A computing power supply node open service architecture and interface structure

Abstract

The present invention relates to the field of computing power network technology, and in particular to an open service architecture and interface structure for a computing power supply node, including a computing power demander, a computing power scheduling node and a computing power supply node. The present invention designs a standardized computing power supply node service architecture and an open interface, so that computing power supply nodes of different specifications have the ability to manage the computing tasks of the node and refresh the status of the node in real time, providing a real-time, standardized interactive interface for upper-layer computing power scheduling, solving the problem that the computing power scheduling layer has difficulty in judging when selecting computing power supply nodes due to huge differences in hardware standards, computing power / storage particle standards, business type acceptance capabilities, security isolation capabilities, etc. of computing power supply nodes of different specifications, thereby improving the efficiency of computing power acceptance.

Description

Technical Field

[0001] The present invention relates to the field of computing power network technology, and specifically to an open service architecture and interface structure for computing power supply nodes. Background Art

[0002] Existing technologies mainly focus on areas such as computing power scheduling, computing power migration, and computing power slicing in computing power networks. For computing power supply nodes that undertake computing tasks, there is a lack of standardized service architecture and open interface design. As a result, the selection of computing power supply nodes of different specifications relies too much on the comprehensive judgment ability of computing power scheduling nodes, which increases the difficulty of scheduling computing power networks.

[0003] In summary, the present invention solves the existing problems by designing a service architecture and interface structure that opens up computing power supply nodes to the outside world. Summary of the Invention

[0004] The purpose of the present invention is to provide a computing power supply node open service architecture and interface structure to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A computing power supply node open service architecture and interface structure includes computing power demanders, computing power scheduling nodes, and computing power supply nodes. The functions of each module in the computing power supply node are designed from the bottom up, as follows:

[0007] 1) Hardware layer: manages hardware, provides monitoring and early warning of CPU, GPU, storage, power supply, temperature, load, and other indicators, and undertakes computing and storage tasks issued by the virtualization layer. It can be a single server or a server cluster.

[0008] 2) Virtualization layer: Provides granularization and virtualization packaging of hardware resources, packaging computing power and storage resources into granular scheduling units, for example: computing power granule: 1 kH / s = 1,000 hashes per second; storage granule: 32MB / unit;

[0009] 3) Task Container Module: This module provides isolated container services for each computing task. Each task container carries independent computing task parameters, computing power, and storage resources, achieving secure isolation between different tasks. The minimum unit of computing power / storage resources is a computing power / storage particle defined by the virtual layer. The specific number of particles is configured according to task requirements.

[0010] 4) Result Verification Module: Verifies the task calculation results based on the result verification parameters issued by the task. If the verification is qualified, the task scheduling module will retrieve the data. If the verification is unqualified, the task container module will be instructed to recalculate.

[0011] 5) Task orchestration module: manages all computing tasks running on computing power supply nodes and maintains a dynamically changing computing task table;

[0012] 6) Status Perception Module: Real-time perception of the current status of the computing power supply node and reporting to the computing power open interface in real time;

[0013] 7) Externally Open Interface: Responsible for interacting with external computing power scheduling nodes, taking on computing tasks, and feeding back computing results; interface interaction parameters include: computing power supply node status, local node computing power / storage granularity standard, security isolation capability, result verification capability, supported business types, local IP address, peer IP address, computing task parameters, and result data.

[0014] As a preferred solution of the present invention, the specific process of the interface interaction is as follows:

[0015] S1, the state perception of computing power supply nodes, and task scheduling, real-time synchronization update of computing power supply node status;

[0016] S2, the computing power scheduling node initiates a status query request to the computing power supply node;

[0017] S3, the external open interface of the computing power supply node transparently transmits the query request to the status perception;

[0018] S4: Status perception feedback (including "node open / closed", computing power / storage granularity standards, computing power / storage margin, security isolation capability, result verification capability, supported business types, and computing power supply node IP") to the external interface.

