Intelligent efficient server

Abstract

The invention discloses an intelligent efficient server which comprises a server module, an intelligent heat dissipation module and an intelligent resource distribution module. The intelligent heat dissipation module comprises a heat conduction module and a cooling device, and the heat conduction module is used for conducting heat generated in the server module; the cooling device is used for cooling the heat conduction module and adjusting a cooling strategy in real time according to the working load and the temperature of the server module; the intelligent resource allocation module comprises a resource monitoring module, a resource allocation decision module, a priority setting module, a dynamic adjustment module and a resource recovery and optimization module, the intelligent heat dissipation module ensures that the server is always kept in the most suitable working temperature range, and performance reduction or faults caused by overheating are effectively avoided. And the intelligent resource allocation module dynamically adjusts resource allocation, so that system crash or data loss caused by resource insufficiency or overload is avoided, and the stability and reliability of the server are improved.

Description

Technical Field

[0001] The present invention relates to the field of servers, and specifically to an intelligent and efficient server. Background Art

[0002] In the current era of rapid development of information technology, as the core device for data processing and storage, the performance and stability of servers are directly related to the operation quality and user experience of various application services. With the wide application of technologies such as cloud computing, big data, and artificial intelligence, the computing tasks faced by servers are becoming increasingly complex and the data volume is growing explosively, which poses unprecedented challenges to the processing power, energy efficiency management, and resource scheduling of servers. Traditional server architectures often rely on fixed hardware configurations and simple resource allocation strategies, making it difficult to adapt to dynamically changing workloads, resulting in low resource utilization, high energy consumption, and prominent heat dissipation problems. Especially in the environment of high-density deployment, server failures caused by overheating occur frequently, seriously affecting the reliability and operation efficiency of the system.

[0003] Traditional server heat dissipation methods mostly adopt passive heat dissipation or simple fan heat dissipation. This heat dissipation method is unable to cope when facing high-load operation, and cannot effectively control the internal temperature of the server, thus affecting the stability and lifespan of the server. At the same time, the traditional resource allocation mechanism lacks intelligence and often divides resources based on static strategies, unable to dynamically adjust according to actual application requirements and the current state of the server, resulting in uneven resource allocation and serious waste. Especially in the scenario of multi-task concurrency, low-priority tasks may occupy a large amount of resources, affecting the execution of high-priority tasks, thereby reducing the overall service quality and response speed. Summary of the Invention

[0004] The purpose of the present invention is to overcome the deficiencies of the prior art and provide an intelligent and efficient server, including a server module, an intelligent heat dissipation module, and an intelligent resource allocation module;

[0005] The intelligent heat dissipation module includes a heat conduction module, a data processing module, and a cooling device. The heat conduction module is used to conduct the heat generated inside the server module; the cooling device is used to cool the heat conduction module and adjust the cooling strategy in real time according to the working load and temperature of the server module;

[0006] The intelligent resource allocation module includes a resource monitoring module, a resource allocation decision module, a priority setting module, a dynamic adjustment module, a resource recycling and optimization module;

[0007] The described resource monitoring module is used to monitor the resource status data in the server in real time. The resource status data includes the CPU usage rate, memory occupancy rate, network bandwidth occupancy rate, storage read / write speed and occupancy, and feeds back the resource status data to the resource allocation decision module;

[0008] The described resource allocation decision module is used to perform resource allocation according to the resource status data provided by the resource monitoring module, as well as the preset policies and task priorities;

[0009] The priority setting module is used to receive the task priorities set by the user;

[0010] The dynamic adjustment module is used to perform dynamic adjustment of resources according to the changes in server load and new task requests;

[0011] The resource recycling and optimization module is used to recycle the released resources after the task is completed.

