Distributed computing server and distributed computing system

Abstract

The present application relates to a distributed computing server and a distributed computing system, including a server main cabinet, which includes at least two first computing modules and a first network switching module; each first computing module is connected to the first network switching module through a network bus; the first network switching module is used to route and forward network data passing through the first network switching module based on the same address, and allocate computing tasks; the network data includes data to be calculated and computing results; the first computing module is used to obtain computing tasks and data to be calculated based on the first network switching module, and execute computing tasks on the data to be calculated to obtain computing results, and send the computing results through the first network switching module, which solves the high CPU performance requirements of existing computing servers, distributes a large amount of data access and forwarding work to each computing module and network switching module, and reduces the burden and power consumption of the CPU.

Description

Technical Field

[0001] The present application relates to the field of server technology, and in particular to a distributed computing server and a distributed computing system. Background Art

[0002] With the rapid development and widespread application of video surveillance technology, video security issues have become increasingly prominent. The industry has been paying more and more attention to video surveillance security, and has introduced various encryption algorithms to encrypt the video collected by the front-end monitoring, and then decrypt it and output it to the back-end server to prevent video information in key places or fields from being obtained by illegal elements. Typical application scenarios include Figure 1 shown.

[0003] Scenario 1: Deploy an encryption and decryption server. The platform sends encryption tasks to the encryption and decryption server. The encryption and decryption server encrypts the front-end IPC plaintext video stream and forwards the encrypted data to the platform. The platform forwards it to the video storage server to store the ciphertext video.

[0004] Scenario 2: Deploy an encryption and decryption server. The platform sends decryption tasks to the encryption and decryption server. The encryption and decryption server decrypts the front-end IPC ciphertext video stream and forwards the decrypted data to the platform. The platform previews the decrypted video.

[0005] The traditional encryption and decryption server architecture is mainly centered on the CPU, and the encryption and decryption chip modules are cascaded through the PCIE bus to perform video encryption and decryption processing, such as Figure 2 As shown in the figure, the CPU pulls in the external data stream (such as video data from the IPC) through the data port, and then forwards it to each encryption and decryption module through the PCIE switch module for encryption and decryption. The encrypted and decrypted video data is then transmitted back to the CPU through the PCIE switch module, and the CPU stores the data. At the same time, the encryption and decryption server must also connect to the platform and forward the encrypted and decrypted data to the platform server. In this way, the CPU load will be very heavy. In order to match the overall performance of all encryption and decryption modules, the CPU needs to be further upgraded and replaced with a higher specification and higher performance CPU, which will increase the power consumption of the entire machine.

[0006] With regard to the problem in related technologies that computing servers have high requirements for CPU performance, no effective solution has been proposed so far. Summary of the invention

[0007] In this embodiment, a distributed computing server and a distributed computing system are provided to solve the problem of high CPU performance requirements of computing servers in related technologies.

[0008] In a first aspect, a distributed computing server is provided in this embodiment, including: a server main cabinet, the server main cabinet including at least two first computing modules and a first network switching module; each of the first computing modules is connected to the first network switching module via a network bus;

[0009] The first network switching module is used to route and forward the network data passing through the first network switching module based on the same address, and to allocate computing tasks; the network data includes data to be calculated and computing results;

[0010] The first computing module is used to obtain the computing task and the data to be calculated based on the first network switching module, and execute the computing task for the data to be calculated to obtain the computing result, and send the computing result through the first network switching module.

[0011] In some embodiments, the first network switching module comprises a three-layer network switching module, and the three-layer network switching module has a processing device embedded therein.

[0012] In some of the embodiments, in an encryption and decryption scenario, the first computing module includes an encryption and decryption module for encrypting or decrypting the data to be calculated.

[0013] In some of the embodiments, in an intelligent analysis scenario, the first computing module includes an intelligent analysis module for performing intelligent analysis on the data to be calculated.