[0019] S5, the external open interface is transparently transmitted to the computing power scheduling node;

[0020] S6, the computing power scheduling node matches computing tasks based on the current status and capabilities fed back by the computing power supply node;

[0021] S7, when the computing task requirements successfully match the current status and capabilities of the computing power supply node, the computing power scheduling node sends the computing task parameters to the external open interface, including: computing parameters, computing power requirements, storage requirements, result verification parameters, etc.;

[0022] S8, open interface to forward computing task parameters to the task container;

[0023] S9, the task container allocates container resources, computing power / storage resources for the new computing task and starts the computing task;

[0024] S10, during the calculation process, the task container reports the task execution progress and current status to the task scheduler in real time according to the refresh interval defined by the task scheduler;

[0025] S11, task scheduling, real-time maintenance and refresh of task table;

[0026] S12, when the task calculation is completed, the task container pushes the calculation task information, result data and verification parameters to the result verification;

[0027] S13, the result verification starts the task verification and informs the task scheduler, the task status changes from calculating to verifying;

[0028] S14, when the result verification is completed, and the verification result meets the requirements, the result verification informs the task scheduler, and the task status becomes completed;

[0029] S15, result verification pushes task information, calculation results and verification results to the open interface;

[0030] S16, the externally open interface transparently transmits the above data information to the computing power scheduling node, and the interaction process is completed.

[0031] As a preferred solution of the present invention, the security isolation between different tasks includes but is not limited to the security isolation of task parameters, computing power, storage, and intermediate data / results.

[0032] As a preferred solution of the present invention, the real-time perception of the current status of the computing power supply node includes but is not limited to whether it undertakes computing tasks, the current amount of parallel computing tasks, the current computing power / storage idle amount and the externally open IP address.

[0033] As a preferred solution of the present invention, the dynamically changing computing task table includes but is not limited to the task ID, the container ID to which it belongs, the task progress and the amount of computing power / storage resources occupied.

[0034] As a preferred solution of the present invention, the computing task parameters sent by the computing power scheduling node to the external open interface include but are not limited to: computing parameters, computing power requirements, storage requirements and result verification parameters.

[0035] Compared with the prior art, the present invention has the following beneficial effects:

[0036] 1. In the present invention, by designing a standardized computing power supply node service architecture and an externally open interface, computing power supply nodes of different specifications are equipped with the ability to manage the computing tasks of the node and refresh the status of the node in real time, providing a real-time, standardized interactive interface for the upper-layer computing power scheduling, solving the problem that the computing power scheduling layer has difficulty in judging when selecting computing power supply nodes due to huge differences in hardware standards, computing power / storage particle standards, business type acceptance capabilities, security isolation capabilities, etc. of computing power supply nodes of different specifications, thereby improving the efficiency of computing power acceptance and effectively solving the problem that the existing technology mainly focuses on the fields of computing power scheduling, computing power migration, computing power slicing, etc. in the computing power network. For computing power supply nodes that undertake computing tasks, there is a lack of standardized service architecture and externally open interface design, which makes the selection of computing power supply nodes of different specifications overly dependent on the comprehensive judgment ability of the computing power scheduling node. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a schematic diagram of the computing power supply node service architecture of the present invention;

[0038] Figure 2 This is a schematic diagram of the virtual layer architecture structure of the present invention;

[0039] Figure 3 This is a schematic diagram of the task container architecture structure of the present invention;

[0040] Figure 4 This is a schematic diagram of the task arrangement architecture of the present invention;

[0041] Figure 5 This is a schematic diagram of the interactive process structure of the present invention;

[0042] Figure 6 For the present invention Figure 5 Interaction process continued structural diagram;

[0043] Figure 7 This is a schematic diagram of the computing task matching structure of the computing power scheduling node of the present invention based on the current status and capabilities fed back by the computing power supply node. DETAILED DESCRIPTION

[0044] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0045] To facilitate understanding of the present invention, the present invention will be described more comprehensively below with reference to the relevant drawings, and several embodiments of the present invention are given. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0046] It should be noted that when an element is referred to as being "fixed on" another element, it may be directly on the other element or there may be a central element. When an element is considered to be "connected to" another element, it may be directly connected to the other element or there may be a central element at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used in this article are for illustrative purposes only.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention pertains. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0048] See also Figure 1-7 , the present invention provides a technical solution:

[0049] A computing power supply node opens its service architecture and interface structure to the outside world, including computing power demanders, computing power scheduling nodes, and computing power supply nodes.

[0050] In this solution, when a computing power demander initiates a computing demand in the computing power network, the computing power scheduling nodes in the computing power network perform computing power scheduling and matching. As a computing power supply node, how can we provide a standard and unified computing power capacity open service architecture and interface for the computing power network?

[0051] 1. Open service architecture for computing power supply nodes

[0052] The computing power supply node service architecture is shown in Figure 1 ;

[0053] The functional design of each module is described from bottom to top as follows:

[0054] 1) Hardware layer: manages hardware, provides monitoring and early warning of CPU, GPU, storage, power supply, temperature, load, and other indicators, and undertakes computing and storage tasks issued by the virtualization layer. It can be a single server or a server cluster.

