area in heat dissipation resource allocation. It can be implemented by querying the mapping table of task priority and heat dissipation resource allocation weight, which is mainly to convert task characteristics into input parameters of the optimization algorithm. The mapping table of task priority and heat dissipation resource allocation weight can be determined in advance through data statistics or expert experience and stored in the local database.

[0030] Among them, the resource allocation algorithm is a mathematical optimization method used to solve how to allocate limited resources to multiple demand parties under given constraints. It can be implemented by using linear programming algorithms, quadratic programming algorithms or model predictive control algorithms. Its main purpose is to find the optimal coolant flow allocation scheme. Minimizing the weighted average temperature of each core area and minimizing the total coolant flow are two optimization goals that the resource allocation algorithm needs to achieve. Minimizing the weighted average temperature aims to preferentially reduce the temperature of high-weight areas and ensure the stability of critical tasks. Minimizing the total coolant flow aims to reduce pump power consumption and improve system energy efficiency. The weighted average temperature is the temperature value obtained by multiplying the real-time temperature values of each core area by the heat dissipation resource allocation weights of the corresponding areas, summing them up and then dividing by the total weight. Among them, the weights of each core area are determined by the heat dissipation resource allocation weights of each computing task and the corresponding relationship between each computing task and each core area, which is mainly to reflect the relative importance of different areas in heat dissipation guarantee.

[0031] The core innovation of this application lies in combining forward-looking task scheduling information, especially task priority, with real-time core area temperature information, and using the resource allocation algorithm. Under the constraint of the total available coolant flow of the system, with the goal of minimizing the weighted average temperature of each core area and minimizing the total coolant flow, dynamically solving and controlling the target coolant flow of each core area, thereby overcoming the lag of traditional pure temperature feedback control, realizing differential heat dissipation guarantee based on task importance, improving the utilization efficiency of heat dissipation resources, and reducing the overall energy consumption of the system.

[0032] Specifically, the working principle of this method is as follows: First, the system obtains the task scheduling information provided by the operating system, including the task priorities of various computing tasks and the corresponding task-assigned core processors, and simultaneously collects the real-time temperature information of each core area. This information provides a comprehensive input basis for subsequent intelligent decision-making. Then, the system determines on which physical heat dissipation area each computing task specifically runs according to the task-assigned core processors of various computing tasks, establishing an association between the abstract task information and the physical heat dissipation area. Next, the system determines the importance of each computing task in the heat dissipation resource allocation according to the task priorities of various computing tasks, and quantifies it as the heat dissipation resource allocation weight. Further, the system comprehensively utilizes the collected real-time temperature information, the determined heat dissipation resource allocation weight, the correspondence between tasks and areas, and the current total available coolant flow rate of the system. Using the weighted average temperature reflecting the task importance and the total coolant flow rate minimizing the system energy consumption as the optimization objectives, and applying the resource allocation algorithm, it calculates the optimal coolant flow rate value that each core area should obtain at this moment, that is, the target coolant flow rate of each core area. Finally, the system sends control instructions to the flow control units corresponding to each core area according to the calculated target coolant flow rate value, precisely adjusting the coolant flow rate flowing through each area. This entire process is executed periodically, enabling the heat dissipation system to adjust the heat dissipation intensity of each area in real time according to the dynamic changes of task scheduling and temperature, achieving intelligent, efficient, and differentiated heat dissipation control.

[0033] As a preferred embodiment, the solution of this application is specifically implemented as follows: On a computing device with multiple core processors, its aluminum-based PCB heat dissipation system includes multiple core areas, each area corresponding to a core processor and equipped with a microvalve as the flow control unit. Through the interface, it obtains the task list provided by the operating system, including the ID, priority, and assigned core number of each task. At the same time, temperature sensors collect the real-time temperature of multiple core areas. According to the assigned core number of the task, it determines on which core area the task runs. According to the task priority, it queries a preset priority-weight mapping table to obtain the heat dissipation resource allocation weight of each task. If there are multiple tasks in an area, the weights can be aggregated (such as taking the mean or maximum value, but not limited to this) to obtain the total area weight. It constructs an optimization problem, and the objective function includes minimizing the weighted average temperature of each core area and minimizing the total coolant flow rate. The constraint conditions include the upper and lower limits of the flow rate of each area and the total flow rate constraint. It uses the sequential quadratic programming algorithm to solve this optimization problem to obtain the target coolant flow rate values of multiple core areas. The control execution module sends control signals to the corresponding microvalves according to the calculated target coolant flow rate values of each area, adjusting their opening degrees to make the actual flow rate close to the target flow rate.

[0034] Through the above solution, the present application realizes differential heat dissipation guarantee for tasks of different importance by introducing task priority information, preferentially ensures the temperature stability of the area where high-priority tasks are located, and reduces the risk brought to the execution of critical tasks due to temperature fluctuations. Combining real-time temperature feedback and forward-looking task information, the system can predict and respond to heat load changes more timely and accurately, improving the responsiveness and accuracy of heat dissipation control. Under the constraint of limited total coolant flow of the system, the resources are intelligently allocated through an optimization algorithm, avoiding resource overrun or inefficient operation caused by simply responding to the temperature rise in all areas, and improving the utilization efficiency of heat dissipation resources. By taking minimizing the total coolant flow as one of the optimization objectives, the operating load of the cooling pump is reduced on the premise of meeting the heat dissipation requirements, thereby effectively reducing the overall power consumption of the system. Overall, this method can effectively cope with the dynamic and uneven heat generation characteristics of multi-core processors, improving the heat dissipation performance, stability and energy efficiency of high-performance computing devices.

[0035] In some embodiments, step A3 includes: A301. Query the task priority and heat dissipation resource allocation weight mapping table according to the task priorities of each computing task to obtain the heat dissipation resource allocation weights of each computing task.

[0036] The task priority and heat dissipation resource allocation weight mapping table refers to a data structure used to store the association relationship between task priorities and corresponding heat dissipation resource allocation weights, which can be implemented in the form of a table, an array, a hash table or a database, etc.