[0014] In some of the embodiments, the server main cabinet further includes: a first expansion board, each of the first computing modules being connected to the first expansion board in a hot-swappable manner;

[0015] The first expansion board is connected to the first network switching module.

[0016] In some of the embodiments, the invention further comprises: at least one server expansion cabinet; the server expansion cabinet is cascade-connected to the server main cabinet;

[0017] The server expansion cabinet includes at least two second computing modules and a second network switching module; each of the second computing modules is connected to the second network switching module via a network bus;

[0018] The second network switching module is used to forward the network data passing through the second network switching module;

[0019] The second computing module is used to obtain the computing task and the data to be calculated based on the first network switching module and the second network switching module; and execute the computing task for the data to be calculated, and send the computing result of the computing task through the second network switching module and the first network switching module.

[0020] In some embodiments, the server expansion cabinet is cascade-connected to the server main cabinet, including:

[0021] The second network switching module is connected to the cascade port of the first network switching module.

[0022] In some embodiments, the server expansion cabinet is cascade-connected to the server main cabinet, and further comprises:

[0023] The second network switching module of the server expansion cabinet of the next level is connected to the cascade port of the second network switching module of the server expansion cabinet of the previous level.

[0024] In some embodiments, the first network switching module is also used to schedule and manage the first computing module and the second computing module.

[0025] In a second aspect, a distributed computing system is provided in this embodiment, including: a platform server, a data source end, and the distributed computing server described in any one of the first aspects;

[0026] The platform server and the data source are respectively connected to the first network switching module of the distributed computing server;

[0027] The first network switching module of the distributed computing server is used to communicate with the platform server based on the same address;

[0028] The first computing module of the distributed computing server is used to obtain the data to be processed from the data source end based on the stream pulling address.

[0029] Compared with the related art, the distributed computing server and distributed computing system provided in this embodiment include a server main cabinet, which includes at least two first computing modules and a first network switching module; each of the first computing modules is connected to the first network switching module through a network bus; the first network switching module is used to route and forward the network data passing through the first network switching module based on the same address, and assign computing tasks; the network data includes data to be calculated and computing results; the first computing module is used to obtain the computing task and the data to be calculated based on the first network switching module, and execute the computing task for the data to be calculated to obtain the computing result, and send the computing result through the first network switching module, which solves the high CPU performance requirements of the existing computing server, distributes a large amount of data access and forwarding work to each computing module and network switching module, and reduces the CPU burden and power consumption.

[0030] Details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0032] Figure 1 It is a structural schematic diagram of a computing system in the prior art;

[0033] Figure 2 This is a schematic diagram of the architecture of an encryption and decryption server in the prior art;

[0034] Figure 3 This is a schematic diagram of the structure of a distributed computing server in one embodiment of the present application;

[0035] Figure 4 It is a structural schematic diagram of another computing system in the prior art;

[0036] Figure 5 This is a schematic diagram of the structure of a distributed computing system in one embodiment of the present application.

[0037] Figure numerals: 100, server main cabinet; 110, first computing module; 120, first network switching module; 130, first expansion board; 200, server expansion cabinet; 210, second computing module; 220, second network switching module; 230, second expansion board; 300, platform server; 400, data source end. DETAILED DESCRIPTION

[0038] In order to more clearly understand the purpose, technical solutions and advantages of the present application, the present application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0039] Unless otherwise defined, the technical terms or scientific terms involved in this application shall have the general meaning understood by people with general skills in the technical field to which this application belongs. The words "one", "a", "the", "these" and the like in this application do not indicate a quantitative limitation, and they may be singular or plural. The terms "include", "comprise", "have" and any variants thereof involved in this application are intended to cover non-exclusive inclusions; for example, a process, method and system, product or device comprising a series of steps or modules (units) is not limited to the listed steps or modules (units), but may include unlisted steps or modules (units), or may include other steps or modules (units) inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like involved in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The "multiple" involved in this application refers to two or more. "And / or" describes the association relationship of associated objects, indicating that there may be three relationships, for example, "A and / or B" may mean: A exists alone, A and B exist at the same time, and B exists alone. Generally, the character " / " indicates that the objects associated with each other are in an "or" relationship. The terms "first", "second", "third", etc. in this application are only used to distinguish similar objects and do not represent a specific ordering of the objects.

