High-performance computing system

Abstract

The invention provides a high-performance computing system, and relates to the field of high-performance computing, the system comprises a task distribution node and a plurality of computing nodes, the task distribution node and each computing node have corresponding hardware architecture and software system; wherein the task allocation node is used for realizing the following steps of: acquiring a processing priority of each calculation task to obtain a processing priority list; determining a current processing task; obtaining a matching degree between the current processing task and each computing node; determining the computing node with the highest matching degree as a target computing node; sending the current processing task to the target computing node, deleting the current processing task, and skipping to'determining the computing task with the highest priority in the processing priority list as the current processing task '; until the processing priority list is empty. According to the method, factors considered by the finally obtained matching degree are more comprehensive, so that the obtained matching degree is more objective and more accurate.

Description

Background Art

[0002] Although existing high-performance computing systems provide more advanced scheduling functions, the overall scheduling strategy may still not be able to meet the needs of all complex and diverse high-performance computing scenarios. For example, for some tasks with real-time requirements and frequent dynamic adjustment of resource requirements, the existing sorting and scheduling methods based on priority and general resource requirements may not achieve optimal allocation, resulting in extended response time for some tasks or the overall computing efficiency not being able to reach the optimal state in certain situations. Summary of the invention

[0003] In response to the above technical problems, the present application provides a high performance computing system that at least partially solves the problems existing in the prior art.

[0004] In a first aspect of the present application, a high-performance computing system is provided, the system comprising a task allocation node and a plurality of computing nodes, the task allocation node and each computing node having a corresponding hardware architecture and software system; wherein the task allocation node is used to implement the following steps: Obtaining a processing priority of each computing task among a plurality of computing tasks to obtain a processing priority list; wherein the processing priority is related to an urgency value, a required amount of computing resources, an importance value, a task dependency, and an estimated completion time of the computing task; and the processing priority list is sorted from high to low according to the processing priority of each computing task; Determine the computing task with the highest priority in the processing priority list as the current processing task; Obtaining the matching degree between the current processing task and each computing node in the plurality of computing nodes; wherein the matching degree is related to the priority of the current processing task and the node attributes of each computing node; the node attributes of the computing node are related to the current remaining memory amount, remaining computing resources amount, historical task completion efficiency and node stability of the computing node; Determine the computing node with the highest matching degree with the current processing task as the target computing node; Send the current processing task to the target computing node, delete the current processing task, and update the node properties of the target computing node; and jump to "determine the computing task with the highest priority in the processing priority list as the current processing task"; until the above processing priority list is empty.

[0005] Optionally, the priority P of the i-th computing task i Meet the following characteristics: , Among them, E i is the urgency value of the i-th computing task; R i is the amount of computing resources required for the i-th computing task; Ii is the importance value of the i-th computing task; D i is the task dependency of the i-th computing task; T i is the estimated completion time corresponding to the i-th computing task; w 1 , w 2 , w 3 , w 4 , w 5 is the priority weight coefficient, and w 1 +w 2 +w 3 +w 4 +w 5 =1.

[0006] Optionally, the node attribute of the jth computation node is S j =(M j , C j , F j , S j ); where M j is the current remaining memory of the jth computing node; C j is the remaining amount of computing resources of the jth computing node; F j is the historical task completion efficiency of the jth computing node; S j is the node stability of the jth computing node.

[0007] Optionally, the matching degree S between the i-th computing task and the j-th computing node ij Meet the following characteristics: ， Among them, RM i is the amount of memory required for the i-th computing task; RC i is the amount of computing resources required for the i-th computing task; F max M is the maximum value of the historical task completion efficiency of the computing node; total is the total memory of the computing node; C total is the total amount of computing resources of the computing node; α, β, γ, δ are matching weight coefficients, and α+β+γ+δ=1.

[0008] Optionally, the hardware architecture includes: X86 architecture and ARM architecture.

[0009] Optionally, the software system includes: centos, ubuntu, windows and parallel file system.

[0010] Optionally, the system also includes a cluster layer, a tool layer and an application layer.

[0011] Optionally, a parallel file system is used to store the data.

[0012] Optionally, the system further comprises a visualization module, wherein the visualization module is used to realize visualization of the CAE result file.

[0013] Optionally, a PBS scheduling system is employed in the calculation process.

