Computing resource access method, system and computing device

Abstract

The present invention discloses a computing power resource access method, system and computing device, which relates to the field of artificial intelligence technology. The method includes: obtaining resource access information and resource usage status information sent by each computing power providing device; receiving a target computing power resource access request sent by a client via a proxy server, and determining a predetermined number of computing power providing devices with the highest idle computing power based on the resource usage status information of each computing power providing device corresponding to the target computing power resource; evenly distributing multiple target computing power resource access requests received within a recent predetermined time to a predetermined number of computing power providing devices, and determining the target computing power providing device corresponding to the target computing power resource access request of the client; sending the target resource access information of the target computing power providing device to the proxy server, so that the proxy server can access the target computing power resources of the target computing power providing device and provide them to the client. The present invention facilitates access to a variety of computing power resources.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a computing resource access method, a computing resource access system and a computing device. Background Art

[0002] With the rapid development of artificial intelligence, computing resources are becoming more and more important, especially intelligent computing resources. These demands have brought new opportunities to the computing market. Therefore, many companies have invested a lot of money to build various computing resources. However, according to existing technologies, it is impossible to interconnect these computing resources, resulting in users being unable to use computing resources conveniently and low resource utilization. How to connect these computing resources to facilitate user use and reasonably schedule them according to computing power usage is an urgent problem to be solved.

[0003] Therefore, a computing resource access method is needed to solve the problems existing in the above technical solutions. Summary of the invention

[0004] To this end, the present invention provides a computing power resource access method and a computing power resource access system to solve or at least alleviate the above problems.

[0005] According to one aspect of the present invention, a computing power resource access method is provided, which is executed in a registration server to access multiple computing power resources, the registration server is in communication connection with a proxy server and multiple computing power providing devices corresponding to each computing power resource, the proxy server is in communication connection with one or more clients, and the method includes: obtaining resource access information and resource usage status information of the computing power providing device sent by each computing power providing device; receiving a target computing power resource access request sent by the client via the proxy server, and determining multiple computing power providing devices corresponding to the target computing power resource based on the target computing power resource access request; determining the target computing power resource based on the resource usage status information of each computing power providing device corresponding to the target computing power resource; The idle computing power of each computing power providing device corresponding to the source, and determine a predetermined number of computing power providing devices with the highest idle computing power, where the idle computing power of the computing power providing device includes the currently available number of cards or the currently available number of cores of the computing power providing device; evenly distribute multiple target computing power resource access requests received within a recent predetermined time to the predetermined number of computing power providing devices, and determine the target computing power providing device corresponding to the target computing power resource access request of the client; send the target resource access information of the target computing power providing device to the proxy server, so that the proxy server can access the target computing power resources provided by the target computing power providing device based on the target resource access information, and provide the target computing power resources to the client.

[0006] Optionally, in the computing power resource access method according to the present invention, a predetermined number of computing power providing devices with the highest idle computing power are determined, including: sorting each computing power providing device corresponding to the target computing power resource based on the idle computing power, and obtaining information of a predetermined number of computing power providing devices with the highest idle computing power to form a target queue; and evenly distributing multiple target computing power resource access requests received within a recent predetermined time to the predetermined number of computing power providing devices, including: evenly distributing multiple target computing power resource access requests received within a recent predetermined time to the predetermined number of computing power providing devices corresponding to the predetermined number of computing power providing device information in the target queue in sequence.

[0007] Optionally, in the computing power resource access method according to the present invention, each computing power resource corresponds to a message channel; obtaining resource access information and resource usage status information of the computing power providing device sent by each of the computing power providing devices includes: obtaining resource usage status information of the computing power providing device sent in real time by each of the computing power providing devices via the message channel corresponding to the computing power resource of the computing power providing device; wherein the resource usage status information of the computing power providing device includes the idle computing power of the computing power providing device and the access request list of the computing power providing device; the access request list of the computing power providing device includes each access request corresponding to the computing power providing device.

[0008] Optionally, the computing power resource access method according to the present invention further includes: in response to the number of access requests in the access request list corresponding to any computing power providing device information in the target queue exceeding a predetermined threshold, moving the any computing power providing device information to the end of the target queue.

[0009] Optionally, in the computing power resource access method according to the present invention, it also includes: generating a resource access information list based on the resource access information of each of the computing power providing devices, the resource access information including computing power providing device information, the communication protocol of the computing power resources and the access address; determining the multiple computing power providing devices corresponding to the target computing power resource based on the target computing power resource access request, including: obtaining the multiple computing power providing device information corresponding to the target computing power resource access request from the resource access information list to determine the multiple computing power providing devices corresponding to the target computing power resource, wherein the target computing power resource access request includes the target communication protocol.

[0010] Optionally, the computing power resource access method according to the present invention further includes: in response to a query request for the resource access information list sent by the client via the proxy server, returning the resource access information list to the client via the proxy server; wherein the client is suitable for selecting the resource access information of any computing power providing device from the resource access information list, and accessing the computing power resources provided by any computing power providing device based on the resource access information of any computing power providing device via the proxy server.

[0011] Optionally, in the computing power resource access method according to the present invention, the target computing power resource access request includes target communication protocol information; the proxy server is suitable for: determining the target communication protocol based on the target computing power resource access request, and sending the interface corresponding to the target communication protocol to the client, so that the client can access the target computing power resource based on the interface corresponding to the target communication protocol.