[0040] In this embodiment, a distributed computing server is provided. Figure 3 FIG. 1 is a schematic diagram of the structure of the distributed computing server in this embodiment. Figure 3 As shown, the distributed computing server includes: a server main cabinet 100, the server main cabinet 100 includes at least two first computing modules 110 and a first network switching module 120; each first computing module 110 is connected to the first network switching module 120 via a network bus.

[0041] The first network switching module 120 is used to route and forward the network data passing through the first network switching module 120 based on the same address, and allocate computing tasks; the network data includes data to be calculated and computing results.

[0042] The first computing module 110 is used to obtain computing tasks and data to be calculated based on the first network switching module 120 , and execute computing tasks on the data to be calculated to obtain computing results, and send the computing results through the first network switching module 120 .

[0043] Specifically, the first network switching module 120 is used to connect to the platform and communicate with the platform server 300. The first network switching module 120 presents an IP address to the platform, that is, a device, which is convenient for the platform to connect and manage. At this time, the server main cabinet 100 is equivalent to a computing power pool, and each computing module is an independent subsystem for scheduling by the first network module in the main cabinet. If any intelligent computing module is abnormal, its related services or tasks will be scheduled to other computing modules to achieve high reliability of the product.

[0044] In the traditional solution, the CPU and the encryption and decryption module (i.e., the computing module) are interconnected by a PCIE bus, forming a master-slave relationship. In this proposal, the CPU and the computing module communicate through the network switching module in a network manner, forming a distributed multi-master relationship. In the traditional solution, the CPU pulls the video stream externally (i.e., obtains the data to be processed) and forwards the data to be processed to the platform server 300. In this proposal, the intelligent module obtains the data to be processed and forwards the calculation results to the platform server 300 through the first network module. The processing device built into the first network switching module 120 can perform some business management, task distribution and scheduling. The CPU capability is weakened, the cost is reduced, and the power consumption is reduced.

[0045] In the present embodiment, the server main cabinet 100 includes at least two first computing modules 110 and a first network switching module 120; each first computing module 110 is connected to the first network switching module 120 through a network bus; the first network switching module 120 is used to route and forward the network data passing through the first network switching module 120 based on the same address, and allocate computing tasks; the network data includes data to be calculated and computing results; the first computing module 110 is used to obtain computing tasks and data to be calculated based on the first network switching module 120, and perform computing tasks on the data to be calculated to obtain computing results, and send the computing results through the first network switching module 120, which solves the high CPU performance requirements of existing computing servers, distributes a large amount of data access and forwarding work to each computing module network switching module, and reduces the CPU burden and power consumption.

[0046] In some of the embodiments, the first network switching module 120 includes a three-layer network switching module, and the three-layer network switching module has an embedded processing device.

[0047] Specifically, the network switching module is a three-layer network switching module with routing and forwarding functions, docking platform, and presenting an IP address; it also has a built-in processing device, such as a system on chip (SOC) using an ARM core, which is used to manage the chassis of distributed computing servers, manage computing modules, allocate tasks to computing modules, migrate tasks when computing modules are abnormal, and other business management functions.

[0048] In some of the embodiments, in an encryption / decryption scenario, the first computing module 110 includes an encryption / decryption module for encrypting or decrypting the data to be calculated.

[0049] Specifically, each encryption and decryption module executes the video streaming task according to the streaming address, and performs plaintext encryption or ciphertext decryption on the video stream, and transmits the video plaintext data or ciphertext data stream back to the platform through the first network switching module 120.

