A network computing system based on network file sharing

Abstract

This invention discloses a network-based computing network system, comprising: a control node for generating control commands and sending them to agent nodes; agent nodes for generating agent commands and sending them to computing nodes connected to the agent nodes via the network; computing nodes for generating file retrieval instructions matching the agent commands and sending them to a secondary file-sharing disk; a secondary file-sharing disk for extracting a target computing power file matching the file retrieval instructions and sending it to the computing node matching the file retrieval instructions; and computing nodes for caching the target computing power file in memory, performing calculation operations on the target computing power file according to the agent commands, obtaining the calculation results, and clearing the memory. Through the technical solution of this invention, the stability, reliability, and computational efficiency of the computing network system are improved, while the data storage pressure of the computing network system is reduced.

Description

Technical Field

[0001] This invention relates to the field of computing networks, and more particularly to a computing network system based on network file sharing. Background Technology

[0002] In the field of modern computing networks, there exists a typical computing network framework structure. This framework consists of several hardware nodes, with the control node primarily responsible for receiving user instructions and distributing computational tasks to the agent nodes below. Each agent node is responsible for further subdividing the computational task and distributing it to its corresponding number of computational nodes. For example, if a specific computing network contains 8 agent nodes, and each agent node has 8 computational nodes, then the entire computing network has a total of 64 computational nodes. This means that the computational task obtained from the control node needs to be broken down into 64 smaller computational tasks, which are then distributed to the corresponding computational nodes by the 8 agent nodes. After completing its own computational task, the computational node transmits the computation result to the agent node, which then feeds back to the control node. After one round of computation, 64 computationally generated files are produced. Only by integrating these files can the complete computational result of a single computational task be obtained.

[0003] In such a complete computing task process, there are numerous file transfer processes, including the control node sending computing tasks and the computing files and configuration files required by each computing node to the agent node, the agent node sending computing tasks and the corresponding computing files and configuration files to the computing nodes, and the computing nodes reporting their own computing results after completing the tasks. The computing results are transmitted in the form of files written to disk.

[0004] However, if each file generated in the above process is transmitted separately, the computing network would require a large number of file transfer request and response operations. These frequent network transmissions significantly impact the stability and reliability of the computing network system. Furthermore, using traditional data transmission methods leads to excessive disk write operations; for example, after a computing node completes a computation task, the resulting file is first written to disk locally before being sent to the upper-level proxy node. The proxy node, after writing the file to disk, then sends it to the management layer node. This file write-to-disk behavior is extremely time-consuming, severely impacting system efficiency and increasing the data storage pressure on the computing network system. Summary of the Invention

[0005] This invention provides a network-based computing system that can solve the problems of low stability, reliability, and computing efficiency in existing computing systems, as well as the high data storage pressure in these systems.

[0006] According to one aspect of the present invention, a network file-sharing-based computing system is provided, the system comprising: a control node, at least one agent node, at least one computing node, a primary file-sharing disk, and a secondary file-sharing disk;

[0007] Specifically, the control node establishes a network connection with at least one agent node, the agent node establishes a network connection with at least one computing node, the primary file-sharing disk establishes a network connection with the control node and each agent node respectively, and the secondary file-sharing disk establishes a network connection with the agent node and the computing node that has a network connection with each agent node respectively; the secondary file-sharing disk stores a pre-configured computing power file.

[0008] The control node is used to generate control commands in response to the user's activation operation and send them to each agent node;

[0009] The proxy node is configured to generate a proxy command in response to the control command after receiving the control command sent by the control node, and send it to the computing node that has a network connection with the proxy node.

[0010] The computing node is used to generate a file retrieval instruction matching the proxy command and send it to the secondary file sharing disk after receiving the proxy command;

[0011] The secondary file-sharing disk is used to extract the target computing power file that matches the file acquisition instruction after receiving the file acquisition instruction, and send it to the computing node that matches the file acquisition instruction;

[0012] The computing node is also used to cache the target computing power file in memory after receiving the target computing power file, perform calculation operations on the target computing power file according to the agent command, obtain the calculation results, and clear the memory.