[0012] Furthermore, the described resource allocation decision module is used to perform resource allocation according to the resource status data provided by the resource monitoring module, as well as the preset policies and task priorities, including:

[0013] Collect the resource usage data of the server module, including CPU usage rate, memory occupancy, network bandwidth, and disk I / O data, and obtain the task information of the running tasks;

[0014] Obtain the load situation of the server module based on the collected resource usage data, and evaluate the tightness of each resource and the resource requirement degree of the tasks according to the load situation;

[0015] Determine the priority order of each task according to the preset rules and user configuration, combine the load situation and task priorities to obtain an evaluation result, and based on the evaluation result, determine the amount of resources to be allocated to each task to obtain a resource allocation result. According to the resource allocation result, allocate the resources of the server module to the corresponding tasks;

[0016] During the running of the tasks, continuously monitor the resource usage situation and load changes, and perform dynamic adjustment.

[0017] Furthermore, the cooling device is used to cool the heat conduction module and adjust the cooling strategy in real time according to the working load and temperature of the server module, including:

[0018] The cooling device obtains the operating power consumption of the server module according to the resource allocation result. The operating power consumption includes the power consumption change area and peak power consumption within the set time period. The cooling device generates a cooling power consumption control strategy according to the power consumption change range and peak power consumption within the set time period.

[0019] Further, obtaining the load condition of the server module based on the collected resource usage data, and evaluating the tightness of each resource and the resource requirement degree of the task, including:

[0020] The load condition includes the occupation amount of each resource. According to the occupation amount of each resource, the tightness of the corresponding resource is obtained respectively. The tightness is the percentage of the occupation amount of the resource. If the percentage of the occupation amount of the resource is greater than the set percentage threshold, the resource is tight.

[0021] According to the ratio of the occupation amount of the resources required by the task to the total amount of the corresponding item of resources, the resource requirement degree of the task is obtained.

[0022] Further, determining the priority order of each task according to the preset rules and user configuration, and combining the load condition and the priority of the task to obtain the evaluation result, including:

[0023] According to the priority of the task, the task priority sorting is obtained. According to the obtained occupation amount of each resource, the remaining amount of each resource is obtained. According to the task priority sorting, the demand amounts of the various resources of the task are accumulated in turn. When a certain resource accumulates to the corresponding resource remaining amount, the accumulation is stopped, and the current executable task sequence is obtained.

[0024] Further, during the operation of the task, continuously monitoring the resource usage situation and load changes, and performing dynamic adjustment, including:

[0025] If there is unreasonable resource allocation or load fluctuation, dynamic adjustment is performed. The unreasonable resource allocation includes: according to the ratio of the actual occupation amount of each resource of the task to the allocated amount, if the ratio is not less than the set ratio threshold, the resource allocation is reasonable, otherwise, the allocated resources are adjusted.

[0026] Further, the cooling device generates a cooling power consumption control strategy according to the power consumption change range and peak power consumption within the set time period, including: obtaining the corresponding cooling power according to the power consumption change range and peak power consumption within the time period.

[0027] The beneficial effects of the present invention are: improving resource utilization rate: The intelligent resource allocation module can perform dynamic adjustment according to the actual application requirements and the current state of the server, ensuring the efficient and fair utilization of resources. Through the priority setting and dynamic adjustment mechanism, the problems of resource waste and uneven allocation are avoided, and the overall processing capacity and response speed of the server are improved.

[0028] Reduce energy consumption: The intelligent cooling module can adjust the cooling strategy in real time according to the server's workload and temperature, avoiding energy waste caused by overcooling. At the same time, the intelligent resource allocation module reduces unnecessary resource consumption by optimizing resource allocation, further reducing the energy consumption of the server.

[0029] Improve stability and reliability: The intelligent cooling module ensures that the server always maintains the most suitable working temperature range, effectively avoiding performance degradation or failures caused by overheating. The intelligent resource allocation module avoids system crashes or data loss caused by insufficient or overloaded resources by dynamically adjusting resource allocation, improving the stability and reliability of the server. Brief Description of the Drawings

[0030] Figure 1 It is a schematic diagram of the principle of an intelligent and efficient server;

[0031] Figure 2 It is a schematic diagram of the principle of the intelligent cooling module;

[0032] Figure 3 It is a schematic diagram of the principle of the intelligent resource allocation module. Detailed Embodiments

[0033] The technical solutions of the present invention will be further described in detail below with reference to the drawings, but the protection scope of the present invention is not limited to the following.