[0012] Optionally, in the computing power resource access method according to the present invention, receiving a target computing power resource access request sent by a client via a proxy server includes: receiving a target computing power resource access request sent by the client via a proxy server based on user authentication information; wherein the user authentication information is used to map the user information of the client with the target computing power providing device information, so as to authenticate the target computing power resource access request from the client based on the user authentication information.

[0013] Optionally, in the computing power resource access method according to the present invention, the multiple computing power resources include intelligent computing power resources, super computing power resources and general computing power resources.

[0014] Optionally, in the computing power resource access method according to the present invention, obtaining resource access information and resource usage status information of the computing power providing device sent by each of the computing power providing devices also includes: obtaining resource access information of the computing power providing device sent by each of the computing power providing devices via a message channel corresponding to the computing power resources of the computing power providing device.

[0015] According to one aspect of the present invention, there is provided a computing power resource access system for accessing a variety of computing power resources, the system comprising: a registration server, which is communicatively connected to a plurality of computing power providing devices corresponding to each computing power resource, and is suitable for executing the method as described above; a proxy server, which is communicatively connected to the registration server, and is suitable for communicatively connecting to a plurality of computing power providing devices corresponding to each computing power resource, so as to access the computing power resources provided by each computing power providing device; and one or more clients, which are communicatively connected to the proxy server, and are suitable for sending a target computing power resource access request to the registration server via the proxy server.

[0016] According to one aspect of the present invention, there is provided a computing device, comprising: at least one processor; and a memory storing program instructions, wherein the program instructions are configured to be suitable for execution by the at least one processor, and the program instructions include instructions for executing the computing power resource access method as described above.

[0017] According to one aspect of the present invention, a computer program product is provided, comprising a computer program / instruction, wherein the computer program / instruction implements the method as described above when executed by a processor.

[0018] According to one aspect of the present invention, a readable storage medium storing program instructions is provided. When the program instructions are read and executed by a computing device, the computing device executes the computing power resource access method as described above.

[0019] According to the technical solution of the present invention, a computing power resource access method is provided, in which a registration server obtains resource access information and resource usage status information of computing power providing devices sent by each computing power providing device, receives a target computing power resource access request sent by a client via a proxy server, and determines multiple computing power providing devices corresponding to the target computing power resources, and determines a predetermined number of computing power providing devices with the highest idle computing power based on the resource usage status information of each computing power providing device corresponding to the target computing power resource, and then evenly distributes multiple target computing power resource access requests received within the most recent predetermined time to a predetermined number of computing power providing devices, and determines the target computing power providing device corresponding to the target computing power resource access request of the client, and sends the target resource access information of the target computing power providing device to the proxy server, so that the proxy server accesses the target computing power resources provided by the target computing power providing device based on the target resource access information, and then provides the target computing power resources to the client for use. Based on this, the interconnection of multiple computing power resources can be realized, so that users can access multiple computing power resources in one stop, and can dynamically schedule computing power resources according to computing power usage, and direct access requests to idle computing power providing devices, thereby improving computing power resource utilization. Moreover, users do not need to consider the communication details of the underlying computing resources.

[0020] Furthermore, if the number of access requests in the access request list corresponding to any computing power providing device information in the target queue exceeds a predetermined threshold, the registration server will move the computing power providing device information to the end of the target queue. Based on this, it can further ensure that computing power providing devices with fewer access requests can be used, realize traffic diversion to idle computing power providing devices, and further improve computing power resource utilization.

[0021] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] To achieve the above and related purposes, certain illustrative aspects are described herein in conjunction with the following description and accompanying drawings, which indicate various ways in which the principles disclosed herein can be practiced, and all aspects and their equivalents are intended to fall within the scope of the claimed subject matter. The above and other purposes, features and advantages of the present disclosure will become more apparent by reading the following detailed description in conjunction with the accompanying drawings. Throughout the present disclosure, the same reference numerals generally refer to the same parts or elements.

[0023] Figure 1 A schematic diagram of a computing resource access system 100 provided according to an embodiment of the present invention is shown;

[0024] Figure 2 A schematic diagram of a computing device 200 provided according to an embodiment of the present invention is shown;

[0025] Figure 3 A schematic diagram of a process of accessing computing resources 300 provided according to an embodiment of the present invention is shown;

[0026] Figure 4 A schematic diagram of the communication method between the computing power providing device for providing each computing power resource and the registration server is shown. DETAILED DESCRIPTION

[0027] The exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0028] In response to the problems in the prior art that multiple computing resources cannot be interconnected and cannot be reasonably scheduled according to the computing power usage, the embodiments of the present invention provide a computing power resource access method, which simplifies the computing power resource access process, can access multiple computing power resources in one stop, realize the interconnection of multiple computing power resources, and can dynamically schedule computing power resources according to the computing power usage.

[0029] The computing power resource access method provided according to the embodiment of the present invention can be implemented in a computing power resource access system so that users can access a variety of computing power resources. The computing power resource access system of the present invention is introduced below.

[0030] Figure 1 A schematic diagram of a computing resource access system 100 provided according to an embodiment of the present invention is shown. The computing resource access system 100 according to the present invention facilitates access to a variety of computing resources and can achieve interconnection and intercommunication of a variety of computing resources.