[0013] The computing node is further configured to: send the computation results to the secondary file-sharing disk based on preset computation storage rules, and generate computation completion information to send to the proxy node, wherein the computation completion information includes the computation storage rules; the proxy node is further configured to, upon receiving the computation completion information sent by the computing node, generate a file retrieval instruction based on the computation storage rules in the computation completion information and send it to the secondary file-sharing disk; the secondary file-sharing disk is further configured to, upon receiving the file retrieval instruction sent by the proxy node, send the corresponding computation results to the proxy node based on the computation storage rules; the proxy node is further configured to cache the received computation results in memory, perform reprocessing operations on the computation results, obtain proxy results, and clear the memory.

[0014] The proxy node is further configured to: after the proxy result is generated, send the calculation result to the first-level file-shared disk based on the preset proxy storage rules, and generate a proxy completion command to send to the proxy node, wherein the proxy completion command includes the proxy storage rules; and the control node is further configured to: after receiving the proxy completion command sent by the proxy node, obtain the calculation result stored in the first-level file-shared disk based on the proxy storage rules in the proxy completion command and cache it in memory, perform reprocessing operations on the calculation result, obtain the control result, and clear the memory.

[0015] Furthermore, the first-level file shared disk specifically includes: a storage unit, used to store each calculation result to the corresponding directory according to the proxy storage rules matching each calculation result after receiving each calculation result; and a result sending unit, used to send the corresponding calculation result to the control node based on the proxy storage rules after receiving the control command sent by the control node.

[0016] The proxy node specifically includes: a verification unit, used to verify the calculation results sent by the secondary file-sharing disk according to preset verification rules after receiving the calculation results, in order to determine the correctness of the verification results; a result cleaning unit, used to perform data cleaning operations on the calculation results that have passed the correctness verification, to obtain the cleaned calculation results and send them to the calculation counting unit; a calculation counting unit, used to obtain the number of calculation nodes matching the proxy node, and when the number of cleaned calculation results is equal to the number of calculation nodes, to send each calculation result to the result sending unit; and a result sending unit, used to combine the calculation results according to preset rules to obtain the proxy result.

[0017] Furthermore, the control node specifically includes: a proxy verification unit, used to verify the calculation results sent by the first-level file-shared disk according to preset verification rules after obtaining the calculation results, so as to determine the correctness of the verification results; a proxy result cleaning unit, used to perform data cleaning operations on the calculation results that have passed the correctness verification, to obtain the cleaned calculation results and send them to the proxy counting unit; a proxy counting unit, used to obtain the number of proxy nodes matching the control node, and when the number of cleaned calculation results is equal to the number of proxy nodes, send each calculation result to the result sending unit; and a result combination unit, used to combine the calculation results sent by the proxy counting unit according to preset rules to obtain the control result.

[0018] Furthermore, the control node is also used to: send the control result to a display device for display.

[0019] Optionally, both the primary file-sharing disk and the secondary file-sharing disk are disks that support network file-sharing and have storage functions.

[0020] Optionally, the computing network system further includes: a monitoring module, used to monitor the network file sharing function of the primary file sharing disk and the secondary file sharing disk according to a preset monitoring frequency, so as to determine whether the computing network system is operating normally; and a restart module, used to restart the computing network system when the monitoring module determines that the computing network system is not operating normally.

[0021] Furthermore, the restart module specifically includes: a node restart unit, used to perform restart operations on the control node, each agent node, and each computing node; a disk restart unit, used to restart the network file sharing function of the primary file-sharing disk and the secondary file-sharing disk; and a reconnection unit, used to send a primary reconnection command to the control node and agent nodes after sensing the completion of the operations of the node restart unit and the disk restart unit, so that the control node and agent nodes can establish a network connection with the primary file-sharing disk according to the primary reconnection command, and simultaneously send a secondary reconnection command to each agent node and each computing node, so that each agent node and each computing node can establish a network connection with the secondary file-sharing disk according to the secondary reconnection command.