[0050] With the same encryption and decryption performance, on the one hand, weakening the CPU can reduce the power consumption and cost of the device; on the other hand, when the power consumption of the device is reduced, the encryption and decryption module of the encryption and decryption server can be made more dense, thereby increasing the number of video encryption and decryption paths and the performance upper limit of the encryption and decryption server.

[0051] In some of the embodiments, in an intelligent analysis scenario, the first computing module 110 includes an intelligent analysis module for performing intelligent analysis on the data to be calculated.

[0052] Specifically, each intelligent analysis module performs video streaming tasks according to the streaming address, and performs decoding, intelligent analysis, and encoding on the video stream to generate intelligent data streams, pictures, events, etc., and transmits the generated calculation results back to the platform. In the intelligent analysis scenario, the first network switching module 120 is also used to load software such as the intelligent computing module running program and the algorithm warehouse.

[0053] With the same intelligent analysis performance, weakening the CPU can reduce the power consumption and cost of the device. On the other hand, when the power consumption of the device is reduced, the intelligent analysis module can be made more dense, thereby increasing the number of video intelligent analysis processing paths and the performance ceiling of the server.

[0054] In some of the embodiments, Figure 3 As shown, the server main cabinet 100 further includes: a first expansion board 130 , each first computing module 110 is connected to the first expansion board 130 in a hot-swappable manner; and the first expansion board 130 is connected to the first network switching module 120 .

[0055] In this embodiment, the expansion board is used to achieve flexible and expandable computing modules and easy maintenance of faults.

[0056] In some embodiments, the distributed computing server further includes: at least one server expansion cabinet 200. Figure 3 As shown, the server expansion cabinet 200 is cascade-connected to the server main cabinet 100. The server expansion cabinet 200 includes at least two second computing modules 210 and a second network switching module 220; each second computing module 210 is connected to the second network switching module 220 via a network bus.

[0057] The second network switching module 220 is used to forward the network data passing through the second network switching module 220 .

[0058] The second computing module 210 is used to obtain computing tasks and data to be calculated based on the first network switching module 120 and the second network switching module 220; and to perform computing tasks on the data to be calculated, and to send the computing results of the computing tasks through the second network switching module 220 and the first network switching module 120.

[0059] The second computing module 210 and the first module use the same type of computing module, for example, both are intelligent analysis modules, or both are encryption and decryption modules. In other embodiments, the second computing module 210 and the first module can use computing modules of different types or specifications. The second computing module 210 is also disposed on the second expansion board 230.

[0060] Specifically, the network cascading method provided in this embodiment, for the platform server 300, no matter how many devices are extended and cascaded, still presents one IP address, one device, and a large computing power pool, which is convenient for the platform to connect and manage.

[0061] In traditional technology, for example, in application scenarios where intelligent analysis of large amounts of video data is required, the platform server 300 sends the intelligent analysis task to the intelligent server and also informs the intelligent server of the streaming address. The intelligent server pulls in the video or image stream from the front-end camera according to the streaming address for intelligent analysis, and forwards the intelligent analysis results to the platform through the network. When the number of intelligent analysis paths of a single intelligent server is insufficient, it is often necessary to expand another device; when expanding multiple intelligent server devices, the platform server 300 needs to connect to multiple intelligent servers, and task scheduling is very complicated. Therefore, it is necessary to add another CV management server, such as Figure 4As shown, the CV management server is used to connect to the platform, and the CV management server is used to manage the two intelligent servers to perform scheduling management of intelligent tasks. The complexity and cost of the solution are quite high. The main reason why the above solution is implemented in this way is that for each intelligent server, the current solution architecture is similar to the encryption and decryption server. It is also a master-slave computing architecture centered on the CPU and interconnected by the PCIE bus. In this architecture mode, the devices are independent of each other. When multiple devices are connected to the platform, multiple IP addresses are presented. The platform server 300 is not easy to manage, and it is usually necessary to use the CV management server to manage the intelligent server cluster.