[0031] It should be noted that any computing power provider can provide a computing power resource, and each computing power resource can correspond to multiple computing power providers (multiple computing power providers from multiple computing power providers). In other words, multiple computing power providers can provide the same computing power resource. In addition, different computing power resources can correspond to different communication protocols.

[0032] like Figure 1 As shown, in an embodiment of the present invention, a computing power resource access system 100 includes one or more clients 110, a proxy server 120, and a registration server 130. Among them, the registration server 130 can be connected to the proxy server 120 and multiple computing power providing devices corresponding to each computing power resource, and the proxy server 120 can be connected to one or more clients 110 in a communication manner, for example, through a wired or wireless network connection.

[0033] Among them, each client 110 can send a target computing resource access request to the registration server 130 via the proxy server 120 to request access to the target computing resource.

[0034] In an embodiment of the present invention, the proxy server 120 may also be communicatively connected with a plurality of computing power providing devices corresponding to each computing power resource to access the computing power resources provided by each computing power providing device, and may provide the accessed computing power resources to the client 110.

[0035] In some embodiments, the various computing power resources may include intelligent computing power resources, super computing power resources, and general computing power resources, but the present invention is not limited to the above computing power resource types.

[0036] It should be noted that intelligent computing resources are computing resources based on GPUs, which are currently commonly used in the field of artificial intelligence. Supercomputing resources are computing resources based on CPUs, which are often used in computing-intensive fields such as scientific computing and engineering computing, and are mainly based on parallel computing. General computing resources mainly refer to traditional CPU server resources, especially single servers, and can also refer to resources that provide services such as storage.

[0037] The client 110 may be a terminal device used by a user, which may be a personal computer such as a desktop computer or a laptop computer, or a mobile terminal such as a mobile phone, a tablet computer, a multimedia device, or a smart wearable device, but is not limited thereto. The client 110 may also be an application program resident in the terminal device.

[0038] The computing power providing device can be implemented as any computing device capable of providing computing power resources in the prior art, and the present invention does not limit the specific type of the computing power providing device. For example, the computing power providing device can be implemented as a desktop computer, a laptop computer, a processor chip, a mobile phone, a tablet computer, etc., but is not limited thereto.

[0039] In an embodiment of the present invention, the registration server 130 obtains resource access information and resource usage status information of the computing power providing device sent by each computing power providing device. In addition, the registration server 130 can receive a target computing power resource access request sent by the client 110 via the proxy server 120, and determine multiple computing power providing devices corresponding to the target computing power resource based on the target computing power resource access request.

[0040] Subsequently, the registration server 130 can determine the idle computing power of each computing power providing device corresponding to the target computing power resource based on the resource usage status information of each computing power providing device corresponding to the target computing power resource obtained in advance, and determine a predetermined number of computing power providing devices with the highest idle computing power from the multiple computing power providing devices corresponding to the target computing power resource. After that, the multiple target computing power resource access requests received within the most recent predetermined time are evenly distributed to the predetermined number of computing power providing devices.

[0041] In addition, the registration server 130 can also determine the target computing power providing device corresponding to the target computing power resource access request of the client 110, and send the target resource access information of the target computing power providing device to the proxy server 120, so that the proxy server 120 can access the target computing power resources provided by the target computing power providing device based on the target resource access information, and provide the target computing power resources to the client 110 for the client user to use the target computing power resources.

[0042] It can be seen that the computing power resource access system 100 provided by the embodiment of the present invention can realize the interconnection of various computing power resources, allowing users to access various computing power resources in one stop, and can dynamically schedule computing power resources according to computing power usage, and direct access requests to idle computing power providing devices, thereby improving computing power resource utilization. Moreover, users do not need to consider the communication details of the underlying computing power resources.

[0043] In an embodiment of the present invention, the registration server 130 in the computing power resource access system 100 is adapted to execute the computing power resource access method 300. The computing power resource access method 300 of the present invention will be described in detail below.

[0044] In one embodiment, the registration server 130 of the present invention can be implemented as a computing device described below, so that the computing resource access method 300 of the present invention can be executed in the computing device. In addition, the client 110, the proxy server 120, and each computing power providing device can be implemented as a computing device described below.

[0045] Figure 2 FIG. 2 is a schematic diagram of a computing device 200 provided according to an embodiment of the present invention. Figure 2 As shown, in a basic configuration, the computing device 200 includes at least one processing unit 202 and a system memory 204. According to one aspect, depending on the configuration and type of the computing device, the processing unit 202 can be implemented as a processor. The system memory 204 includes, but is not limited to, volatile storage (e.g., random access memory), non-volatile storage (e.g., read-only memory), flash memory, or any combination of such memories. According to one aspect, the system memory 204 includes an operating system 205.

[0046] According to one aspect, operating system 205 is suitable for controlling the operation of computing device 200, for example. Furthermore, examples are practiced in conjunction with graphics libraries, other operating systems, or any other application programs, and are not limited to any particular application or system. Figure 2 The basic configuration is shown in FIG. 2 by those components within the dashed lines. According to one aspect, the computing device 200 has additional features or functionality. For example, according to one aspect, the computing device 200 includes additional data storage devices (removable and / or non-removable), such as magnetic disks, optical disks, or tapes. Such additional storage Figure 2 2 is illustrated by a removable storage device 209 and a non-removable storage device 210.