[0022] The technical solution of this invention involves a control node responding to a user's activation operation, generating a control command, and sending it to each proxy node. Upon receiving the control command, each proxy node generates a proxy command and sends it to a computing node with a network connection. Upon receiving the proxy command, the computing node generates a file retrieval instruction matching the proxy command and sends it to a secondary file-sharing disk. Upon receiving the file retrieval instruction, the secondary file-sharing disk extracts the target computing power file matching the instruction and sends it to the computing node matching the instruction. Upon receiving the target computing power file, the computing node caches it in memory, performs computation on the file according to the proxy command, obtains the computation result, and clears the memory. This solution addresses the problems of low stability, reliability, and computational efficiency in existing computing network systems, as well as high data storage pressure. It improves the stability, reliability, and computational efficiency of the computing network system while reducing its data storage pressure.

[0023] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram of the structure of a network computing system based on network file sharing according to Embodiment 1 of the present invention;

[0026] Figure 2 This is a schematic diagram of a computing network system based on network file sharing according to Embodiment 2 of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] Example 1

[0030] Figure 1 This invention provides a computing network system based on network file sharing, which can be applied to the construction of computing network systems using network file sharing.

[0031] like Figure 1 As shown, the system includes: a control node 110, at least one agent node 120, at least one computing node 130, a primary file-sharing disk 140, and a secondary file-sharing disk 150.

[0032] Specifically, the control node 110 establishes a network connection with at least one agent node 120, the agent node 120 establishes a network connection with at least one computing node 130, the primary file-sharing disk 140 establishes a network connection with the control node 110 and each agent node 120, and the secondary file-sharing disk 150 establishes a network connection with the agent node 120 and each computing node 130 that has a network connection with each agent node 120; the secondary file-sharing disk 150 stores a pre-configured computing power file.

[0033] For example, the pre-configured computing power files may include: node performance parameter files, task allocation record files, computing power resource statistics files, computing node status files, specific algorithm computing power requirement files, historical computing power usage trend files, computing power adjustment strategy files, and computing power resource distribution map files, etc. Those skilled in the art should understand that the number and types of computing power files are determined by the specific computing power tasks of the computing power system. The computing power files stored in the secondary file-sharing disks corresponding to each node can be the same or different; this embodiment does not impose any restrictions on this. Furthermore, due to the inherent requirements of the computing nodes themselves, the computing power files applied to the current computing node should be pre-configured manually in a secondary directory related to the computing power task, and each computing node corresponding to the same secondary file-sharing disk should be pre-configured with one and only one secondary directory matching the specific computing node. When a computing node receives a corresponding computing command, it will directly retrieve the file in the secondary directory matching the current computing node from the secondary file-sharing disk. Since the configuration method of computing power files in the storage device is a mature existing technology, this embodiment will not elaborate on it here.

[0034] The control node 110 is used to generate control commands in response to the user's activation operation and send them to each agent node 120;

[0035] The proxy node 120 is configured to generate a proxy command in response to the control command after receiving the control command sent by the control node 110, and send it to the computing node 130 which has a network connection with the proxy node 120.

[0036] The computing node 130 is used to generate a file retrieval instruction matching the proxy command after receiving the proxy command and send it to the secondary file sharing disk 150.

[0037] The secondary file-sharing disk 150 is used to extract the target computing power file that matches the file acquisition instruction after receiving the file acquisition instruction, and send it to the computing node 130 that matches the file acquisition instruction;

[0038] The computing node 130 is also used to cache the target computing power file in memory after receiving the target computing power file, perform calculation operations on the target computing power file according to the agent command, obtain the calculation results, and clear the memory.