[0062] In this embodiment, there is no need to connect an additional CV management server to manage the intelligent server cluster. By cascading the server expansion cabinet 200 on the basis of the server main cabinet 100, the computing power pool can be expanded. No matter how many devices are cascaded, one IP address is still presented, which is easy to manage. The server expansion cabinet 200 is plug-and-play, which is more convenient to use and the overall solution is low in cost.

[0063] In some embodiments, the server expansion cabinet 200 is cascade-connected to the server main cabinet 100 , including: the second network switching module 220 is connected to the cascade port of the first network switching module 120 .

[0064] Specifically, a cascade port is provided on the first network switching module 120 of the server main cabinet 100, and a first-level expansion cabinet (i.e., a server expansion cabinet 200 directly connected to the server main cabinet 100) can be cascaded through the network cascade port of the server main cabinet 100. If it is necessary to further expand computing resources, other first-level expansion cabinets can be connected through other cascade ports to form a horizontal cascade; another expansion cabinet can also be cascaded through the network cascade port of the first-level expansion cabinet, and further, the second network switching module 220 of the next-level server expansion cabinet 200 is connected to the cascade port of the second network switching module 220 of the previous-level server expansion cabinet 200 to form a vertical cascade.

[0065] In some embodiments, the first network switching module 120 is further used to schedule and manage the first computing module 110 and the second computing module 210. The processing device in the first network switching module 120 is used to schedule and manage the first computing module 110 and the second computing module 210.

[0066] Specifically, each computing module in the server main cabinet 100 and the server expansion cabinet 200 is an independent subsystem, which is scheduled by the three-layer network module in the main cabinet. If any computing module has an abnormality, its related services or tasks will be scheduled to other modules to achieve high reliability of the product.

[0067] In this embodiment, a distributed computing server 300 and a data source terminal 400 are also provided. The platform server 300 and the data source terminal 400 are respectively connected to the first network switching module 120 of the distributed computing server; the first network switching module 120 of the distributed computing server is used to communicate with the platform server 300 based on the same address; the first computing module 110 of the distributed computing server is used to obtain the data to be processed from the data source terminal 400 based on the pull stream address. Among them, the data source terminal 400 can be a video source terminal, such as a network camera.

[0068] In this embodiment, the first network switching module 120 is used to replace the traditional CPU, and all computing modules are interconnected through the network to realize a distributed architecture, and key tasks such as business management, data storage, data computing, data flow access and forwarding are separated and executed in a distributed manner, that is, after the CPU is decentralized, each computing module performs data flow access and forwarding, and the first network switching module 120 performs task allocation and scheduling, module and chassis management, etc. The overall power consumption and cost of the device are reduced.

[0069] The present embodiment is described and illustrated below through a preferred embodiment in an intelligent analysis scenario.

[0070] Distributed computing system distributed computing server, platform server 300 and video source end. The distributed computing server includes a server main cabinet 100 and a server expansion cabinet 200. The server main cabinet 100 includes a plurality of intelligent computing modules and a three-layer network switching module; each intelligent computing module is connected to the three-layer network switching module through a network bus. The server expansion cabinet 200 includes a plurality of intelligent computing modules and a two-layer network switching module; each intelligent computing module is connected to the two-layer network switching module through a network bus; the two-layer network switching module is connected to the cascade port of the three-layer network switching module.

[0071] Among them, the intelligent computing module and the three-layer network switching module are interconnected through the network bus to form a distributed computing system. The three-layer network switching module has the functions of routing and forwarding network data, and has an embedded ARM core with certain CPU management functions; the two-layer network switching module only has the function of data forwarding.

[0072] The three-layer network switching module is used to connect to the platform server 300 and communicate with the platform server 300. The three-layer network switching presents an IP address, that is, a device, to the platform, which is convenient for the platform to connect and manage.