[0047] As stated above, according to one aspect, a program module 203 is stored in the system memory 204. According to one aspect, the program module 203 may include one or more application programs, and the present invention does not limit the type of application program, for example, the application program may include: email and contact application programs, word processing application programs, spreadsheet application programs, database application programs, slide show application programs, drawing or computer-aided application programs, web browser application programs, etc.

[0048] In an embodiment according to the present invention, the program module 203 includes multiple program instructions for executing the computing resource access method 300 of the present invention.

[0049] According to one aspect, the examples may be practiced on a circuit comprising discrete electronic components, a packaged or integrated electronic chip containing logic gates, a circuit utilizing a microprocessor, or a single chip containing electronic components or a microprocessor. Figure 2 Each or many components shown in can be integrated into a system on a chip (SOC) on a single integrated circuit to practice examples. According to one aspect, such a SOC device may include one or more processing units, a graphics unit, a communication unit, a system virtualization unit, and various application functions, all of which are integrated (or "burned") into a chip substrate as a single integrated circuit. When operated via SOC, the functions described in this article can be operated via a dedicated logic integrated with other components of the computing device 200 on a single integrated circuit (chip). Other technologies that can perform logical operations (such as AND, OR, and NOT) can also be used to practice embodiments of the present invention, including but not limited to mechanical, optical, fluid, and quantum technologies. In addition, embodiments of the present invention can be practiced in a general-purpose computer or in any other circuit or system.

[0050] According to one aspect, the computing device 200 may also have one or more input devices 212, such as a keyboard, a mouse, a pen, a voice input device, a touch input device, etc. Output devices 214 may also be included, such as a display, a speaker, a printer, etc. The aforementioned devices are examples and other devices may also be used. The computing device 200 may include one or more communication connections 216 that allow communication with other computing devices 218. Examples of suitable communication connections 216 include, but are not limited to: RF transmitters, receivers and / or transceiver circuits; Universal Serial Bus (USB), parallel and / or serial ports.

[0051] The term computer-readable medium as used in this article includes computer storage media. Computer storage media can include volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (e.g., computer-readable instructions, data structures, or program modules). System memory 204, removable storage device 209, and non-removable storage device 210 are all examples of computer storage media (i.e., memory storage). Computer storage media can include random access memory (RAM), read-only memory (ROM), electrically erasable read-only memory (EEPROM), flash memory or other memory technology, CD-ROM, digital versatile disk (DVD) or other optical storage, cassettes, tapes, disk storage or other magnetic storage devices, or any other products that can be used to store information and can be accessed by computing device 200. According to one aspect, any such computer storage medium can be a part of computing device 200. Computer storage media does not include carrier waves or other propagated data signals.

[0052] According to one aspect, communication media is implemented by computer readable instructions, data structures, program modules, or other data in a modulated data signal (e.g., a carrier wave or other transport mechanism), and includes any information delivery media. According to one aspect, the term "modulated data signal" describes a signal that has one or more characteristics set or changed in a manner that encodes information in the signal. By way of example and not limitation, communication media includes wired media such as a wired network or direct-wired connection, and wireless media such as acoustic, radio frequency (RF), infrared, and other wireless media.

[0053] In an embodiment according to the present invention, the computing device 200 is configured to execute the computing power resource access method 300. The computing device 200 includes one or more processors, and one or more readable storage media storing program instructions. When the program instructions are configured to be executed by one or more processors, the computing device executes the computing power resource access method 300 in the embodiment of the present invention, so that the computing device 200 can realize the interconnection of multiple computing power resources by executing the computing power resource access method 300 of the present invention, so that users can access multiple computing power resources in one stop, and can dynamically schedule computing power resources according to computing power usage.

[0054] In some embodiments, the computing device that executes the computing power resource access method 300 in the embodiments of the present invention may be a registration server 130, that is, the computing power resource access method 300 may be executed in the registration server 130 to access a variety of computing power resources, which may include intelligent computing power resources, super computing power resources, and general computing power resources.

[0055] The computing power resource access method 300 in the embodiment of the present invention is described in detail below.

[0056] Figure 3 FIG. 3 is a flow chart of a computing resource access method 300 provided according to an embodiment of the present invention. Figure 3 As shown, the computing power resource access method 300 includes the following steps 310~350.

[0057] Step 310: The registration server 130 obtains resource access information and resource usage status information of the computing power providing device sent by each computing power providing device.

[0058] Step 320: The registration server 130 receives a target computing resource access request (i.e., a request to access the target computing resource) sent by the client 110 (which may be referred to as the "first client") via the proxy server 120, and determines a plurality of computing power providing devices corresponding to the target computing resource based on the target computing resource access request. Here, it can be understood that the target computing resource may be provided by a plurality of computing power providing devices.

[0059] In an embodiment of the present invention, the client 110 inputs the relevant parameter information of the target computing power resource, and sends the relevant parameter information of the target computing power resource to the proxy server 120, so as to send the target computing power resource access request via the proxy server 120 based on the relevant parameter information of the target computing power resource. The target computing power resource access request contains the relevant parameter information of the target computing power resource, so that the registration server 130 can determine that the target computing power resource corresponds to multiple computing power providing devices based on the target computing power resource access request (the relevant parameter information of the target computing power resource). It should be noted that the present invention does not specifically limit the relevant parameter information. In some embodiments, the relevant parameter information may include communication protocol information.