[0034] The features and performance of the present invention will be further described in detail below with reference to the embodiments.

[0035] As Figure 1 shown, an intelligent and efficient server includes a server module, including an intelligent cooling module and an intelligent resource allocation module;

[0036] As Figure 2 shown, the intelligent cooling module includes a heat conduction module, a data processing module, and a cooling device. The heat conduction module is used to conduct the heat generated inside the server module; the cooling device is used to cool the heat conduction module and adjust the cooling strategy in real time according to the workload and temperature of the server module;

[0037] As Figure 3 shown, the intelligent resource allocation module includes a resource monitoring module, a resource allocation decision module, a priority setting module, a dynamic adjustment module, a resource recycling and optimization module;

[0038] The resource monitoring module is used to monitor the resource status data in the server in real time. The resource status data includes the usage rate of the central processing unit, the memory occupancy rate, the network bandwidth occupancy rate, the storage read and write speed and occupancy, and feeds the resource status data back to the resource allocation decision module;

[0039] The described resource allocation decision module is used to perform resource allocation based on the resource status data provided by the resource monitoring module, as well as preset policies and task priorities;

[0040] The priority setting module is used to receive the task priorities set by the user;

[0041] The dynamic adjustment module is used to perform dynamic adjustment of resources according to changes in server load and new task requests;

[0042] The resource recycling and optimization module is used to recycle the released resources after the task is completed.

[0043] The described resource allocation decision module is used to perform resource allocation based on the resource status data provided by the resource monitoring module, as well as preset policies and task priorities, including:

[0044] Collect the resource usage data of the server module, including CPU usage rate, memory occupancy, network bandwidth, disk I / O data, and obtain the task information of the running tasks;

[0045] Obtain the load situation of the server module based on the collected resource usage data, and evaluate the tightness of each resource and the resource demand degree of the tasks according to the load situation;

[0046] Determine the priority order of each task according to preset rules and user configurations, combine the load situation and task priorities to obtain an evaluation result, based on the evaluation result, determine the amount of resources to be allocated to each task to obtain a resource allocation result, and allocate the resources of the server module to the corresponding tasks according to the resource allocation result;

[0047] During the task execution, continuously monitor the resource usage situation and load changes, and perform dynamic adjustment.

[0048] The cooling device is used to cool the heat conduction module and adjust the cooling strategy in real time according to the working load and temperature of the server module, including:

[0049] The cooling device obtains the operating power consumption of the server module according to the resource allocation result. The operating power consumption includes the power consumption change area and peak power consumption within a set time period. The cooling device generates a cooling power consumption control strategy according to the power consumption change range and peak power consumption within the set time period.

[0050] The obtaining the load situation of the server module based on the collected resource usage data, and evaluating the tightness of each resource and the resource demand degree of the tasks according to the load situation, includes:

[0051] The load conditions therein include the occupancy of various resources. According to the occupancy of each resource, the tightness of the corresponding resource is obtained respectively. The tightness is the percentage of the occupancy of the resource. If the percentage of the occupancy of the resource is greater than the set percentage threshold, the resource is tight.

[0052] According to the ratio of the occupancy of the resources required by the task to the total amount of the corresponding resources, the resource requirement degree of the task is obtained.

[0053] According to the preset rules and user configurations, determine the priority order of each task, and combine the load conditions and the priorities of the tasks to obtain the evaluation results, including:

[0054] According to the priorities of the tasks, obtain the task priority sorting. According to the obtained occupancy of each resource, obtain the remaining amount of each resource. According to the task priority sorting, sequentially accumulate the demand for each resource of the task. When a certain resource accumulates to the corresponding remaining amount of the resource, stop the accumulation to obtain the currently executable task sequence.