[0037] This solution provides a specific, operable, and stable implementation mechanism by defining a specific method for obtaining the weight of heat dissipation resource allocation. After obtaining the task scheduling information, including the task priorities of various computing tasks, the system no longer relies on vague rules or empirical judgments to determine the weight of heat dissipation resource allocation. Instead, according to the task priorities of various computing tasks, it directly queries the pre-established mapping table of task priorities and heat dissipation resource allocation weights. This mapping table corresponds different task priority levels or values to the preset heat dissipation resource allocation weight values one by one. Through a simple table lookup operation, the system can quickly, accurately, and stably obtain the heat dissipation resource allocation weight that matches the current task priority. The obtained heat dissipation resource allocation weight is then used in the subsequent resource allocation algorithm as an important input affecting the calculation of the target coolant flow in each core area. This table-based method converts the abstract task priority into a specific and quantifiable weight, providing reliable basic data for subsequent optimization calculations. In this way, the system can ensure that high-priority tasks obtain a higher heat dissipation resource allocation weight, so that when the resource allocation algorithm solves for the target coolant flow, it can preferentially meet the heat dissipation requirements of the core area where high-priority tasks are located. Even when the total coolant flow is limited, it can achieve differential heat dissipation control based on task priorities. The combination of this clear and stable weight acquisition mechanism and the subsequent resource allocation algorithm makes the entire heat dissipation control process more accurate and reliable.

[0038] For example, the task priority can be divided into multiple levels, such as 1 to 10, where 10 represents the highest priority. The mapping table of task priorities and heat dissipation resource allocation weights can be preset as follows: priority 1 corresponds to weight value 1, priority 2 corresponds to weight value 2,..., priority 10 corresponds to weight value 10. When the system obtains that the task priority of a certain computing task is 8, it queries this mapping table according to priority 8 and obtains the corresponding heat dissipation resource allocation weight of 8. If the task priority of another computing task is 3, then querying the mapping table gives a weight of 3. These obtained weight values are then input into the resource allocation algorithm to calculate the target coolant flow in each core area.

[0039] Preferably, the task scheduling information further includes the task type and task volume information of each computing task; After step A301, it may further include: A302. Determine the energy consumption level of each computing task according to the task type and task volume information of each computing task; A303. Adjust the heat dissipation resource allocation weight of each computing task according to the energy consumption level of each computing task to obtain the final heat dissipation resource allocation weight.

[0040] Specifically, in addition to task priority and the core processor to which the task is assigned, the task scheduling information further includes the task type and task volume information of each computing task. The task type can indicate the computing characteristics of the task, such as whether it is compute-intensive, I / O-intensive, memory-intensive, etc. The task volume information can indicate the scale of the task, such as the amount of data to be processed (corresponding to compute-intensive), the number of instructions to be executed (corresponding to I / O-intensive), or the duration (corresponding to memory-intensive), etc. These information provide a richer description of task attributes than a single priority and are the basis for evaluating the impact of the task on the processor's thermal load during actual execution. According to the task type and task volume information of each computing task, the energy consumption level of each computing task is determined.

[0041] The energy consumption level is a quantitative or classified representation of the heat or power consumption level expected to be generated by the task during execution. The process of determining the energy consumption level can be based on preset rules, models, or look-up table methods to convert the task type and task volume information into an indicator reflecting its thermal load potential. For example, tasks can be classified into different levels such as high energy consumption, medium energy consumption, and low energy consumption. According to the energy consumption level of each computing task, the initially obtained heat dissipation resource allocation weights are adjusted to obtain the final heat dissipation resource allocation weights. This adjustment process aims to incorporate the actual thermal load potential of the task into the consideration of heat dissipation resource allocation. The adjustment can be achieved through a function, a look-up table, or a set of rules (the specific function or rules can be set according to actual needs and are not limited here), and the initial weights based on priority are corrected according to the energy consumption level of the task. For example, for tasks with a high energy consumption level, their initial weights may be adjusted upward; for tasks with a low energy consumption level, their initial weights may be adjusted downward. This adjustment enables the final heat dissipation resource allocation weights to more comprehensively reflect the importance of the task and its actual demand for the heat dissipation system.

[0042] Based on the preliminary heat dissipation resource allocation weights obtained by querying the mapping table according to the task priorities of various computing tasks, this solution further improves the process of determining the weights. The task scheduling information obtained by the system not only includes task priorities and the core processors to which tasks are assigned, but also adds information on the task types and task volumes of various computing tasks. After obtaining the preliminary weights, the system uses this newly added task type and task volume information to determine the energy consumption levels of various computing tasks through an evaluation process. The energy consumption level reflects the heat level expected to be generated during the execution of the task. Subsequently, the system corrects the preliminary heat dissipation resource allocation weights obtained based on priorities previously. For example, for tasks with the same priority but different energy consumption levels, the task with a higher energy consumption level will be assigned a higher final heat dissipation resource allocation weight, while the task with a lower energy consumption level may be assigned a lower final weight. This adjustment enables the final heat dissipation resource allocation weights to more accurately reflect the importance of the tasks and their actual requirements for the heat dissipation system. Thus, when solving for the target coolant flow rate based on these weights subsequently, the system can more effectively prioritize the allocation of limited coolant resources to the core areas where those important and high-heat-generating tasks are located, thereby better balancing the heat dissipation requirements of each core area under the constraint of the total coolant flow rate, ensuring the temperature stability of critical tasks first, and at the same time avoiding waste caused by over-allocation of resources to low-energy-consuming tasks. This method of determining the heat dissipation resource allocation weights by combining task priorities, task types, task volumes, and energy consumption levels can more precisely evaluate the actual heat dissipation requirements of tasks compared to relying solely on task priorities, improving the pertinence and efficiency of heat dissipation control.