[0014] This application has at least the following beneficial effects: The high-performance computing system provided by the present application first obtains the processing priority of each computing task in multiple computing tasks. The processing priority is related to the urgency value of the computing task, the amount of computing resources required, the importance value, the task dependency and the estimated completion time, and then the computing task with the highest priority in the processing priority list is determined as the current processing task. After that, the matching degree between the current processing task and each computing node in a number of computing nodes is obtained; wherein the above matching degree is related to the priority of the current processing task and the node attributes of each computing node; the node attributes of the above computing nodes are related to the current memory remaining amount, computing resource remaining amount, historical task completion efficiency and node stability of the computing node. And the computing node with the highest matching degree with the current processing task is determined as the target computing node. Here, the highest matching degree means that the computing node is most suitable for the current processing task, and it is determined as the target computing node. Finally, the current processing task is sent to the target computing node, the current processing task is deleted, and the node attributes of the target computing node are updated; and jump to "determine the computing task with the highest priority in the processing priority list as the current processing task"; until the above processing priority list is empty. By comprehensively considering the various features of each computing task and obtaining the features of multiple computing nodes, matching the corresponding dimensions is performed. The final matching degree takes more comprehensive factors into consideration, making the matching degree more objective and accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A structural block diagram of a high performance computing system provided in an embodiment of the present application; Figure 2 A system architecture diagram of a high-performance computing system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of this application.

[0018] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0019] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein may be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present application, it should be understood by those skilled in the art that an aspect described herein may be implemented independently of any other aspect, and two or more of these aspects may be combined in various ways. For example, any number of aspects described herein may be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein may be used to implement this device and / or practice this method.

[0020] Please refer to Figure 1 As shown, an embodiment of the present application provides a high-performance computing system 100, the system comprising: a task allocation node 110 and a plurality of computing nodes 120, the task allocation node and each computing node having a corresponding hardware architecture and software system; wherein the task allocation node is used to implement the following steps: S100, obtaining the processing priority of each computing task among a number of computing tasks to obtain a processing priority list; wherein the above processing priority is related to the urgency value of the computing task, the amount of computing resources required, the importance value, the task dependency and the estimated completion time; the above processing priority list is sorted from high to low according to the processing priority of each computing task.

[0021] Specifically, the priority of the i-th computing task is Pi Meet the following characteristics: , Among them, E i is the urgency value of the i-th computing task; R i is the amount of computing resources required for the i-th computing task; I i is the importance value of the i-th computing task; D i is the task dependency of the i-th computing task; T i is the estimated completion time corresponding to the i-th computing task; w 1 , w 2 , w 3 , w 4 , w 5 is the priority weight coefficient, and w 1 +w 2 +w 3 +w 4 +w 5 =1.

[0022] E i Indicates the urgency of the task, with a value range of 0-1, 1 being the most urgent; 1 is the priority weight coefficient of urgency, w 1 E i Indicates the contribution of urgency to task priority. 2 (1 / R i ) means that the less resources are needed, the greater the improvement in priority. i The value range is 0-1. The higher the importance, the greater the contribution to the priority. i The value range of is 0-1. The higher the dependency, the greater the impact on the priority. The shorter the estimated completion time, the greater the improvement in priority. The denominator plays a normalization role to avoid the deviation of priority calculation caused by excessive or too small eigenvalues, so that the calculation result is within a relatively reasonable range. The other two trigonometric functions are nonlinear adjustment factors. Among them, With E i From 0 to 1, From 0 to 0, but through this formula, the impact of urgency on priority will show nonlinear changes. When the urgency is high, the improvement of priority is more significant. With I i From 0 to 1, Changing from 1 to -1, this formula makes the impact of importance on priority also show nonlinear changes. When the importance is higher, the improvement of priority is more prominent.

[0023] S200: Determine the computing task with the highest priority in the processing priority list as the current processing task.

[0024] Specifically, the highest priority indicates that the computing task should be processed first, and therefore, is determined as the current processing task.

[0025] S300, obtaining the matching degree between the current processing task and each computing node in a number of computing nodes; wherein the matching degree is related to the priority of the current processing task and the node attributes of each computing node; the node attributes of the computing node are related to the current remaining memory of the computing node, the remaining computing resources, the historical task completion efficiency and the node stability.

[0026] Specifically, the node attribute of the jth computing node is S j =(M j , C j , F j , S j ); where M j is the current remaining memory of the jth computing node; C j is the remaining amount of computing resources of the jth computing node; F j is the historical task completion efficiency of the jth computing node; S j is the node stability of the jth computing node, ranging from 0 to 1, with 1 representing the most stable. The matching degree S between the i-th computing task and the j-th computing node ij Meet the following characteristics:

[0027] ， Among them, RM i is the amount of memory required for the i-th computing task; RC i is the amount of computing resources required for the i-th computing task; F max M is the maximum value of the historical task completion efficiency of the computing node; total is the total memory of the computing node; C total is the total amount of computing resources of the computing node; α, β, γ, δ are matching weight coefficients, and α+β+γ+δ=1.