[0060] In some embodiments, the target computing power resources may be any one of intelligent computing power resources, super computing power resources, and general computing power resources.

[0061] Step 330: The registration server 130 determines the idle computing power of each computing power providing device corresponding to the target computing power resource based on the resource usage status information of each computing power providing device corresponding to the target computing power resource, and determines a predetermined number of computing power providing devices with the highest idle computing power from the multiple computing power providing devices corresponding to the target computing power resource.

[0062] In an embodiment of the present invention, the idle computing power of the computing power providing device can be represented by the current number of available cards of the computing power providing device or the number of cores of the computing power providing device. In other words, the idle computing power of the computing power providing device includes the current number of available cards or the current number of available cores of the computing power providing device.

[0063] Step 340: The registration server 130 evenly distributes multiple target computing resource access requests (including the target computing resource access request from the first client) received within the most recent predetermined time to the predetermined number of computing power providing devices, and can determine the target computing resource access request from the client 110 (the first client) The corresponding target computing resource access request is assigned to the computing power providing device. In this way, resource access requests can be allocated to idle computing power providing devices.

[0064] It should be noted that the target computing power providing device is the computing power providing device to which the target computing power resource access request of the client 110 (the first client) is allocated.

[0065] In an embodiment of the present invention, each user of the client 110 can send a target computing resource access request (i.e., an access request to the target computing resource) to the registration server 130 via the proxy server 120. In step 340, the multiple target computing resource access requests can be multiple target computing resource access requests sent by multiple client users via the proxy server 120.

[0066] In some embodiments, in step 340, the most recent scheduled time may be the last minute, that is, the registration server 130 may evenly distribute multiple target computing power resource access requests received within the last minute to the above-mentioned predetermined number of computing power providing devices in sequence, and determine the target computing power providing device corresponding to the target computing power resource access request of the client 110.

[0067] Step 350: The registration server 130 sends the target resource access information of the target computing power providing device to the proxy server 120, so that the proxy server 120 can access the target computing power resources provided by the target computing power providing device based on the target resource access information, and provide the target computing power resources to the client 110 (first client) for the client user to use the target computing power resources.

[0068] Based on this, it is possible to realize the interconnection of various computing resources, allowing users to access various computing resources in one stop, and dynamically schedule computing resources according to computing power usage, directing access requests to idle computing power provision devices, thereby improving computing power resource utilization. Moreover, users do not need to consider the communication details of the underlying computing resources.

[0069] Figure 4 A schematic diagram of the communication method between the computing power providing device for providing each computing power resource and the registration server is shown.

[0070] like Figure 4As shown, in some embodiments, each computing power resource corresponds to a message channel, and each computing power resource can report information to the registration server 130 based on the corresponding message channel. Specifically, the computing power providing device used to provide each computing power resource can communicate with the registration server 130 via the message channel corresponding to the computing power resource. For example, the computing power providing device used to provide each computing power resource can report information to the registration server 130 via the message channel corresponding to the computing power resource. In one embodiment, each computing power providing device can send the resource usage status information of the computing power providing device to the registration server 130 in real time via the message channel corresponding to the computing power resource provided by the computing power providing device. In addition, each computing power providing device can send the resource access information of the computing power providing device to the registration server 130 via the message channel corresponding to the computing power resource provided by the computing power providing device.

[0071] That is, in step 310, the registration server 130 can specifically obtain the resource usage status information of the computing power providing device sent in real time by each computing power providing device via the message channel corresponding to the computing power resources of the computing power providing device. In addition, the registration server 130 can obtain the resource access information of the computing power providing device sent by each computing power providing device via the message channel corresponding to the computing power resources of the computing power providing device.

[0072] In some embodiments, each message channel can be implemented as a kafka middleware or a rabbitmq middleware.

[0073] In some embodiments, the resource usage status information of the computing power providing device may include: the idle computing power of the computing power providing device and the access request list of the computing power providing device. Based on this, the registration server 130 can determine the idle computing power of each computing power providing device corresponding to the target computing power resource based on the resource usage status information of each computing power providing device corresponding to the target computing power resource.

[0074] The access request list of the computing power providing device may include various access requests corresponding to the computing power providing device (ie, access requests from the client 110 to the computing power resources of the computing power providing device).

[0075] In some embodiments, in step 330, after the registration server 130 determines the idle computing power of each computing power providing device corresponding to the target computing power resource based on the resource usage status information of each computing power providing device corresponding to the target computing power resource, it can further determine the predetermined number of computing power providing devices with the highest idle computing power in the following manner: sort the computing power providing devices corresponding to the target computing power resource based on the idle computing power, obtain the information of the predetermined number of computing power providing devices with the highest idle computing power, and form a target queue based on the information of the predetermined number of computing power providing devices with the highest idle computing power. In one embodiment, the predetermined number is N, and N is a positive integer greater than or equal to 2. In other words, the registration server 130 can obtain the information of the N computing power providing devices with the highest idle computing power and form a target queue.