[0055] During the operation of the task, continuously monitor the resource usage conditions and load changes, and perform dynamic adjustment, including:

[0056] If there is unreasonable resource allocation or load fluctuation, perform dynamic adjustment. The unreasonable resource allocation includes: according to the ratio of the actual occupancy of each resource of the task to the allocated amount, if the ratio is not less than the set ratio threshold, the resource allocation is reasonable, otherwise, adjust the allocated resources.

[0057] The cooling device generates a cooling power consumption control strategy according to the power consumption change range and the peak power consumption within the set duration, including: obtaining the corresponding cooling power according to the power consumption change range and the peak power consumption within the duration.

[0058] Specifically, the intelligent and efficient server of the present invention specifically includes a server module, and an intelligent heat dissipation module and an intelligent resource allocation module are embedded in the server module. The two work together to jointly optimize the operating state of the server.

[0059] The intelligent cooling module consists of a heat conduction module and a cooling device. The heat conduction module is made of high thermal conductivity materials and is closely attached to the main heat-generating components (such as CPUs, GPUs, etc.) inside the server, effectively conducting the heat generated by these components. The cooling device is responsible for cooling the heat conduction module. Its innovation lies in being able to adjust the cooling strategy in real time according to the workload and temperature of the server module. Specifically, the cooling device is built with intelligent sensors that monitor the internal temperature and workload of the server in real time, analyze and predict the heat generation situation in the next period of time through algorithms, and accordingly adjust the cooling power and methods (such as fan speed, liquid cooling flow rate, etc.) to ensure that the server always maintains within the most suitable working temperature range, effectively avoiding performance degradation or failures caused by overheating.

[0060] The cooling device also dynamically generates a cooling power consumption control strategy based on the resource allocation result, that is, the operating power consumption of the server module (including the power consumption change area and peak power consumption within the set duration). This means that the working efficiency of the cooling system is closely related to the working state of the server, avoiding both the energy waste caused by overcooling and ensuring timely and effective heat dissipation under high load.

[0061] The intelligent resource allocation module consists of a resource monitoring module, a resource allocation decision module, a priority setting module, a dynamic adjustment module, and a resource recycling and optimization module, realizing the comprehensive monitoring and intelligent scheduling of server resources.

[0062] Resource monitoring module: It collects the resource status data inside the server in real time, including key indicators such as the usage rate of the central processing unit (CPU), memory occupancy rate, network bandwidth occupancy rate, storage read and write speed, and occupancy, providing data support for resource allocation decisions.

[0063] Resource allocation decision module: Based on the data provided by the resource monitoring module, combined with preset strategies and task priorities, it conducts resource allocation. This module first collects the resource usage data of the server module, evaluates the tightness of each resource and the resource requirements of tasks; then determines the execution order of tasks according to preset rules and the task priorities configured by users; finally, combines the load situation and task priorities to generate a resource allocation result, ensuring that high-priority tasks obtain sufficient resource support.

[0064] Priority setting module: It allows users to set the priorities of tasks according to actual needs, providing a basis for resource allocation decisions and enhancing the flexibility and user-friendliness of the system.

[0065] Dynamic adjustment module: During the task execution, continuously monitor the resource usage and load changes. Once it is found that the resource allocation is unreasonable or the load fluctuates, immediately conduct dynamic adjustment. For example, when the resources actually used by a certain task far exceed the allocated amount, the system will automatically adjust its resource quota, or reallocate some resources to more needy tasks to ensure the efficient use of resources.

[0066] Resource recycling and optimization module: After the task is completed, promptly recycle the released resources and conduct resource optimization, such as defragmenting the memory, cleaning up useless files, etc., to keep the server resources in good condition and create conditions for the efficient execution of subsequent tasks.