[0028] Here, F j is the historical task completion efficiency of the j-th computing node, which can be calculated comprehensively through indicators such as the number of tasks completed by the node in the past period of time and the average completion time, reflecting the speed and ability of the node to process tasks.

[0029] S jis the node stability of the jth computing node, with a value range of 0-1, 1 being the most stable. Stability can be measured by factors such as the node failure rate and the number of interruptions. Stable nodes are more suitable for assigning tasks. max It is the maximum value of the historical task completion efficiency of the computing node.

[0030] These matching weight coefficients are used to adjust the relative importance of various factors when calculating fitness, and can be flexibly adjusted according to actual business scenarios and needs. Indicates the ratio of the current remaining memory of computing node j to the memory required by computing task i. The larger the ratio, the more the node's memory can meet the task's requirements. is a node stability S j A related nonlinear adjustment factor. When the node stability is moderate, it will have an additional effect on the fitness, while when the stability is too low or too high, the improvement is relatively small. It represents the ratio of the current remaining computing resources of computing node j to the computing resources required by computing task i. The larger the ratio, the more the computing resources of the node can meet the task requirements. is related to the node's historical task completion efficiency F j Related nonlinear adjustment factors. When the efficiency of historical task completion is at a low level, there will be a certain additional improvement, but when the efficiency is too high, the improvement effect will be reduced, so as to avoid over-reliance on a single efficient node. The historical task completion efficiency of node j is normalized to between 0 and 1 to facilitate comparison with other nodes. is the ratio of the remaining memory of node j A related nonlinear adjustment factor. It takes into account the impact of the remaining ratio of node memory on fitness. When the remaining ratio of memory is moderate, there will be an additional improvement in fitness. S j It directly reflects the stability of the node. The higher the stability, the greater the contribution to fitness. is the remaining ratio of computing resources to node j The related nonlinear adjustment factor takes into account the impact of the remaining ratio of computing resources on fitness. When the remaining ratio of computing resources is moderate, there will be an additional improvement in fitness.

[0031] The above formula takes multiple factors into consideration and uses nonlinear adjustment factors to more comprehensively and carefully evaluate the suitability of computing nodes for tasks.

[0032] S400: Determine the computing node with the highest matching degree with the current processing task as the target computing node.

[0033] Specifically, if the matching degree is the highest, it means that the computing node is most suitable for the current processing task, and it is determined as the target computing node.

[0034] S500, send the current processing task to the target computing node, delete the current processing task, and update the node properties of the target computing node; and jump to "determine the computing task with the highest priority in the processing priority list as the current processing task"; until the above processing priority list is empty.

[0035] Specifically, the current processing task is sent to the target computing node, and the target computing node is used to process the current task. At this time, the current processing task is deleted, and since the target computing node receives the current processing task, its corresponding memory and computing resources are changed, because this embodiment also updates the node attributes of the target computing node. In addition, since the current processing task has been deleted from the priority list, that is, the current processing task is already being processed, at this time, jump to step S200 to determine the updated computing task with the highest priority as the new current processing task, and repeat the above steps until the processing priority list is empty, that is, each computing task has been assigned.

[0036] The system provided in this embodiment first obtains the processing priority of each computing task in multiple computing tasks. The processing priority is related to the urgency value of the computing task, the amount of computing resources required, the importance value, the task dependency and the estimated completion time. Then, the computing task with the highest priority in the processing priority list is determined as the current processing task. After that, the matching degree between the current processing task and each computing node in a number of computing nodes is obtained; wherein the above matching degree is related to the priority of the current processing task and the node attributes of each computing node; the node attributes of the above computing nodes are related to the current memory remaining amount, computing resource remaining amount, historical task completion efficiency and node stability of the computing node. And the computing node with the highest matching degree with the current processing task is determined as the target computing node. Here, the highest matching degree means that the computing node is most suitable for the current processing task, and it is determined as the target computing node. Finally, the current processing task is sent to the target computing node, the current processing task is deleted, and the node attributes of the target computing node are updated; and jump to "determine the computing task with the highest priority in the processing priority list as the current processing task"; until the above processing priority list is empty. By comprehensively considering the various features of each computing task and obtaining the features of multiple computing nodes, matching the corresponding dimensions is performed. The final matching degree takes more comprehensive factors into consideration, making the matching degree more objective and accurate.