[0039] In a specific implementation scenario of this embodiment, the control node 110 can be a central processing unit or a specific control device with powerful computing and instruction distribution capabilities. When it receives an activation operation performed by a user through an input device or a specific interface, it can respond quickly and accurately and generate a corresponding control command. This control command is then accurately sent to each agent node 120 via a data transmission channel. Furthermore, the agent node 120, acting as an intermediate bridge connecting the control node 110 and the computing node 130, can be, for example, a server or processing unit with certain data processing and transmission capabilities. Upon successfully receiving a control command sent by the control node 110, it can immediately respond sensitively to the control command and generate an agent command. This agent command can be a specific task allocation instruction or operation instruction, and then this agent command is sent via a network link to the computing nodes 130 that have a stable network connection with the agent node 120. Furthermore, the computing node 130 can be multiple computing units distributed in the computing network, such as high-performance computers or dedicated computing devices. Upon receiving a proxy command, the system can generate a file retrieval instruction that matches the proxy command, such as an instruction to retrieve data files or resources in a specific second-level directory of the second-level file-sharing disk 150, and then send it to the second-level file-sharing disk 150. Furthermore, the second-level file-sharing disk 150 can be a centralized or distributed storage area, such as a large-capacity disk array or cloud storage. When a file retrieval instruction is received, the system can accurately extract the target computing power file matching the instruction from its large amount of stored data. This target computing power file can be various computing-related configuration files, data files, etc., and then send it to the computing node 130 matching the file retrieval instruction. Furthermore, the computing node 130 also has the ability to cache the target computing power file in its own memory after receiving it; this memory can be a high-speed random access memory. It can also perform complex computational operations on the target computing power file according to the proxy command, ultimately obtaining the computation result through a series of calculations. Simultaneously, to release resources, it will clear the memory to prepare for the next round of computational tasks.

[0040] Optionally, the computing node 130 is further configured to: send the calculation results to the secondary file-sharing disk 150 based on preset calculation storage rules, and generate calculation completion information and send it to the proxy node 120, wherein the calculation completion information includes the calculation storage rules; the proxy node 120 is further configured to, upon receiving the calculation completion information sent by the computing node 130, generate a file retrieval instruction based on the calculation storage rules in the calculation completion information and send it to the secondary file-sharing disk 150; the secondary file-sharing disk 150 is further configured to, upon receiving the file retrieval instruction sent by the proxy node 120, send the corresponding calculation results to the proxy node 120 based on the calculation storage rules; the proxy node 120 is further configured to cache the received calculation results in memory, perform reprocessing operations on the calculation results, obtain proxy results, and clear the memory.

[0041] Specifically, the computing node 130 can also accurately send the processed calculation results to the secondary file-shared disk 150 according to the pre-set calculation storage rules. At the same time, it will also generate calculation completion information containing the calculation storage rules and send it to the agent node 120. The calculation storage rules can be the pre-set storage location of the calculation results uploaded to the secondary file-shared disk. After successfully receiving the computation completion information sent by the computing node 130, the proxy node 120 generates a file retrieval instruction based on the computation storage rules contained in the computation completion information, and then sends the file retrieval instruction to the secondary file-sharing disk 150. Furthermore, in addition to the functions described above, the secondary file-sharing disk 150, upon receiving the file retrieval instruction from the proxy node 120, sends the corresponding computation result to the proxy node 120 according to the computation storage rules. Finally, after caching the received computation result in its own memory, the proxy node 120 performs further reprocessing operations on the computation result, ultimately obtaining the proxy result through a series of processing steps. After completing these operations, it promptly clears the memory to prepare for subsequent tasks.

[0042] Furthermore, the proxy node 120 is also configured to: after the proxy result is generated, send the calculation result to the first-level file-shared disk 140 based on the preset proxy storage rules, and generate a proxy completion command to send to the proxy node 120, wherein the proxy completion command includes the proxy storage rules; and the control node 110 is also configured to: after receiving the proxy completion command sent by the proxy node 120, obtain the calculation result stored in the first-level file-shared disk 140 based on the proxy storage rules in the proxy completion command and cache it in memory, perform reprocessing operations on the calculation result, obtain the control result and clear the memory.

[0043] The proxy storage rule can be a pre-defined storage location for the calculation results uploaded to the primary file-sharing disk.