[0067] Specific implementation of resource allocation decision-making

[0068] During the resource allocation process, the resource allocation decision-making module takes the following steps:

[0069] Data collection and analysis: Comprehensively collect key data such as CPU usage rate, memory occupancy, network bandwidth, disk I / O of the server module, as well as information about running tasks, such as task type, required resource amount, etc.

[0070] Load assessment: According to the collected data, analyze the current load situation of the server, and evaluate the tightness of various resources and the resource requirements of tasks. The tightness is measured by the percentage of resource occupancy. If it exceeds the set threshold, the resource is considered tight. The resource requirement degree of a task is determined by the ratio of the required resource amount of the task to the total amount of the corresponding resource.

[0071] Priority sorting and resource allocation: Sort the tasks according to the preset rules and the task priorities set by the user. Then, combined with the load assessment results, allocate reasonable amounts of resources to each task. During the allocation process, the remaining amount of resources needs to be considered to ensure that the resource allocation does not exceed the total server resource limit.

[0072] Dynamic adjustment: During the task execution, continuously monitor the resource usage and load changes, and reasonably adjust the resource allocation. Especially when it is found that the resource allocation is unreasonable (such as the actual resource occupancy of a task far exceeds the allocated amount) or the load fluctuates, it is necessary to make timely adjustments to ensure the efficient and fair use of resources.

[0073] Dynamic generation of cooling strategy

[0074] The cooling device dynamically generates a cooling power consumption control strategy based on the resource allocation result, that is, the operating power consumption of the server module.

[0075] The specific implementation is as follows:

[0076] Power consumption prediction: Based on the resource allocation result, predict the power consumption change range and peak power consumption of the server within the set time period.

[0077] Cooling power determination: Based on the power consumption prediction results, combined with the internal temperature of the server and the performance parameters of the cooling system, determine the required cooling power.

[0078] Strategy generation and execution: According to the determined cooling power, generate specific cooling strategies, such as adjusting the fan speed, liquid cooling flow rate, etc., and execute them through the control system to ensure that the server always remains within the most suitable working temperature range.

[0079] Embodiment 1

[0080] In a large-scale cloud computing data center, the intelligent and efficient server proposed by the present invention is deployed. This server hosts a variety of different types of business applications, including online transaction processing, big data analysis, machine learning training, etc. These applications have different requirements for computing resources, storage resources, and network resources, and often need to run simultaneously, posing extremely high requirements on the server's resource management and heat dissipation capabilities.

[0081] Thermal conduction module design: Heat sinks made of high thermal conductivity materials (such as graphene composites) are installed on the main heat-generating components such as the CPU and GPU of the server. These heat sinks are closely attached to the surface of the heat-generating components and can quickly conduct heat out.

[0082] Cooling device configuration: An advanced liquid cooling system is integrated inside the server as the cooling device. This system includes a coolant circulation pump, a coolant storage tank, a radiator, and intelligent sensors. The intelligent sensors continuously monitor the internal temperature of the server and the working load of each component, and predict the heat generation trend in the next period of time through the built-in algorithm.

[0083] Dynamic cooling strategy: According to the server operation power consumption data provided by the resource allocation module (including the power consumption change range and peak power consumption), the cooling device dynamically adjusts the flow rate of the coolant and the fan speed of the radiator. For example, when it is detected that the server is about to enter a high-load state, the cooling system will increase the flow rate of the coolant and the fan speed in advance to ensure that the server temperature remains within the safe range. Conversely, in the low-load state, the system will correspondingly reduce the cooling power to save energy.

[0084] Resource monitoring: The resource monitoring module continuously collects the CPU usage rate, memory occupancy rate, network bandwidth occupancy rate of the server, as well as the read / write speed and occupancy of the storage system. These data are transmitted to the resource allocation decision module in real time through the high-speed bus.