[0043] Preferably, step A302 may include: Query the task type and processor frequency impact factor mapping table according to the task types of various computing tasks to obtain the corresponding frequency impact factors; Estimate the processor frequency increase value during the execution of each computing task based on the task volume information and frequency impact factors of each computing task; Calculate the energy consumption levels of various computing tasks based on the processor frequency increase value and task volume information.

[0044] Among them, the task type and processor frequency impact factor mapping table refers to a data structure that stores the association relationship between different computing task types and the corresponding frequency impact factors, which can be implemented using a lookup table, a database, or a configuration file.

[0045] Among them, the frequency impact factor refers to a numerical index that quantifies the tendency of a specific task type to trigger the processor to increase its operating frequency, which can be determined using pre-calibrated or empirical values.

[0046] Among them, the processor frequency increase value refers to the increase in the processor operating frequency during task execution relative to the base frequency, which can be estimated using a calculation model or prediction algorithm based on the task volume and frequency impact factor. For example, the following calculation formula can be used to estimate the processor frequency increase value: △P = P0 * (1 + k * λ * K), where △P is the processor frequency increase value, P0 is the base frequency, k is the task volume ratio (the ratio of the task volume to the preset base task volume), λ is the frequency impact factor, and K is the proportionality coefficient (a preset adjustment parameter).

[0047] Among them, the energy consumption level is a grading index for measuring the heat generation intensity during the execution of a computing task, which can be represented by discrete levels or continuous values. A formula or a look-up table can be used to determine the energy consumption level of each computing task. For example, the energy consumption level can be proportional to the product of the task volume and the frequency increase value, or the tasks can be further divided into three energy consumption levels: low, medium, and high according to the calculation results. For example, if the calculation result falls within a certain range, it is determined as the "high energy consumption level". In this way, an energy consumption level reflecting the expected heat generation intensity can be calculated for each computing task.

[0048] This solution further refines the specific process of determining the energy consumption level of each computing task. This process first queries the preset mapping table of task type and processor frequency impact factor according to the task type of each computing task to obtain the corresponding frequency impact factor. This step identifies the potential impact of different task types on the dynamic adjustment of the processor frequency. Then, combining the task volume information of each computing task and the obtained frequency impact factor, the possible frequency increase value of the processor during the execution of each computing task is estimated. The task volume reflects the total computing requirements of the task, and the frequency impact factor reflects the tendency of the task type to drive frequency increase. Combining the two can more accurately predict the working intensity that the processor may reach to complete the task. The increase in the processor frequency is the key factor leading to a significant increase in power consumption and heat generation. Therefore, accurately estimating the frequency increase value is crucial for evaluating the actual heat generation potential. Finally, according to the estimated processor frequency increase value and the task volume information, the energy consumption level of each computing task is calculated. The energy consumption level comprehensively reflects the total workload of the task and the processing intensity per unit time, thus providing an index that is closer to the actual heat generation situation during task execution.

[0049] Through the above steps, this solution can more comprehensively and accurately evaluate the energy consumption level of each computing task, because it not only considers the task type and task amount, but also further considers the potential impact of task execution on processor frequency. Applying this more accurate energy consumption level information to the subsequent cooling resource allocation weight adjustment process can enable the cooling control system to more accurately predict and respond to local hotspots, especially when the processor is running at high load or performing critical tasks, thereby optimizing the cooling effect and improving system stability and energy efficiency. This method of determining energy consumption level by considering task type, task amount, and the potential impact of the task on processor frequency can more accurately capture the heat generation characteristics of the task than methods that rely solely on task type and task amount, providing a more solid foundation for forward-looking cooling control based on task characteristics.

[0050] Preferably, step A303 may include: Taking each core processor as the target core in turn, obtain the set of all core processors in the neighborhood centered on the target core as the neighborhood core set (including the target core itself); Based on the preset heat transfer coefficient and the distance between each neighborhood core and the target core in the neighborhood core set, the heat impact factor of each domain core on the target core is calculated; the heat impact factor of the target core on itself is 1, and the heat impact factors of other domain cores on the target core are proportional to the heat transfer coefficient and inversely proportional to the distance; The sum of the products of the thermal impact factors of all neighboring cores and the energy consumption levels of the corresponding computing tasks is calculated to adjust the heat dissipation resource allocation weight of the target core and obtain the final heat dissipation resource allocation weight.

[0051] The neighborhood core set refers to a group of core processors physically adjacent to or close to the target core processor. The range of the neighborhood core set can be determined based on the physical layout and thermal conductivity of the processors. Specifically, the size and shape of the neighborhood core set can be predetermined based on the physical layout and thermal conductivity of the processors. For example, it can be a circular area with a preset radius.

[0052] The preset thermal conductivity coefficient refers to a parameter reflecting the thermal conductivity of a heat dissipation medium such as an aluminum-based PCB, and the parameter can be obtained through experiments or table lookup based on actual material properties.

[0053] The thermal impact factor refers to a value that quantifies the degree to which the heat generated by the neighboring core affects the temperature of the target core through heat conduction. This value reflects the efficiency of heat conduction from the neighboring core to the target core.