[0073] As mentioned above, the main cabinet in the solution is connected to the platform to provide services such as intelligence, compression, coding, OSD, etc. At the same time, the three-layer network switching module is embedded with an ARM CPU, which is also responsible for the management of the equipment chassis, the management of the intelligent module, the task allocation of the intelligent module, the running program of the intelligent computing module, and the loading of software such as the algorithm warehouse.

[0074] In the solution, the distributed intelligent server all-in-one expansion cabinet is connected to the distributed intelligent all-in-one main cabinet in a network cascade manner to expand more computing resources. After the distributed expansion cabinet is connected to the distributed main cabinet, the intelligent computing module of the expansion cabinet is automatically registered and loaded into the main cabinet, and the three-layer network switching module in the main cabinet is used for unified scheduling and task allocation, realizing convenient and seamless expansion of computing power.

[0075] The main cabinet and expansion cabinet of the distributed intelligent all-in-one machine present an IP address to the platform, so they present a device, or a large computing power pool. At the same time, each intelligent computing module in the main cabinet and the expansion cabinet is an independent subsystem, which is dispatched by the three-layer network module in the main cabinet. If any intelligent computing module is abnormal, its related business or tasks will be dispatched to other intelligent modules to achieve high reliability of the product.

[0076] Similarly, if you need to cascade another computing power expansion cabinet, you can cascade another expansion cabinet through the network cascade port of the first-level expansion cabinet to form a vertical cascade; you can also cascade horizontally through other network cascade ports of the main cabinet to form a larger computing power pool.

[0077] When implementing the business, the main workflow includes:

[0078] Step 1: The distributed intelligent server main cabinet 100 is powered on, the three-layer network switching module performs network configuration, and the intelligent computing module performs network port configuration and mapping.

[0079] Step 2, the distributed intelligent server main cabinet 100 is connected to the platform, and the platform sends intelligent tasks and streaming addresses to the three-layer network switching module of the distributed intelligent server main cabinet 100. The three-layer network switching module assigns intelligent tasks to each intelligent computing module and informs it of the streaming address.

[0080] Step 3: The intelligent computing module executes the video streaming task according to the address, and decodes, intelligently analyzes, and encodes the video stream to generate intelligent data streams, pictures, events, etc., and returns these data streams to the platform.

[0081] Step 4: The distributed intelligent server expansion cabinet 200 is powered on and connected to the main cabinet in a network cascade manner. The intelligent computing module in the expansion cabinet performs network configuration and port mapping, and is mounted in the main cabinet, and is uniformly scheduled and managed by the three-layer network switching module in the main cabinet.

[0082] Step 5: The Layer 3 network switching module in the main cabinet dispatches the intelligent task to the intelligent computing module in the expansion cabinet through the cascade network (Layer 2 network switching module). The intelligent computing module executes the video streaming task according to the address, and decodes, intelligently analyzes, and encodes the video stream to generate intelligent data streams, pictures, events, etc., and returns these data streams to the platform.

[0083] In this preferred embodiment, after the CPU is decentralized, each intelligent computing module is responsible for accessing the data stream, encrypting and decrypting, and forwarding plaintext and ciphertext, and the built-in ARM CPU of the three-layer network switching module is responsible for task allocation and scheduling, module and chassis management, etc. The overall power consumption and cost of the equipment are reduced. At the same time, the intelligent computing modules constitute a distributed cluster. If any module is abnormal, the business will be scheduled to other modules for execution, ensuring the continuity of the equipment business and fully guaranteeing the reliability of the equipment. In addition, through the network cascading method, for the platform server 300, no matter how many devices are expanded and cascaded, it still presents an IP address, a device, and a large computing power pool, which is convenient for the platform to connect and manage, and there is no need to connect an additional CV management server to manage the intelligent server cluster. The overall solution has a low implementation cost, and the network computing power expansion cabinet is plug-and-play, which is more convenient to use.