[0085] Resource Allocation Decision: Based on the collected data, the resource allocation decision module first evaluates the tightness of each current resource. For example, when the CPU usage rate exceeds 80%, the CPU resource is considered tight. At the same time, according to the requirements of tasks, such as a big data analysis task requiring a large amount of memory and CPU resources, while an online transaction processing task pays more attention to network bandwidth and storage I / O performance, the system will allocate reasonable resources to different tasks according to these requirements and preset priority rules.

[0086] Dynamic Adjustment and Priority Setting: Users can set priorities for different business applications through the priority setting module. During the task execution process, the dynamic adjustment module continuously monitors the resource usage. If it is found that a low-priority task occupies too many resources, the system will automatically adjust its resource quota and reallocate some resources to higher-priority tasks. In addition, when a new high-priority task arrives, the system will also dynamically adjust the resource allocation according to the current load situation and the remaining resources to ensure that high-priority tasks can be responded to in a timely manner.

[0087] Resource Recycling and Optimization: After the task is completed, the resource recycling and optimization module automatically recycles the released resources and performs operations such as memory fragmentation sorting and useless file cleaning to ensure that the server resources are in the best state.

[0088] Embodiment 2

[0089] In a high-performance computing (HPC) cluster, multiple intelligent and efficient servers of the present invention are deployed to support large-scale computing tasks such as scientific research and engineering simulation. These tasks usually have extremely high requirements for computing resources and storage resources, and have long running times and large amounts of data, posing severe challenges to the resource management and heat dissipation capabilities of the servers.

[0090] Thermal Conductivity Module Optimization: For the high-performance CPUs and GPUs commonly used in HPC applications, the thermal conductivity module adopts a customized heat dissipation solution, including using materials with higher thermal conductivity coefficients and increasing the heat dissipation area to ensure that heat can be quickly conducted out under high loads.

[0091] Cooling Device Upgrade: The server adopts advanced phase change cooling technology, utilizing the heat absorption and heat release characteristics of the coolant during the phase change process to achieve more efficient heat dissipation. Intelligent sensors continuously monitor the internal temperature of the server and the working status of each component, predict the heat generation trend through machine learning algorithms, and adjust the operating status of the cooling system accordingly.

[0092] Cooling Strategy Optimization: Based on the resource allocation results and the prediction of server operating power consumption, the cooling device dynamically adjusts the cooling strategy. For example, during high-load tasks such as large-scale matrix operations or computational fluid dynamics simulations, the system will increase the cooling power in advance to ensure that the server temperature does not get too high. During task intervals or low-load periods, the system will reduce the cooling power to save energy and reduce noise.

[0093] Implementation of the Intelligent Resource Allocation Module

[0094] Refined Resource Monitoring: The resource monitoring module not only collects the status data of conventional resources such as CPU, memory, and network bandwidth, but also, in light of the characteristics of HPC applications, adds the monitoring of indicators such as GPU utilization rate, storage I / O performance, and the call frequency of specific scientific computing libraries.

[0095] Intelligent Resource Allocation: The resource allocation decision-making module makes refined resource allocations based on the collected data, taking into account the characteristics of HPC tasks and the priorities set by users. For example, for deep learning training tasks that require a large amount of GPU resources, the system will give priority to allocating GPU resources; for simulation tasks that require high-concurrency I / O operations, it will give priority to allocating storage resources and network bandwidth.

[0096] Dynamic Adjustment and Resource Optimization: During task execution, the dynamic adjustment module will adjust the resource allocation in real time according to the changes in server load and task requirements. At the same time, the resource recycling and optimization module automatically recycles the released resources after the task is completed and performs necessary optimization operations such as memory fragmentation defragmentation and cache cleaning to ensure that the server resources are always in the best state.