[0054] The technical solution of this application further adjusts the cooling resource allocation weight determined based on the energy consumption level of the task itself by introducing the consideration of the thermal coupling effect between cores in a multi-core processor system. Specifically, the solution processes each core processor as the target core in sequence. For each target core, first, the set of core processors within its physical neighborhood, that is, the neighborhood core set, is determined. Then, according to the heat conduction characteristics of the cooling medium and the physical distances between each core in the neighborhood and the target core, the thermal influence factor of each neighborhood core on the target core is calculated. The thermal influence factor of the target core itself is set to 1, indicating that the heat generated by itself has a direct and maximum impact on its own temperature. The thermal influence factors of other neighborhood cores are set to be proportional to the thermal conductivity coefficient and inversely proportional to the distance, which conforms to the basic physical law of heat conduction: the better the thermal conductivity of the medium and the closer the distance, the greater the heat conduction influence. Subsequently, the solution calculates the sum of the products of the thermal influence factors of all cores within the neighborhood core set and the energy consumption levels of the tasks running on them. This sum of products comprehensively reflects the impact of the total heat load from its own heat generation and neighborhood heat conduction on the temperature of the target core in the target core area. Finally, this sum of products is used to adjust the initially determined cooling resource allocation weight of the target core to obtain the final cooling resource allocation weight that can better reflect the actual heat load distribution. For example, if the energy consumption of the task on the target core itself is not high, but there are multiple high-energy-consuming tasks in its neighborhood and the distance is relatively close, the calculated sum of products will be relatively large, so that the cooling resource allocation weight of the target core is adjusted upward to cope with the additional heat load from the neighborhood. This adjustment method combines the energy consumption information of the task with the thermal coupling effect between cores, making the cooling resource allocation weight consider not only the cooling requirements of the task itself but also the thermal environment impact of its physical location, so that the subsequent cooling resource allocation based on this weight is more accurate and effective. Compared with the adjustment based only on the energy consumption of the task itself, this adjustment method considering thermal coupling can more accurately reflect the actual heat load distribution. Especially when there are local hot spots or areas with high-energy-consuming task aggregations, it can more effectively identify and cope with the cooling requirements of these areas.

[0055] In a specific embodiment, the neighborhood core set can be defined as all core processors whose Manhattan distance from the target core in the two-dimensional processor array is less than or equal to a preset threshold (for example, the threshold is 2). The distance between cores can be represented by their grid distance in the processor array. The preset thermal conductivity coefficient can be set according to the thermal conductivity of the aluminum-based PCB material. The thermal influence factor can be specifically calculated as follows: for the target core itself, the thermal influence factor is 1; for other cores in the neighborhood, the thermal influence factor can be calculated as h / d, where h is a constant positively correlated with the thermal conductivity coefficient, and d is the distance between the neighborhood core and the target core. After calculating the sum of the products of the thermal influence factors of all neighborhood cores and the energy consumption levels of the corresponding computing tasks, the sum value can be multiplied by an adjustment coefficient and then added to or multiplied by the heat dissipation resource allocation weight initially determined by the target core based on task priority and its own energy consumption to obtain the final weight. For example, A1 = A0 + M * (∑(ri * Ei)), where A1 is the final heat dissipation resource allocation weight, A0 is the heat dissipation resource allocation weight before adjustment, M is the preset adjustment coefficient, ri is the thermal influence factor of the i-th neighborhood core, Ei is the energy consumption level of the computing task corresponding to the i-th neighborhood core, and ∑(ri * Ei) represents the sum of the products of the thermal influence factors of all neighborhood cores and the energy consumption levels of the corresponding computing tasks.

[0056] Through the above technical solution, the heat dissipation resource allocation weight can more accurately reflect the actual thermal load distribution in the multi-core processor system, including the heat generated by the task itself and the thermal coupling effect between cores. This enables subsequent heat dissipation resource allocation to more effectively respond to dynamically changing local hotspots, prioritize ensuring the temperature stability of the area where high-priority tasks are located, and optimize the overall heat dissipation effect and improve the heat dissipation efficiency under limited total heat dissipation resources.

[0057] Preferably, in step A4, the resource allocation algorithm can adopt a linear programming algorithm, a quadratic programming algorithm, or a model predictive control algorithm.

[0058] Among them, the linear programming algorithm refers to a mathematical optimization method for solving problems where both the objective function and the constraint conditions are linear functions. The quadratic programming algorithm refers to a mathematical optimization method for solving problems where the objective function is a quadratic function and the constraint conditions are linear functions. The model predictive control algorithm refers to a control strategy that predicts the future behavior of the system based on the system model and solves an optimization problem in each control cycle to determine the control action.

[0059] By limiting the resource allocation algorithm to a linear programming algorithm, a quadratic programming algorithm, or a model predictive control algorithm, the present application can quickly and accurately solve the target coolant flow rate in each core area, overcome the problems of solving efficiency and accuracy that may be caused by not limiting the algorithm type, enhance the processing ability for complex constraint conditions, and thus improve the real-time response ability and optimization effect of the system.

[0060] In some possible implementation manners, step A4 includes: A401. Construct an optimization objective function, where the optimization objective function includes minimizing the weighted average temperature of each core region and minimizing the total coolant flow rate; wherein, the weight of each core region is determined by the heat dissipation resource allocation weight of each computing task and the corresponding relationship between each computing task and each core region, and the weighted average temperature is obtained by multiplying the temperature information of each core region by the corresponding weight, summing them up, and then dividing by the total weight; A402. Adopt an optimization algorithm based on sequential quadratic programming, use the coolant flow rate of each core region as the optimization variable, and solve the optimization objective function under the constraints of the constraint conditions to obtain the target coolant flow rate of each core region; the constraint conditions include the upper and lower limits of the coolant flow rate of each core region and the total available coolant flow rate of the system.

[0061] Among them, the optimization objective function refers to a mathematical expression used to quantify the objective to be optimized, such as minimizing temperature or flow rate. It can be a single function that combines multiple objectives (such as weighted summation) or a multi-objective function. The weighted average temperature refers to an index for measuring the overall temperature state, which reflects its relative importance or temperature sensitivity by assigning different weights to different regions, so as to highlight the temperature impact of key regions when calculating the average value.

[0062] Among them, the optimization algorithm based on sequential quadratic programming refers to a method for iteratively solving non-linear constrained optimization problems. It approximates the original problem by solving a quadratic programming sub-problem in each iteration. The optimization variable refers to the variable whose value needs to be determined during the optimization process to achieve the optimal objective. In this solution, it is the coolant flow rate of each core region. The constraint conditions refer to the restrictions that the values of the optimization variables must satisfy. These restrictions can be physical (such as flow rate upper and lower limits) or system resource-related (such as total available flow rate).