[0044] In this embodiment, the first-level file sharing disk 140 specifically includes: a storage unit, used to store each calculation result to a corresponding directory according to the proxy storage rules matching each calculation result after receiving each calculation result; and a result sending unit, used to send the corresponding calculation result to the control node 110 based on the proxy storage rules after receiving the control command sent by the control node 110.

[0045] Furthermore, the proxy node 120 specifically includes: a verification unit, used to verify the calculation results sent by the secondary file sharing disk 150 according to preset verification rules after receiving the calculation results, in order to determine the correctness of the verification results. For example, the verification unit can verify whether the calculation results conform to preset format or specifications, and whether there are missing or incorrect data after receiving the calculation results. Optionally, if any abnormality or non-compliance is found, the proxy node 120 may take corresponding measures, such as requesting the calculation node to recalculate or correct errors.

[0046] Furthermore, the result cleaning unit is used to perform data cleaning operations on the calculation results that have passed the correctness verification, obtain the cleaned calculation results, and send them to the calculation counting unit; the calculation counting unit is used to obtain the number of calculation nodes 130 that match the proxy node 120, and when the number of cleaned calculation results is equal to the number of calculation nodes 130, send each calculation result to the result sending unit; the result sending unit is used to combine each calculation result according to a preset rule to obtain the proxy result.

[0047] Specifically, the control node 110 includes a proxy verification unit, which verifies the calculation results sent by the primary file-sharing disk 140 according to preset verification rules after receiving the calculation results, in order to determine the correctness of the verification results. For example, the proxy verification unit can verify whether the calculation results conform to preset formats or specifications and whether there are missing or incorrect data after receiving the calculation results. Optionally, if any abnormality or non-compliance is found, the control node 110 may take corresponding measures, such as requesting the calculation node to recalculate or correct errors.

[0048] Furthermore, the proxy result cleaning unit is used to perform data cleaning operations on the calculation results that have passed the correctness verification, obtain the cleaned calculation results, and send them to the proxy counting unit; the proxy counting unit is used to obtain the number of proxy nodes 120 that match the control node 110, and when the number of cleaned calculation results is equal to the number of proxy nodes 120, it sends each calculation result to the result sending unit; the result combination unit is used to combine the calculation results sent by the proxy counting unit according to a preset rule to obtain the control result.

[0049] Optionally, the control node 110 is further configured to: send the control result to a display device for display.

[0050] It should be noted that both the primary file-sharing disk 140 and the secondary file-sharing disk 150 are disks that support network file sharing and have storage functions. For example, the primary file-sharing disk 140 and the secondary file-sharing disk 150 can be network storage disks that support network file sharing services such as NFS or SSHFS. It should be noted that the specific network file-sharing services supported by each file-sharing disk can be configured by relevant personnel according to actual computing power needs and scenario requirements, and this embodiment does not impose any restrictions on this.

[0051] The technical solution of this invention involves a control node responding to a user's activation operation, generating a control command, and sending it to each proxy node. Upon receiving the control command, each proxy node generates a proxy command and sends it to a computing node with a network connection. Upon receiving the proxy command, the computing node generates a file retrieval instruction matching the proxy command and sends it to a secondary file-sharing disk. Upon receiving the file retrieval instruction, the secondary file-sharing disk extracts the target computing power file matching the instruction and sends it to the computing node matching the instruction. Upon receiving the target computing power file, the computing node caches it in memory, performs computation on the file according to the proxy command, obtains the computation result, and clears the memory. This improves the stability, reliability, and computational efficiency of the computing network system while reducing the data storage pressure on the system.

[0052] Example 2

[0053] Figure 2 This embodiment of the invention provides a computing network system based on network file sharing. This embodiment is applicable to the construction of computing network systems using network file sharing.

[0054] like Figure 2As shown, the network file sharing-based computing system includes: a control node, at least one agent node, at least one computing node, a primary file sharing disk, a secondary file sharing disk, a monitoring module 210, and a restart module 220.