Claims

1. An intelligent and efficient server, comprising a server module, characterized in that: Including intelligent heat dissipation module and intelligent resource allocation module; The intelligent heat dissipation module includes a heat conduction module, a data processing module and a cooling device. The heat conduction module is used to conduct the heat generated inside the server module; the cooling device is used to cool the heat conduction module and adjust the cooling strategy in real time according to the workload and temperature of the server module; The intelligent resource allocation module includes a resource monitoring module, a resource allocation decision module, a priority setting module, a dynamic adjustment module, and a resource recovery and optimization module; The resource monitoring module is used to monitor the resource status data in the server in real time, including the CPU usage rate, memory occupancy rate, network bandwidth occupancy rate, storage read and write speed and occupancy, and feed back the resource status data to the resource allocation decision module; The resource allocation decision module is used to allocate resources according to the resource status data provided by the resource monitoring module and the preset strategies and task priorities; The priority setting module is used to receive the task priority set by the user; The dynamic adjustment module is used to dynamically adjust resources according to changes in server load and new task requests; The resource recovery and optimization module is used to recover released resources after the task is completed.

2. The intelligent and efficient server according to claim 1, characterized in that: The resource allocation decision module is used to allocate resources according to the resource status data provided by the resource monitoring module and the preset strategy and task priority, including: Collect resource usage data of the server module, including CPU usage, memory usage, network bandwidth, disk I / O data, and obtain task information of running tasks; Obtain the load of the server module based on the collected resource usage data, and evaluate the intensity of each resource and the resource demand of the task based on the load situation; According to the preset rules and user configuration, the priority order of each task is determined, and the evaluation result is obtained by combining the load situation and the priority of the task. Based on the evaluation result, the amount of resources to be allocated to each task is determined to obtain the resource allocation result. According to the resource allocation result, the resources of the server module are allocated to the corresponding tasks; During the task execution, resource usage and load changes are continuously monitored and dynamic adjustments are made.

3. The intelligent and efficient server according to claim 2, characterized in that: The cooling device is used to cool the heat conduction module and adjust the cooling strategy in real time according to the workload and temperature of the server module, including: The cooling device obtains the operating power consumption of the server module according to the resource allocation result. The operating power consumption includes the power consumption variation area and the peak power consumption within the set time period. The cooling device generates a cooling power consumption control strategy according to the power consumption variation range and the peak power consumption within the set time period.

4. The intelligent and efficient server according to claim 2, characterized in that: The method of obtaining the load of the server module based on the collected resource usage data and evaluating the intensity of various resources and the resource demand of the task based on the load condition includes: The load condition includes the occupied amount of each resource. According to the occupied amount of each resource, the tension degree of the corresponding resource is obtained respectively. The tension degree is the percentage of the occupied amount of the resource. If the percentage of the occupied amount of the resource is greater than the set percentage threshold, the resource is in tension. The resource requirement of the task is obtained based on the ratio of the occupied amount of resources required by the task to the total amount of corresponding resources.

5. The intelligent and efficient server according to claim 2, characterized in that: The priority order of each task is determined according to the preset rules and user configuration, and the evaluation result is obtained by combining the load situation and the priority of the task, including: According to the priority of the task, the task priority ranking is obtained. According to the occupancy of each resource, the remaining amount of each resource is obtained. According to the task priority ranking, the demand for each resource of the task is accumulated in turn. When a resource accumulates to the corresponding remaining amount of resources, the accumulation is stopped to obtain the currently executable task sequence.

6. The intelligent and efficient server according to claim 2, characterized in that: During the task running process, the resource usage and load changes are continuously monitored and dynamically adjusted, including: If there is unreasonable resource allocation or load fluctuation, dynamic adjustment is performed. The unreasonable resource allocation includes: based on the ratio of the actual resource occupancy of each task to the allocated amount, if the ratio is not less than the set ratio threshold, the resource allocation is reasonable, otherwise, the allocated resources are adjusted.

7. The intelligent and efficient server according to claim 3, characterized in that: The cooling device generates a cooling power consumption control strategy according to the power consumption variation range and peak power consumption within a set time period, including: obtaining corresponding cooling power according to the power consumption variation range and peak power consumption within a set time period.

Images