[0063] On the basis of the basic method, this solution further refines the steps for solving the target coolant flow rate in each core area. First, by constructing a clear optimization objective function, the heat dissipation resource allocation problem is transformed into a computable mathematical problem. This objective function comprehensively considers two interrelated objectives: minimizing the weighted average temperature in each core area and minimizing the total coolant flow rate. The introduction of the weighted average temperature, combined with the weights determined by factors such as task priorities, enables the optimization process to prioritize the temperature stability of core areas that are more sensitive to temperature or carry more important tasks. At the same time, the objective of minimizing the total coolant flow rate is directly related to the reduction of system energy consumption. This dual-objective design can balance the heat dissipation requirements of key areas and the reduction of overall energy consumption through the adjustment of weights. The specific calculation method of the weighted average temperature is also clarified, providing a basis for subsequent mathematical solutions. Second, the solution clearly adopts an optimization algorithm based on sequential quadratic programming to solve the constructed optimization problem. The sequential quadratic programming algorithm is an effective method for dealing with nonlinear constrained optimization problems. Its iterative solution process can gradually approach the optimal solution and usually has good convergence. By setting the coolant flow rate in each core area as the optimization variable, the algorithm searches for the flow rate combination that minimizes the value of the optimization objective function under the constraints of the upper and lower limits of the flow rate in each area and the total available coolant flow rate of the system. This method provides a structured and operable solution process, which can dynamically calculate the optimal coolant flow rate allocation scheme for each area according to real-time temperature, task information, and system resource conditions.

[0064] For example, in a specific implementation, the optimization objective function can be constructed in the form of a weighted sum, such as: Min(w_t*(Sum(we_i*T_i) / Sum(we_i))+w_f*Sum(F_i)), where w_t and w_f are weight coefficients used to balance the temperature objective and the flow rate objective, T_i is the temperature of the i-th core region, we_i is the weight of the i-th core region, and F_i is the coolant flow rate of the i-th core region. The weight we_i of the core region can be calculated based on the heat dissipation resource allocation weights of the various computing tasks assigned to this region and their corresponding relationships. For example, the maximum value or weighted average of the task weights in this region can be taken. The optimization algorithm based on sequential quadratic programming can be implemented using existing numerical optimization libraries. For example, an optimization solver that supports the sequential quadratic programming algorithm can be used. During the solution process, the coolant flow rate F_i of each core region is used as the optimization variable, and constraint conditions are added, such as F_i_min <= F_i <= F_i_max (where F_i_min and F_i_max are the physical limits of the flow control unit of the i-th core region) and Sum(F_i) <= Fz (where Fz is the total available flow rate provided by the coolant pump). The optimization algorithm iteratively calculates until the flow rate values of each region that satisfy the constraint conditions and minimize the objective function value are found, that is, the target coolant flow rates of each core region are obtained.

[0065] By constructing a clear optimization objective function and using an optimization algorithm based on sequential quadratic programming for solution, this solution provides a specific and operable implementation method for the resource allocation algorithm. This enables the system to dynamically calculate the optimal coolant flow rate allocation scheme that satisfies the constraint conditions according to real-time temperature and task information. This solution can effectively balance the heat dissipation requirements of different core regions, prioritize ensuring the temperature stability of the regions where high-priority tasks are located, and at the same time minimize the total coolant flow rate, so as to achieve more accurate and efficient heat dissipation control under limited total heat dissipation resources and improve the overall temperature stability and energy efficiency of the system.

[0066] Preferably, step A402 may include: B1. Initialize the parameters of the sequential quadratic programming algorithm. The initialized parameters include the initial coolant flow rate values of each core region, the Hessian matrix, the Lagrange multiplier estimate value, and the convergence tolerance; B2. Based on the coolant flow rate values of each core region in the current iteration step, calculate the objective function value and its gradient of the optimization objective function, and calculate the constraint conditions and their Jacobian matrix; B3. Construct a quadratic programming sub - problem, which centers around the coolant flow rate values of each core region at the current iteration step. Based on the calculated optimal objective function value, its gradient, and the estimated Lagrange multiplier value, perform a quadratic approximation of the optimal objective function and a linear approximation of the constraint conditions; among them, the quadratic approximation uses the Hessian matrix for approximation, and the linear approximation uses the Jacobian matrix for approximation; B4. Use the active set method to solve the quadratic programming sub - problem to obtain the search direction and search step size of the coolant flow rate values of each core region; B5. Update the coolant flow rate values of each core region according to the search direction and search step size of the coolant flow rate values of each core region, and adjust the search step size according to the rmijo line search strategy to ensure that the objective function value decreases and the constraint conditions are satisfied; B6. Determine whether the convergence condition determined by the convergence tolerance is met. If it is met, use the latest coolant flow rate values of each core region as the target coolant flow rate of each core region; otherwise, update the Hessian matrix and the estimated Lagrange multiplier value using the BFGS formula and return to step B2.

[0067] Among them, the parameters of the sequential quadratic programming algorithm refer to a set of numerical values required to start and control the sequential quadratic programming iteration process, which may include the initial coolant flow rate values of each core region for starting the iteration, the initial Hessian matrix for approximating the second - order information of the objective function, the initial estimated Lagrange multiplier value for handling constraints, and the convergence tolerance for determining when the algorithm stops. The Hessian matrix refers to a square matrix containing the second - order partial derivatives of the objective function, which can be used to perform a quadratic approximation of the objective function at the current point. The estimated Lagrange multiplier value refers to the current estimated value of the auxiliary variable introduced in the constrained optimization problem, which can be used to incorporate the constraint conditions into the objective function to form the Lagrangian function and reflect the influence of the constraints on the optimal solution during the iteration process. The convergence tolerance refers to a preset small positive value, which can be used to define the threshold for determining whether the algorithm reaches a sufficiently accurate solution. For example, when the norm of the iteration step size or the gradient is less than this tolerance, the algorithm can be considered to have converged (i.e., the convergence condition is met).