[0055] The monitoring module 210 establishes communication connections with the primary file-sharing disk and the secondary file-sharing disk, and the restart module 220 establishes communication connections with the control node, each agent node, each computing node, the primary file-sharing disk, and the secondary file-sharing disk.

[0056] The monitoring module 210 is used to monitor the network file sharing function of the primary file sharing disk and the secondary file sharing disk according to a preset monitoring frequency, so as to determine whether the computing network system is operating normally.

[0057] The restart module 220 is used to restart the computing network system when the monitoring module determines that the computing network system is not operating normally.

[0058] Furthermore, the restart module 220 specifically includes: a node restart unit, used to perform restart operations on the control node, each agent node, and each computing node; a disk restart unit, used to restart the network file sharing function of the primary file sharing disk and the secondary file sharing disk; and a reconnection unit, used to send a primary reconnection command to the control node and agent nodes after sensing the completion of the operations of the node restart unit and the disk restart unit, so that the control node and agent nodes can establish a network connection with the primary file sharing disk according to the primary reconnection command, and simultaneously send a secondary reconnection command to each agent node and each computing node, so that each agent node and each computing node can establish a network connection with the secondary file sharing disk according to the secondary reconnection command.

[0059] For example, in a specific implementation scenario of this embodiment, the monitoring module 210 executes an operation similar to `sudo systemctl status nfs-server` every 10 seconds to check the status of network file sharing services such as NFS. If it displays "running," it indicates that the service is running normally; otherwise, it means there is an anomaly. If an anomaly occurs, the restart module 220 can attempt to restart the service, and then further monitoring commands are used to determine the service status. If the monitoring module 210 detects an anomaly in the network file sharing service, the node restart unit of the restart module 220 can perform a restart operation on the node corresponding to the abnormal file-sharing disk. After the network file sharing service returns to normal, the reconnection unit of the restart module 220 needs to send a reconnection command to the restarted node and its subordinate nodes. Specifically, this can be done using SSH Pass to send a reconnection command to the subordinate nodes to connect to the network file sharing service, thereby restoring network file sharing operations and achieving disaster recovery.

[0060] The technical solution of this invention involves a control node responding to a user's activation operation, generating a control command, and sending it to each proxy node. Upon receiving the control command, each proxy node generates a proxy command and sends it to a computing node with a network connection. Upon receiving the proxy command, the computing node generates a file retrieval instruction matching the proxy command and sends it to a secondary file-sharing disk. Upon receiving the file retrieval instruction, the secondary file-sharing disk extracts the target computing file matching the instruction and sends it to the computing node matching the instruction. Upon receiving the target computing file, the computing node caches it in memory, performs computation on the target computing file according to the proxy command, obtains the computation result, and clears the memory. A monitoring module monitors the network file-sharing function of both the primary and secondary file-sharing disks at a preset monitoring frequency to determine if the computing network system is operating normally. A restart module restarts the computing network system when the monitoring module determines it is not operating normally. This improves the stability, reliability, and computational efficiency of the computing network system while reducing its data storage pressure.

[0061] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A network file sharing-based algorithm network system, characterized by, The application relates to a file sharing system, which comprises a control node, at least one agent node, at least one computing node, a first file sharing disk and a second file sharing disk. The control node is connected with the at least one agent node, the agent node is connected with the at least one computing node, the first file sharing disk is connected with the control node and the agent node, the second file sharing disk is connected with the agent node and the computing node connected with the agent node, and the second file sharing disk stores a preconfigured computing power file. The control node generates a control command in response to an activation operation of a user and sends the control command to the agent node. The agent node generates an agent command in response to the control command and sends the agent command to the computing node connected with the agent node after receiving the control command sent by the control node. The computing node generates a file acquisition instruction matched with the agent command and sends the file acquisition instruction to the second file sharing disk after receiving the agent command. The second file sharing disk extracts a target computing power file matched with the file acquisition instruction and sends the target computing power file to the computing node matched with the file acquisition instruction after receiving the file acquisition instruction. The computing node caches the target computing power file to the memory, performs a computing operation on the target computing power file according to the agent command, obtains a computing result and clears the memory after receiving the target computing power file. The computing node sends the computing result to the second file sharing disk based on a preset computing storage rule and sends a computing completion information to the agent node, wherein the computing completion information comprises the computing storage rule. The agent node sends a file acquisition instruction to the second file sharing disk based on the computing storage rule in the computing completion information after receiving the computing completion information sent by the computing node. The second file sharing disk sends the corresponding computing result to the agent node based on the computing storage rule after receiving the file acquisition instruction sent by the agent node. The agent node caches the received computing result to the memory, performs a reprocessing operation on the computing result, obtains an agent result and clears the memory. The agent node sends the computing result to the first file sharing disk based on a preset agent storage rule after the agent result is generated, and sends an agent completion command to the control node, wherein the agent completion command comprises the agent storage rule.