[0076] In this embodiment, in step 340, multiple target computing power resource access requests received within the most recent predetermined time may be evenly distributed in sequence to a predetermined number of computing power providing devices corresponding to a predetermined number of computing power providing device information in the target queue, and the target computing power providing device to which the target computing power resource access request of client 110 (first client) is assigned may be determined.

[0077] In some embodiments, the registration server 130 may also determine whether the number of access requests (i.e., the total number of all access requests to the computing power providing device from each client 110) in the access request list corresponding to any computing power providing device information in the target queue (i.e., the access request list of the computing power providing device corresponding to any computing power providing device information) exceeds a predetermined threshold. If it is determined that the number of access requests in the access request list corresponding to any computing power providing device information in the target queue exceeds the predetermined threshold, the registration server 130 may move any computing power providing device information to the end of the target queue in response to the number of access requests in the access request list corresponding to any computing power providing device information in the target queue exceeding the predetermined threshold. Based on this, it can be further ensured that computing power providing devices with fewer access requests can be used, so as to provide target computing power resources to users through idle computing power providing devices with fewer access requests, thereby achieving traffic diversion to idle computing power providing devices.

[0078] Furthermore, the registration server 130 may determine whether the number of access requests in the access request list corresponding to any computing power providing device information in the target queue exceeds a predetermined threshold within a predetermined time period. If the threshold is exceeded, the registration server 130 may move the information of any computing power providing device to the end of the target queue. The predetermined time period is, for example, 1 day (the same day). Based on this, it can be further ensured that the computing power providing device with fewer access requests can be used, so as to provide the target computing power resources to the user through the computing power providing device with fewer access requests.

[0079] In some embodiments, after obtaining the resource access information of the computing power providing device sent by each computing power providing device, the registration server 130 can also generate a resource access information list based on the resource access information of each computing power providing device. It can be understood that the resource access information list includes the resource access information of each computing power providing device.

[0080] It should be noted that the resource access information of the computing power providing device may include: computing power providing device information, the communication protocol of the computing power resource and the access address of the computing power resource. The computing power providing device information may include, for example, the computing power providing device identification.

[0081] Based on this, in step 320, the specific method in which the registration server 130 determines the multiple computing power providing devices corresponding to the target computing power resource based on the target computing power resource access request is as follows: the registration server 130 can obtain the multiple computing power providing device information corresponding to the target computing power resource access request of the client 110 proxy server 120 (that is, the multiple computing power providing device information adapted to the target computing power resource access request) from the resource access information list, and then, based on the multiple computing power providing device information, it can determine the multiple computing power providing devices corresponding to the target computing power resource (that is, the multiple computing power providing devices that can provide the target computing power resource).

[0082] In some embodiments, the user may input the target communication protocol information on the client 110 to request access to the corresponding target computing power resources. The client 110 may send a target computing power resource access request to the registration server 130 via the proxy server 120 based on the target communication protocol information to request access to the target computing power resources corresponding to the target communication protocol information. In this way, the target computing power resource access request includes the target communication protocol information. In step 320, the registration server 130 may respond to the target computing power resource access request sent by the client 110 via the proxy server 120, and determine the multiple computing power providing devices corresponding to the target computing power resource based on the target communication protocol information in the target computing power resource access request. Specifically, the registration server 130 may obtain the multiple computing power providing device information corresponding to the target communication protocol information from the resource access information list, and determine the multiple computing power providing devices corresponding to the target computing power resource based on the multiple computing power providing device information corresponding to the target communication protocol information.

[0083] Furthermore, in step 330, the registration server 130 may obtain a predetermined number of computing power providing device information with the highest computing power from multiple computing power providing device information corresponding to the target computing power resource access request, and form a target queue based on the predetermined number of computing power providing device information with the highest computing power.

[0084] In one embodiment, the resource access information list includes the following information:

[0085] <reslist>

[0086] <res type="notebook">

[0087] <notebook providercode="00001"> notebook.aaa.com< / notebook>

[0088] <notebook providercode="00002"> notebook.bbb.com< / notebook>

[0089] < / res>

[0090] <res type="ssh">

[0091] <ssh providercode="00001"> ssh.aaa.com-p22< / ssh>

[0092] <ssh providercode="00002"> ssh.bbb.com -p 22< / ssh>

[0093] < / res>

[0094] <res type ="vnc".

[0095] <wnc providercode="00001">xxx.xxx.xx x.xxx -p 5900

[0096] <wnc providercode="00002">yyy.yyy.yy yyyy -p 5900

[0097]

[0098] <res type="restful">

[0099] <restful providercode="00001"> http: / / apiaaa.com / v1< / restful>

[0100] <restful providercode="00002"> http: / / apibbb.com / v1< / restful>

[0101] < / res>

[0102] < / wnc> < / wnc> < / reslist>

[0103] In the above resource access information list, notebook, ssh, vnc, and restful are different communication protocols. That is, in one embodiment, the communication protocols of the computing power resources may include notebook protocol, ssh protocol, vnc protocol, and restful protocol.

[0104] In some embodiments, the client 110 may send a query request for a resource access information list to the registration server 130 via the proxy server 120, and the registration server 130 may return the resource access information list to the client 110 via the proxy server 120 in response to the query request for the resource access information list sent by the client 110 via the proxy server 120.