[0068] Among them, the quadratic programming sub - problem refers to a simplified optimization problem constructed and solved in each iteration of the sequential quadratic programming algorithm, which can have a quadratic - form objective function and linear - form constraint conditions. By solving this sub - problem, the search direction and step size of the current iteration point can be obtained.

[0069] Among them, the active set method refers to an iterative algorithm for solving quadratic programming problems. It can maintain a set of constraints that are currently considered "active" or binding, and adjust this set in each iteration until a solution that satisfies the KKT conditions is found. The search direction refers to the direction in which the variables should move at the current iteration point. It can be obtained by solving a quadratic programming subproblem and points to the region where the objective function decreases and the constraints are satisfied. The search step size refers to the distance moved along the search direction. It can be determined by a line search method to ensure sufficient decrease of the objective function while satisfying the constraints.

[0070] Among them, the rmijo line search strategy refers to a non-exact line search method that can be used to determine an appropriate search step size to balance ensuring sufficient decrease of the objective function and satisfying the constraints, and avoid slow convergence caused by too small a step size.

[0071] Among them, the BFGS formula refers to an update formula in a common quasi-Newton method. It can be used to iteratively update the approximation of the Hessian matrix without calculating the exact second-order derivatives, thereby improving the computational efficiency of the algorithm.

[0072] This technical solution provides a systematic optimization method by specifying in detail the iterative process of solving the target coolant flow rate based on the sequential quadratic programming algorithm. First, by initializing key parameters, a foundation is laid for the start of the algorithm. Then, in each iteration, the first-order information of the objective function and constraints is calculated based on the current flow rate value, which is the basis for constructing a local approximation model. Next, a quadratic programming subproblem is constructed and solved. This subproblem transforms the original non-linear problem into an easily solvable quadratic programming problem by making a quadratic approximation of the objective function and a linear approximation of the constraints. Solving this subproblem gives the search direction and step size, indicating how to approach the optimal solution. Subsequently, the flow rate value is updated according to the search direction and step size, and the line search strategy is used to adjust the step size to ensure the stability and effectiveness of the iterative process, guarantee the decrease of the objective function and satisfy the constraints. Finally, it is judged whether the optimal solution is reached by checking the convergence conditions. If not, the Hessian matrix and the Lagrange multiplier estimate value are updated to prepare for the next round of iteration. This iterative process continues until the convergence conditions are met, and the finally obtained flow rate value is the target coolant flow rate of each core region. Applying this detailed optimization and solution process to the optimization problem constructed based on information such as task priority and temperature can effectively determine the optimal coolant flow rate distribution of each core region under constraints such as the total system flow rate, thereby achieving precise control of the aluminum-based PCB microchannel heat dissipation system and coping with dynamic and uneven heat generation characteristics.

[0073] In one embodiment, the implementation of the sequential quadratic programming algorithm can be assisted by using a numerical computation library. For example, when initializing parameters, the initial coolant flow rate value can be set according to experience or preset values, the initial Hessian matrix can be set as the identity matrix, the initial Lagrange multiplier estimate can be set as a zero vector, and the convergence tolerance can be set as a small positive number. When calculating the objective function value, its gradient, constraint conditions, and their Jacobian matrices, symbolic differentiation or numerical differentiation methods can be used for calculation according to the specific function and constraint expressions. When constructing the quadratic programming subproblem, a standard quadratic programming problem form can be formed based on the objective function value, gradient, Hessian matrix approximation, and Jacobian matrix of the current iteration point. When solving the quadratic programming subproblem, an active set method solver implemented in an existing optimization library can be called. When updating the flow rate value and adjusting the search step size, according to the search direction obtained by the solver and combined with the rmijo line search criterion, different step sizes can be iteratively tried to find a step size that meets the conditions. When updating the Hessian matrix and Lagrange multiplier estimate, calculations can be performed according to the BFGS formula using the gradient, flow rate value, and Lagrange multiplier information of the current and previous iteration steps. The entire process is executed in a control loop until the convergence condition is met.

[0074] Through the above detailed sequential quadratic programming iteration steps, a technical solution for solving the target coolant flow rate is provided, which is specific, operable, and helps to improve the algorithm performance and convergence. It solves the problem that only stating the use of the algorithm without detailed limitation of its specific implementation steps may lead to difficulties in algorithm implementation and ineffective guarantee of convergence or computational efficiency. This solution can effectively determine the optimal coolant flow rate distribution in each core area considering various factors such as task priority, temperature, and total flow rate constraint, so as to achieve precise control of the aluminum-based PCB microchannel heat dissipation system.

[0075] Preferably, after step A402, the following steps may further be included: A403. Obtain the task content categories of each computing task, query the preset mapping table of task content category and temperature sensitivity level, and obtain the corresponding temperature sensitivity levels of each computing task; A404. According to the temperature sensitivity levels of each computing task, use the preset adjustment rules to adjust the target coolant flow rate in each core area.

[0076] Among them, the task content category refers to the classification identifier of the computing task content. For example, it can be a real-time control task, a data processing task, a background service task, a graphics rendering task, etc. It can be identified by using the task attribute information provided by the operating system or through the task name, process ID, etc.

[0077] Among them, the temperature sensitivity level refers to a quantitative or graded representation that measures the tolerance of a computing task to temperature changes or the requirement for temperature stability. For example, it can be divided into high sensitivity, medium sensitivity, low sensitivity, etc., and can be represented by preset enumerated values or numerical ranges.

[0078] Among them, the mapping table between task content categories and temperature sensitivity levels refers to a data structure that stores the association relationship between different task content categories and their corresponding temperature sensitivity levels, and can be stored and queried in the form of a lookup table, hash table, or database, etc.

[0079] Among them, the preset adjustment rule refers to the logic or algorithm for correcting the preliminary target coolant flow rate according to the temperature sensitivity level of the task, and can be implemented in ways such as based on a lookup table, piecewise function, scale factor, or fuzzy control.