2. The system of claim 1, wherein, The control node obtains the computing result stored in the first file sharing disk based on the agent storage rule in the agent completion command and caches the computing result to the memory after receiving the agent completion command sent by the agent node, and performs a reprocessing operation on the computing result, obtains a control result and clears the memory. The first file sharing disk comprises a storage unit for storing the computing result to a corresponding directory according to the agent storage rule matched with the computing result after receiving the computing result.

3. The system of claim 2, wherein, ​ ​ The result sending unit is configured to send the corresponding calculation result to the control node based on the agent storage rule after receiving the control command sent by the control node.

4. The system of claim 1, wherein, The agent node specifically comprises: The verification unit is configured to verify the calculation result according to a preset verification rule after obtaining the calculation result sent by the secondary file sharing disk, to determine the correctness of the calculation result. The result cleaning unit is configured to perform a data cleaning operation on the calculation result that passes the correctness verification, to obtain the cleaned calculation result and send the cleaned calculation result to the calculation counting unit. The calculation counting unit is configured to obtain the number of calculation nodes matched with the agent node, and send the calculation result to the result sending unit when the number of cleaned calculation results is equal to the number of calculation nodes. The result sending unit is configured to combine the calculation results according to a preset rule, to obtain the agent result.

5. The system of claim 2, wherein, The control node specifically comprises: The agent verification unit is configured to verify the calculation result according to a preset verification rule after obtaining the calculation result sent by the primary file sharing disk, to determine the correctness of the calculation result. The agent result cleaning unit is configured to perform a data cleaning operation on the calculation result that passes the correctness verification, to obtain the cleaned calculation result and send the cleaned calculation result to the agent counting unit. The agent counting unit is configured to obtain the number of agent nodes matched with the control node, and send the calculation result to the result sending unit when the number of cleaned calculation results is equal to the number of agent nodes. The result combination unit is configured to combine the calculation results sent by the agent counting unit according to a preset rule, to obtain the control result.

6. The system of claim 5, wherein, The control node is further configured to: Send the control result to a display device for display.

7. The system of claim 1, wherein, The primary file sharing disk and the secondary file sharing disk are both disks supporting a network file sharing function and having a storage function.

8. The system of claim 1, wherein, The computing network system further comprises: The listening module is configured to listen to the network file sharing function of the primary file sharing disk and the secondary file sharing disk according to a preset listening frequency, to determine whether the computing network system is normally running. The restart module is configured to restart the computing network system when the listening module determines that the computing network system is not normally running.

9. The system of claim 8, wherein, The restart module specifically comprises: The node restart unit is configured to perform a restart operation on the control node, the agent nodes, and the calculation nodes. The disk restart unit is configured to perform a restart operation on the network file sharing function of the primary file sharing disk and the secondary file sharing disk. The reconnection unit is configured to send a primary reconnection command to the control node and the agent nodes after sensing that the operation of the node restart unit and the disk restart unit is completed, to enable the control node and the agent nodes to establish a network connection with the primary file sharing disk according to the primary reconnection command, and to send a secondary reconnection command to the agent nodes and the calculation nodes, to enable the agent nodes and the calculation nodes to establish a network connection with the secondary file sharing disk according to the secondary reconnection command.

Images