[0105] Subsequently, the client 110 can select the resource access information of any computing power providing device from the resource access information list, and access the computing power resources provided by any computing power providing device based on the resource access information of any computing power providing device via the proxy server 120. In addition, the proxy server 120 can provide the computing power resources it accesses to the client 110 for the client 110 to use the computing power resources. It should be understood that in this embodiment, the client user selects a determined and unique computing power providing device, so the client 110 can directly access the computing power resources provided by the computing power providing device via the proxy server 120.

[0106] In some embodiments, different communication protocols correspond to different interfaces. When the proxy server 120 receives the target computing resource access request from the client 110, it can determine the target communication protocol corresponding to the target computing resource access request based on the target computing resource access request from the client 110 (including the target communication protocol information), and send the interface corresponding to the target communication protocol to the client 110, so that the client 110 can access and use the target computing resource based on the interface corresponding to the target communication protocol.

[0107] It should be noted that the proxy server 120 can use a plug-in mechanism to manage the client 110's access to computing resources. Specifically, the proxy server 120 can integrate plug-ins corresponding to a variety of communication protocols and interfaces, and can support a variety of different communication protocols and interfaces through multiple plug-ins. In this way, it can be expanded according to actual needs and support dynamic loading. At the same time, it can be customized and developed through overloading interfaces according to the needs of users or computing power providers.

[0108] In some embodiments, the user needs to enter user authentication information when requesting access to the target computing resource at the client 110. The client 110 can obtain the user authentication information entered by the user, and send a target computing resource access request to the registration server 130 via the proxy server 120 based on the user authentication information.

[0109] Accordingly, in step 320 , the registration server 130 may receive a target computing resource access request sent by the client 110 via the proxy server 120 based on user authentication information.

[0110] It should be noted that the user authentication information is used to map the user information of the client 110 with the target computing power providing device information (for example, the account information of the target computing power providing device), so that the registration server 130 can authenticate the target computing power resource access request from the client 110 based on the user authentication information. For example, the registration server 130 can use the Oauth2 mechanism instead of the target computing power providing device to authenticate the target computing power resource access request from the client 110 based on the user authentication information. In this way, it can ensure that the client 110 has legal access to the target computing power resources provided by the target computing power providing device.

[0111] According to the computing power resource access method 300 of the present invention, the registration server obtains the resource access information and resource usage status information of the computing power providing device sent by each computing power providing device, receives the target computing power resource access request sent by the client via the proxy server, and determines the multiple computing power providing devices corresponding to the target computing power resource, and determines the predetermined number of computing power providing devices with the highest idle computing power based on the resource usage status information of each computing power providing device corresponding to the target computing power resource, and then evenly distributes the multiple target computing power resource access requests received within the most recent predetermined time to the predetermined number of computing power providing devices, and determines the target computing power providing device corresponding to the target computing power resource access request of the client, and sends the target resource access information of the target computing power providing device to the proxy server, so that the proxy server accesses the target computing power resource provided by the target computing power providing device based on the target resource access information, and then provides the target computing power resource to the client for use. Based on this, it is possible to realize the interconnection of multiple computing power resources, enable users to access multiple computing power resources in one stop, and dynamically schedule computing power resources according to computing power usage, and direct access requests to idle computing power providing devices, thereby improving computing power resource utilization. Moreover, users do not need to consider the communication details of the underlying computing resources.

[0112] Furthermore, if the number of access requests in the access request list corresponding to any computing power providing device information in the target queue exceeds a predetermined threshold, the registration server will move the computing power providing device information to the end of the target queue. Based on this, it can further ensure that computing power providing devices with fewer access requests can be used, realize traffic diversion to idle computing power providing devices, and further improve computing power resource utilization.

[0113] The various techniques described herein may be implemented in combination with hardware or software, or a combination thereof. Thus, the method and apparatus of the present invention, or some aspects or parts of the method and apparatus of the present invention may be in the form of program codes (i.e., instructions) embedded in a tangible medium, such as a removable hard disk, a USB flash drive, a floppy disk, a CD-ROM, or any other machine-readable storage medium, wherein when the program is loaded into a machine such as a computer and executed by the machine, the machine becomes a device for practicing the present invention.

[0114] In the case where the program code is executed on a programmable computer, the mobile terminal generally includes a processor, a storage medium readable by the processor (including volatile and non-volatile memory and / or storage elements), at least one input device, and at least one output device. The memory is configured to store the program code; the processor is configured to execute the computing resource access method of the present invention according to the instructions in the program code stored in the memory.

[0115] By way of example and not limitation, readable media include readable storage media and communication media. Readable storage media stores information such as computer readable instructions, data structures, program modules or other data. Communication media generally embody computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and include any information transfer medium. Any combination of the above is also included within the scope of readable media.

[0116] In the description provided herein, algorithms and displays are not inherently related to any particular computer, virtual system or other device. Various general purpose systems can also be used together with the examples of the present invention. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the present invention is not directed to any specific programming language either. It should be understood that various programming languages ​​can be utilized to implement the content of the present invention described herein, and the above description of specific languages ​​is for the purpose of disclosing the best mode of the present invention.

[0117] In the description provided herein, a large number of specific details are described. However, it is understood that embodiments of the present invention can be practiced without these specific details. In some instances, well-known methods, structures and techniques are not shown in detail so as not to obscure the understanding of this description.