[0037] In an exemplary embodiment of the present application, Figure 2 The figure shows a system architecture diagram of a high performance computing system of the present application, wherein the hardware architecture includes: X86 architecture and ARM architecture. The software system includes: centos, ubuntu, windows and parallel file system.

[0038] Among them, X86 architecture and ARM architecture provide basic hardware facilities for SuperHPC high-performance computing. The system layer is mainly used to install the required operating systems, such as centos, ubuntu, windows and parallel file systems. The above system also includes cluster layer, tool layer and application layer. Among them, the cluster layer is mainly for the construction and deployment of system clusters, cluster monitoring, job scheduling, cluster management, etc. The tool layer is mainly used for the use of software tools, such as parallel tools MPICH, inte100 years together, GUl compiler, and the application layer is mainly used for simulation design, fluid mechanics, etc.

[0039] In an exemplary embodiment of the present application, the above parallel file system is used to store data, wherein NAS can also be used for storage.

[0040] The platform management nodes are divided into Windows and Linux management nodes, which are installed with Windows and Linux respectively, as well as the corresponding versions of PBS Professional and PPAS, namely the SOA layer of PBS Professional, PBS ProfessionalApplication Service components; Access is installed as a service on the login node and the analysis report service is also installed on this node; the management network uses Gigabit network, the 10 network uses Infiniband, the file system supports SAN or NAS, and 10 nodes can be configured with one or more; new and existing hardware resources are managed in a unified manner, and Windows and Linux are scheduled and used through the same portal, but logically become multiple logical groups, and resources between logical groups do not call each other; storage uses NAS or parallel file system.

[0041] In an exemplary embodiment of the present application, the system further includes a visualization module, wherein the visualization module is used to realize visualization of CAE result files.

[0042] Here, with the help of the integrated Result Service, remote visualization of CAE result files is achieved. This is extremely convenient for users such as researchers. They do not need to install complex professional software locally or go to a specific terminal to view the results. Through remote access, they can intuitively see the calculation results, which is convenient for timely analysis and evaluation of calculation results, and speeds up the advancement of scientific research and projects.

[0043] Data optimization can also be performed. The PBS Analytics component collects and deeply analyzes the data generated by PBS Professional. By mining and analyzing a large amount of task execution data, resource usage data, etc., administrators can have a clear insight into key indicators such as the cluster's operating status, resource usage trends, and task execution efficiency. Based on these analysis results, the cluster can be better managed and optimized in a targeted manner, such as adjusting resource allocation strategies, optimizing scheduling algorithms, etc., to continuously improve the performance of the entire high-performance computing platform.

[0044] The web interface provided by PBS Control allows administrators to easily manage all PBS Works components. Whether it is job scheduling, resource allocation adjustment, or configuration management of each component, all operations can be centralized on this unified interface, which greatly simplifies the management process, reduces the difficulty of management for administrators, improves management efficiency, and makes the operation and maintenance of the entire high-performance computing platform more convenient and efficient.

[0045] The optional PBS Access component can provide more flexible access control for different users. You can accurately set the operation permissions for cluster resources, task submission, result viewing, etc. according to the user's role, department, project and other factors. While ensuring system security, it meets the diverse usage needs of different user groups and ensures that the entire high-performance computing environment runs safely and orderly.

[0046] In addition, PBS Works is not only applicable to traditional HPC clusters, but also plays an excellent role in cloud computing environments. In particular, the PBS Works on Azure version can realize automatic management and scheduling of Azure virtual machines, which means that it can adapt well to both traditional local high-performance computing scenarios and emerging cloud computing large-scale scientific computing scenarios, providing stable and efficient resource management and job scheduling services for users with different computing needs, expanding its application scope and applicable scenarios.

[0047] In an exemplary embodiment of the present application, a PBS scheduling system is used in the calculation process.

[0048] This application integrates all computing resources of the site through SuperHPC, implements the resource utilization strategy of the site, and maximizes the utilization efficiency of the site's HPC resources through its powerful resource management and job scheduling functions.

[0049] In an exemplary embodiment of the present application, an electronic device is also provided.

[0050] Those skilled in the art will appreciate that various aspects of the present application may be implemented as a system, method or program product. Therefore, various aspects of the present application may be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to as "circuit", "module" or "system" herein.

[0051] The electronic device according to this embodiment of the present application is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0052] The electronic device is presented in the form of a general-purpose computing device. The components of the electronic device may include, but are not limited to: the at least one processor mentioned above, the at least one storage device mentioned above, and a bus connecting different system components (including storage devices and processors).