[0080] Based on the preliminary target coolant flow rates of each core area obtained by solving through an optimization algorithm, this solution further introduces the task temperature sensitivity as an adjustment factor. First, by obtaining the task content categories of the various computing tasks currently being executed on each core area and querying the mapping relationship between the preset task content categories and temperature sensitivity levels, the system can identify the specific tolerance of each task to temperature fluctuations or the stability requirements. This process transforms the abstract task content categories into temperature sensitivity information that can be used for heat dissipation control decisions. Subsequently, based on the temperature sensitivity levels of the various computing tasks obtained, the system combines the preset adjustment rules to correct the preliminary target coolant flow rates of each core area calculated by the previous optimization algorithm. For example, for a core area carrying a high-temperature sensitivity task, even if the preliminary optimization result has considered its priority and current temperature, the adjustment rule may further increase the coolant flow rate allocated to this area to provide stronger heat dissipation capacity and ensure that the temperature is maintained within a stricter range. On the contrary, for a core area carrying a low-temperature sensitivity task, the adjustment rule may appropriately reduce the allocated flow rate without affecting the task stability, so as to allocate the saved coolant resources to areas that need it more, or reduce the total system flow rate requirement and optimize energy consumption.

[0081] Through this secondary adjustment based on task temperature sensitivity, the final coolant flow rate allocation scheme becomes more refined, can more accurately match the actual heat dissipation requirements of different tasks, and in particular can give priority to ensuring the operating stability of temperature-sensitive key tasks. This adjustment mechanism combined with the optimization algorithm based on priority and temperature forms a more comprehensive and intelligent heat dissipation resource allocation strategy, overcoming the limitation that only relying on priority and temperature for optimization may ignore the inherent temperature characteristics of tasks.

[0082] For example, in a specific embodiment, the system can obtain a list of tasks currently running on each core processor and their task content categories from the operating system. The preset mapping table between task content categories and temperature sensitivity levels can be stored as a configuration file or a database table. For example, real-time control tasks are mapped to high sensitivity, data analysis tasks are mapped to medium sensitivity, and background log tasks are mapped to low sensitivity. The system determines the task content category currently mainly carried by each core area based on the task allocation core processor information, and queries the mapping table to obtain the corresponding temperature sensitivity level. The preset adjustment rules can be defined as follows: for the core area where high-sensitivity tasks are located, increase the preliminary target coolant flow by a first preset percentage (e.g., 10%); for the area where medium-sensitivity tasks are located, keep the preliminary flow unchanged or make fine adjustments; for the area where low-sensitivity tasks are located, reduce the preliminary flow by a second preset percentage (e.g., 5%), but ensure that it is not lower than the minimum flow limit. The system applies these adjustment factors to the preliminary target coolant flow value output by the optimization algorithm to obtain the target coolant flow for each core area finally used to control the flow control unit.

[0083] Reference Figure 2 , this application provides an aluminum-based PCB microchannel heat dissipation control system for controlling the aluminum-based PCB microchannel heat dissipation system of a computing device with multiple core processors. The aluminum-based PCB microchannel heat dissipation system has multiple core areas and multiple flow control units. Each of the core processors is respectively arranged at each of the core areas, and each of the flow control units is respectively used to adjust the coolant flow of each of the core areas; the system includes: An information acquisition module 1, configured to obtain task scheduling information and temperature information of each core area; the task scheduling information includes the task priorities of various computing tasks and the corresponding task allocation core processors (the specific process can refer to step A1 in the previous text); A mapping module 2, configured to determine the correspondence between various computing tasks and each core area according to the task allocation core processors of various computing tasks (the specific process can refer to step A2 in the previous text); A weight determination module 3, configured to obtain the heat dissipation resource allocation weights of various computing tasks according to the task priorities of various computing tasks (the specific process can refer to step A3 in the previous text); An optimization calculation module 4, configured to use a resource allocation algorithm to solve the target coolant flow of each core area with the goal of minimizing the weighted average temperature of each core area and minimizing the total coolant flow based on the temperature information of each core area, the heat dissipation resource allocation weights of various computing tasks, the correspondence between various computing tasks and each core area, and the total available coolant flow of the system (the specific process can refer to step A4 in the previous text); The control execution module 5 is used to control each flow control unit to adjust the coolant flow rate of each core area according to the target coolant flow rate of each core area (for the specific process, please refer to step A5 in the previous text).

[0084] The above are only embodiments of the present application and are not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various changes and modifications. 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. An aluminum-based PCB microchannel heat dissipation control method for controlling an aluminum-based PCB microchannel heat dissipation system of a computing device with multiple core processors. The aluminum-based PCB microchannel heat dissipation system has multiple core areas and multiple flow control units. Each of the core processors is respectively disposed at each of the core areas, and each of the flow control units is respectively used to adjust the coolant flow rate of each of the core areas; characterized in that, The steps of the method include: A1. Obtain task scheduling information and temperature information of each core area; the task scheduling information includes the task priorities of various computing tasks and the corresponding task-assigned core processors; A2. Determine the corresponding relationship between each computing task and each core area according to the task-assigned core processors of various computing tasks; A3. Obtain the heat dissipation resource allocation weights of various computing tasks according to the task priorities of various computing tasks; A4. Based on the temperature information of each core area, the heat dissipation resource allocation weights of various computing tasks, the corresponding relationship between various computing tasks and each core area, and the total available coolant flow of the system, with the goal of minimizing the weighted average temperature of each core area and minimizing the total coolant flow, use a resource allocation algorithm to solve the target coolant flow of each core area; A5. Control each flow control unit to adjust the coolant flow of each core area according to the target coolant flow of each core area.

2. The aluminum-based PCB microchannel heat dissipation control method according to claim 1, characterized in that Step A3 includes: A301. Query the task priority and heat dissipation resource allocation weight mapping table according to the task priorities of various computing tasks to obtain the heat dissipation resource allocation weights of various computing tasks.