[0118] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the present invention, various features of the present invention are sometimes grouped together into a single embodiment, figure, or description thereof.

[0119] Those skilled in the art will appreciate that the modules or units or components of the devices in the examples disclosed herein may be arranged in the devices described in the embodiment, or alternatively may be located in one or more devices different from the devices in the examples. The modules in the foregoing examples may be combined into one module or may be divided into multiple submodules.

[0120] Unless otherwise specified, the use of ordinal numbers "first", "second", "third", etc. to describe common objects merely indicates that different instances of similar objects are involved and is not intended to imply that the objects so described must have a given order in time, space, order or in any other manner.

Claims

1. A computing power resource access method is executed in a registration server to access multiple computing power resources, wherein the multiple computing power resources include intelligent computing power resources, super computing power resources, and general computing power resources, each computing power resource corresponds to a message channel, the registration server is connected in communication with a proxy server and multiple computing power providing devices corresponding to each computing power resource, the proxy server is connected in communication with one or more clients, and the method includes: Obtain resource access information and resource usage status information of the computing power providing device sent in real time by each computing power providing device via a message channel corresponding to the computing power resources of the computing power providing device, wherein the resource usage status information of the computing power providing device includes the idle computing power of the computing power providing device and the access request list of the computing power providing device, and the access request list of the computing power providing device includes each access request corresponding to the computing power providing device; Receive a target computing resource access request sent by a client via a proxy server, and determine a plurality of computing power providing devices corresponding to the target computing resource based on the target computing resource access request; Based on the resource usage status information of each computing power providing device corresponding to the target computing power resource, the idle computing power of each computing power providing device corresponding to the target computing power resource is determined, and a predetermined number of computing power providing devices with the highest idle computing power is determined, wherein determining the predetermined number of computing power providing devices with the highest idle computing power includes: sorting the computing power providing devices corresponding to the target computing power resource based on the idle computing power, and obtaining information of the predetermined number of computing power providing devices with the highest idle computing power to form a target queue; the idle computing power of the computing power providing device includes the current number of available cards or the current number of available cores of the computing power providing device; In response to the number of access requests in the access request list corresponding to any computing power providing device information in the target queue exceeding a predetermined threshold, moving the any computing power providing device information to the end of the target queue; Distribute multiple target computing power resource access requests received within the most recent predetermined time evenly to the predetermined number of computing power providing devices corresponding to the predetermined number of computing power providing device information in the target queue, and determine the target computing power providing device corresponding to the target computing power resource access request of the client; The target resource access information of the target computing power providing device is sent to the proxy server, so that the proxy server can access the target computing power resources provided by the target computing power providing device based on the target resource access information, and provide the target computing power resources to the client.

2. The method of claim 1, wherein: Also includes: Generate a resource access information list based on the resource access information of each computing power providing device, wherein the resource access information includes computing power providing device information, a communication protocol and an access address of the computing power resource; Determining multiple computing power providing devices corresponding to the target computing power resource based on the target computing power resource access request includes: From the resource access information list, obtain information of multiple computing power providing devices corresponding to the target computing power resource access request to determine the multiple computing power providing devices corresponding to the target computing power resource, wherein the target computing power resource access request includes a target communication protocol.

3. The method of claim 2, wherein: Also includes: In response to a query request for the resource access information list sent by the client via the proxy server, returning the resource access information list to the client via the proxy server; Among them, the client is suitable for selecting resource access information of any computing power providing device from the resource access information list, and accessing the computing power resources provided by any computing power providing device based on the resource access information of any computing power providing device via the proxy server.

4. The method according to any one of claims 1 to 3, wherein: The target computing resource access request includes target communication protocol information; The proxy server is suitable for: determining the target communication protocol based on the target computing power resource access request, and sending the interface corresponding to the target communication protocol to the client, so that the client can access the target computing power resource based on the interface corresponding to the target communication protocol.

5. The method according to any one of claims 1 to 3, wherein: Receive a target computing resource access request sent by the client via the proxy server, including: Receiving a target computing resource access request sent by the client via a proxy server based on user authentication information; The user authentication information is used to map the user information of the client with the target computing power providing device information, so as to authenticate the target computing power resource access request from the client based on the user authentication information.

6. A computing resource access system for accessing multiple computing resources, wherein the multiple computing resources include intelligent computing resources, super computing resources, and general computing resources, each computing resource corresponds to a message channel, and the system includes: A registration server, which is in communication connection with a plurality of computing power providing devices corresponding to each computing power resource, and is suitable for executing the method as described in any one of claims 1 to 5; A proxy server, which is connected to the registration server in communication, and is adapted to be connected to a plurality of computing power providing devices corresponding to each computing power resource, so as to access the computing power resources provided by each computing power providing device; One or more clients are connected to the proxy server for communication, and are suitable for sending a target computing resource access request to the registration server via the proxy server.

7. A computing device comprising: at least one processor; and A memory storing program instructions, wherein the program instructions are configured to be processed by the at least one processor, and the program instructions include instructions for processing the method according to any one of claims 1 to 5.

8. A computer program product comprising a computer program / instructions, wherein: When the computer program / instructions are executed by a processor, the method according to any one of claims 1 to 5 is implemented.

9. A readable storage medium storing program instructions, which, when read and processed by a computing device, enables the computing device to process the method according to any one of claims 1 to 5.

Images