[0053] The storage stores program codes, which can be executed by the processor, so that the processor executes the steps described in the above “Exemplary Method” section of this specification according to various exemplary embodiments of the present application.

[0054] The memory may include readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory, and may further include read only memory (ROM).

[0055] The storage may also include a program / utility having a set (at least one) of program modules, such program modules including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0056] The bus may represent one or more of several types of bus structures including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures.

[0057] The electronic device may also communicate with one or more external devices (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may communicate with one or more devices that enable a user to interact with the electronic device, and / or may communicate with any device (e.g., routers, modems, etc.) that enables the electronic device to communicate with one or more other computing devices. This communication may be performed through an input / output (I / O) interface. In addition, the electronic device may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter. As shown in the figure, the network adapter communicates with other modules of the electronic device through a bus. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0058] Through the description of the above implementation methods, it is easy for those skilled in the art to understand that the example implementation methods described here can be implemented by software, or by combining software with necessary hardware. Therefore, the technical solution according to the implementation method of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the implementation method of the present application.

[0059] In an exemplary embodiment of the present application, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present application can also be implemented in the form of a program product, which includes a program code. When the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present application described in the above "Exemplary Method" section of the present specification.

[0060] The program product may adopt any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, a system, device or component of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination thereof. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof.

[0061] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0062] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0063] Program code for performing the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., using an Internet service provider to connect through the Internet).

[0064] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present application, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0065] It should be noted that, although several modules or units of the equipment for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present application, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into being embodied by multiple modules or units.

[0066] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A high performance computing system, characterized in that: The system includes a task allocation node and several computing nodes, wherein the task allocation node and each computing node have corresponding hardware architecture and software system; wherein the task allocation node is used to implement the following steps: Obtaining a processing priority of each computing task among a plurality of computing tasks to obtain a processing priority list; wherein the processing priority is related to an urgency value, a required amount of computing resources, an importance value, a task dependency, and an estimated completion time of the computing task; and the processing priority list is sorted from high to low according to the processing priority of each computing task; Determine the computing task with the highest priority in the processing priority list as the current processing task; Obtaining the matching degree between the current processing task and each computing node in the plurality of computing nodes; wherein the matching degree is related to the priority of the current processing task and the node attribute of each computing node; the node attribute of the computing node is related to the current remaining amount of memory, remaining amount of computing resources, historical task completion efficiency and node stability of the computing node; Determine the computing node with the highest matching degree with the current processing task as the target computing node; Send the current processing task to the target computing node, delete the current processing task, and update the node properties of the target computing node; and jump to "determine the computing task with the highest priority in the processing priority list as the current processing task"; until the processing priority list is empty.

2. The high performance computing system according to claim 1, characterized in that: The priority P of the i-th computing task i Meet the following characteristics: ， Among them, E i is the urgency value of the i-th computing task; R i is the amount of computing resources required for the i-th computing task; I i is the importance value of the i-th computing task; D i is the task dependency of the i-th computing task; T i is the estimated completion time corresponding to the i-th computing task; w1, w2, w3, w4, w5 are priority weight coefficients, and w1+w2+w3+w4+w5=1.

3. The high performance computing system according to claim 2, characterized in that: The node attribute of the jth computing node is S j =(M j , C j , F j , S j ); Among them, M j is the current remaining memory of the jth computing node; C j is the remaining amount of computing resources of the jth computing node; F j is the historical task completion efficiency of the jth computing node; S j is the node stability of the jth computing node.

4. The high performance computing system according to claim 3, characterized in that: The matching degree S between the i-th computing task and the j-th computing node ij Meet the following characteristics: ， Among them, RM i is the amount of memory required for the i-th computing task; RC i is the amount of computing resources required for the i-th computing task; F max M is the maximum value of the historical task completion efficiency of the computing node; total is the total memory of the computing node; C total is the total amount of computing resources of the computing node; α, β, γ, δ are matching weight coefficients, and α+β+γ+δ=1.

5. The high performance computing system according to claim 1, characterized in that: The hardware architecture includes: X86 architecture and ARM architecture.

6. The high performance computing system according to claim 1, characterized in that: The software system includes: centos, ubuntu, windows and parallel file system.

7. The high performance computing system according to claim 1, characterized in that: The system also includes a cluster layer, a tool layer and an application layer.

8. The high performance computing system according to claim 6, characterized in that: The parallel file system is used to store data.

9. The high performance computing system according to claim 1, characterized in that: The system further comprises a visualization module, wherein the visualization module is used to realize visualization of CAE result files.

10. The high performance computing system according to claim 1, characterized in that: The PBS scheduling system is used in the calculation process.

Images