[0084] It should be understood that the specific embodiments described herein are only used to explain the application, rather than to limit it. Based on the embodiments provided in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the protection scope of this application.

[0085] Obviously, the drawings are only some examples or embodiments of the present application. For ordinary technicians in the field, the present application can also be applied to other similar situations based on these drawings without creative work. In addition, it is understandable that although the work done in this development process may be complicated and lengthy, for ordinary technicians in the field, certain changes in design, manufacturing or production based on the technical content disclosed in this application are only conventional technical means and should not be regarded as insufficient content disclosed in this application.

[0086] The term "embodiment" in this application refers to a specific feature, structure or characteristic described in conjunction with the embodiment that can be included in at least one embodiment of the present application. The appearance of this phrase in various locations in the specification does not necessarily mean the same embodiment, nor does it mean that it is mutually exclusive with other embodiments and is independent or optional. It is clearly or implicitly understood by those of ordinary skill in the art that the embodiments described in this application can be combined with other embodiments without conflict.

[0087] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of patent protection. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the scope of protection of the present application. Therefore, the scope of protection of the present application shall be subject to the attached claims.

Claims

1. A distributed computing server, characterized in that: include: A server main cabinet, wherein the server main cabinet includes at least two first computing modules and a first network switching module; Each of the first computing modules is connected to the first network switching module via a network bus; The first network switching module is used to route and forward the network data passing through the first network switching module based on the same address, and to allocate computing tasks; the network data includes data to be calculated and computing results; The first computing module is used to obtain the computing task and the data to be calculated based on the first network switching module, and perform the computing task on the data to be calculated to obtain the computing result, and send the computing result through the first network switching module; The distributed computing server further comprises: at least one server expansion cabinet; the server expansion cabinet is cascade-connected to the server main cabinet; The server expansion cabinet includes at least two second computing modules and a second network switching module; each of the second computing modules is connected to the second network switching module via a network bus; The second network switching module is used to forward the network data passing through the second network switching module; The second computing module is used to obtain the computing task and the data to be calculated based on the first network switching module and the second network switching module; and execute the computing task for the data to be calculated, and send the computing result of the computing task through the second network switching module and the first network switching module.

2. The distributed computing server according to claim 1, characterized in that: The first network switching module comprises a three-layer network switching module, and the three-layer network switching module is embedded with a processing device.

3. The distributed computing server according to claim 1, characterized in that: In the encryption and decryption scenario, the first computing module includes an encryption and decryption module for encrypting or decrypting the data to be calculated.

4. The distributed computing server according to claim 1, characterized in that: In an intelligent analysis scenario, the first computing module includes an intelligent analysis module for performing intelligent analysis on the data to be calculated.

5. The distributed computing server according to claim 1, characterized in that: The server main cabinet further includes: a first expansion board, each of the first computing modules being connected to the first expansion board in a hot-swappable manner; The first expansion board is connected to the first network switching module.

6. The distributed computing server according to claim 1, characterized in that: The server expansion cabinet is cascade-connected to the server main cabinet, including: The second network switching module is connected to the cascade port of the first network switching module.

7. The distributed computing server according to claim 1, characterized in that: The server expansion cabinet is cascade-connected to the server main cabinet, and further comprises: The second network switching module of the server expansion cabinet of the next level is connected to the cascade port of the second network switching module of the server expansion cabinet of the previous level.

8. The distributed computing server according to claim 1, characterized in that: The first network switching module is also used to schedule and manage the first computing module and the second computing module.

9. A distributed computing system, characterized in that: include: A platform server, a data source terminal, and a distributed computing server as described in any one of claims 1 to 8; The platform server and the data source are respectively connected to the first network switching module of the distributed computing server; The first network switching module of the distributed computing server is used to communicate with the platform server based on the same address; The first computing module of the distributed computing server is used to obtain the data to be processed from the data source end based on the stream pulling address.

Images