3. The aluminum-based PCB microchannel heat dissipation control method according to claim 2, wherein, The task scheduling information further includes the task types and task volume information of various computing tasks; After step A301, it includes: A302. Determine the energy consumption levels of various computing tasks according to the task types and task volume information of various computing tasks; A303. Adjust the heat dissipation resource allocation weights of various computing tasks according to the energy consumption levels of various computing tasks to obtain the final heat dissipation resource allocation weights.

4. A method for controlling heat dissipation of a microchannel of an aluminum-based PCB according to claim 3, characterized in that, Step A302 includes: Query the task type and processor frequency impact factor mapping table according to the task types of various computing tasks to obtain the corresponding frequency impact factors; Estimate the processor frequency increase value during the execution of various computing tasks according to the task volume information and frequency impact factors of various computing tasks; Calculate the energy consumption levels of various computing tasks according to the processor frequency increase value and task volume information.

5. A method for controlling heat dissipation of an aluminum-based PCB microchannel according to claim 3, characterized in that, Step A303 includes: Successively take each core processor as the target core, and obtain the set of all core processors within the neighborhood centered on the target core as the neighborhood core set; Calculate the thermal impact factor of each neighborhood core on the target core according to the preset thermal conductivity and the distance between each neighborhood core and the target core within the neighborhood core set; among them, the thermal impact factor of the target core on itself is 1, and the thermal impact factor of other neighborhood cores on the target core is proportional to the thermal conductivity and inversely proportional to the distance; Calculate the sum of the products of the thermal impact factors of all neighborhood cores and the corresponding energy consumption levels of computing tasks to adjust the heat dissipation resource allocation weight of the target core to obtain the final heat dissipation resource allocation weight.

6. The aluminum-based PCB microchannel heat dissipation control method according to claim 1, characterized in that The resource allocation algorithm is a linear programming algorithm, a quadratic programming algorithm or a model predictive control algorithm.

7. A method for controlling the heat dissipation of an aluminum-based PCB microchannel according to claim 1, characterized in that, Step A4 includes: A401. Construct an optimization objective function, where the optimization objective function includes minimizing the weighted average temperature of each core region and minimizing the total coolant flow rate; among them, the weight of each core region is determined by the heat dissipation resource allocation weight of each computing task and the corresponding relationship between each computing task and each core region, and the weighted average temperature is obtained by multiplying the temperature information of each core region by the corresponding weight, summing them up, and then dividing by the total weight; A402. Adopt an optimization algorithm based on sequential quadratic programming, with the coolant flow rate of each core region as the optimization variable, and solve the optimization objective function under the constraints to obtain the target coolant flow rate of each core region; the constraints include the upper and lower limits of the coolant flow rate of each core region and the total available coolant flow rate of the system.

8. A method for controlling heat dissipation of an aluminum-based PCB microchannel according to claim 7, characterized in that Step A402 includes: B1. Initialize the parameters of the sequential quadratic programming algorithm. The initialized parameters include the initial coolant flow rate value of each core region, the Hessian matrix, the Lagrange multiplier estimate value, and the convergence tolerance; B2. Based on the coolant flow rate values of each core region at the current iteration step, calculate the optimization objective function value and its gradient, and calculate the constraints and their Jacobian matrix; B3. Construct a quadratic programming sub-problem. The quadratic programming sub-problem takes the coolant flow rate values of each core region at the current iteration step as the center, and based on the calculated optimization objective function value and its gradient and the Lagrange multiplier estimate value, performs a quadratic approximation on the optimization objective function and a linear approximation on the constraints; among them, the quadratic approximation is approximated by the Hessian matrix, and the linear approximation is approximated by the Jacobian matrix; B4. Use the active set method to solve the quadratic programming sub-problem to obtain the search direction and search step size of the coolant flow rate values of each core region; B5. According to the search direction and search step size of the coolant flow rate values of each core region, update the coolant flow rate values of each core region, and adjust the search step size according to the rmijo line search strategy to ensure that the objective function value decreases and the constraints are satisfied; B6. Determine whether the convergence condition determined by the convergence tolerance is satisfied. If satisfied, use the latest coolant flow rate values of each core region as the target coolant flow rate of each core region. Otherwise, update the Hessian matrix and the Lagrange multiplier estimate value using the BFGS formula, and return to step B2.

9. A method for controlling heat dissipation of an aluminum-based PCB microchannel according to claim 7, characterized in that, After step A402, it further includes: A403. Obtain the task content category of each computing task, query the preset mapping table of task content category and temperature sensitivity level, and obtain the temperature sensitivity level corresponding to each computing task; A404. According to the temperature sensitivity level of each computing task, use the preset adjustment rule to adjust the target coolant flow rate of each core region.

10. An aluminum-based PCB microchannel heat dissipation control system is used to control the aluminum-based PCB microchannel heat dissipation system of a computing device with multiple core processors. The aluminum-based PCB microchannel heat dissipation system has multiple core regions and multiple flow control units. Each of the core processors is respectively disposed at each of the core regions, and each of the flow control units is respectively used to adjust the coolant flow rate of each of the core regions; characterized in that, The system includes: An information collection module, which is used to obtain task scheduling information and the temperature information of each core region; the task scheduling information includes the task priority of each computing task and the corresponding task-assigned core processor; A mapping module, which is used to determine the corresponding relationship between each computing task and each core region according to the task-assigned core processor of each computing task; A weight determination module, configured to obtain the heat dissipation resource allocation weights for each computing task according to the task priorities of each computing task; An optimization calculation module, configured to use a resource allocation algorithm to solve the target coolant flow rates of each core area with the goal of minimizing the weighted average temperature of each core area and minimizing the total coolant flow rate, based on the temperature information of each core area, the heat dissipation resource allocation weights for each computing task, the correspondence between each computing task and each core area, and the total available coolant flow rate of the system; A control execution module, configured to control each flow control unit to adjust the coolant flow rate of each core area according to the target coolant flow rate of each core area.

Images