[0055] 2) Virtualization layer: Provides granular and virtualized packaging of hardware resources, packaging computing power / storage resources into granular scheduling units, for example: computing power granule: 1 kH / s = 1,000 hashes per second; storage granule: 32MB / piece, see the diagram Figure 2 ;

[0056] 3) Task container module: provides isolated container services for each computing task. Each task container carries independent computing task parameters, computing power and storage resources, achieving secure isolation between different tasks (task parameters, computing power, storage, intermediate data / results, etc. are all securely isolated). The minimum unit of computing power / storage resources is a computing power / storage particle defined by the virtual layer. The specific number of particles is configured according to task requirements. See the schematic diagram. Figure 3 ;

[0057] 4) Result Verification Module: Verifies the task calculation results based on the result verification parameters issued by the task. If the verification is qualified, the task scheduling module will retrieve the data. If the verification is unqualified, the task container module will be instructed to recalculate.

[0058] 5) Task Scheduling Module: manages all computing tasks running on computing power supply nodes; maintains a dynamically changing computing task table, including task ID, container ID, task progress, computing power / storage resource usage, etc. (See the schematic diagram for details). Figure 4 ;

[0059] 6) Status Perception Module: This module perceives the current status of the computing power supply node in real time (including whether it is accepting computing tasks, the current number of parallel computing tasks, the current amount of idle computing power / storage, and the externally exposed IP address), and reports this to the computing power external interface in real time.

[0060] 7) Externally Open Interface: Responsible for interacting with external computing power scheduling nodes, accepting computing tasks, and feeding back computing results. Interface interaction parameters include: computing power supply node status, computing power / storage granularity standard of this node, security isolation capability, result verification capability, supported business types, local IP address, peer IP address, computing task parameters, and result data. The interface interaction process is detailed below.

[0061] 2. Hashrate Open Interface Interaction Process

[0062] See the interactive process diagram Figure 5 and Figure 6 ;

[0063] 1) Status awareness of computing power supply nodes, task orchestration, and real-time synchronization of computing power supply node status updates;

[0064] 2) The computing power scheduling node initiates a status query request to the computing power supply node;

[0065] 3) The externally open interface of the computing power supply node transparently transmits the query request to the status perception;

[0066] 4) Status perception feedback: "node open / closed, node computing power / storage granularity standard, computing power / storage margin, security isolation capability, result verification capability, supported business types, computing power supply node IP" to the external open interface;

[0067] 5) The external open interface is transparently transmitted to the computing power scheduling node;

[0068] 6) The computing power scheduling node matches computing tasks based on the current status and capabilities of the computing power supply node. For example, see Figure 7 ;

[0069] 7) When the computing task requirements successfully match the current status and capabilities of the computing power supply node, the computing power scheduling node sends the computing task parameters to the external open interface, including: computing parameters, computing power requirements, storage requirements, result verification parameters, etc.

[0070] 8) Open interfaces to forward computing task parameters to task containers;

[0071] 9) The task container allocates container resources, computing power / storage resources for the new computing task and starts the computing task;

[0072] 10) During the computation process, the task container reports the task execution progress and current status to the task orchestration in real time according to the refresh interval defined by the task orchestration;

[0073] 11) Task scheduling: Real-time maintenance and refresh of task tables;

[0074] 12) When the task calculation is completed, the task container pushes the calculation task information, result data and verification parameters to the result verification;

[0075] 13) Result verification starts the task verification and notifies the task scheduler. The task status changes from calculating to verifying.

[0076] 14) When the result verification is completed and the verification result meets the requirements, the result verification notifies the task scheduler and the task status changes to completed;

[0077] 15) Result verification pushes task information, calculation results, and verification results to the open interface;

[0078] 16) The open interface transparently transmits the above data information to the computing power scheduling node, and the interaction process is completed.

[0079] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A computing power supply node open service architecture and interface structure, including computing power demanders, computing power scheduling nodes and computing power supply nodes, characterized by: The functions of each module in the computing power supply node are designed from bottom to top, as follows: 1) Hardware layer: manages hardware, provides CPU, GPU, storage, power, temperature, and load indicator monitoring and early warning, and undertakes computing and storage tasks issued by the virtualization layer. It uses a single server or a server cluster. 2) Virtualization layer: Provides granularization and virtualization packaging of hardware resources, packaging computing power and storage resources into granular scheduling units, where: computing power granule: 1 kH / s = 1,000 hashes per second; storage granule: 32MB / unit; 3) Task Container Module: This module provides isolated container services for each computing task. Each task container carries independent computing task parameters, computing power, and storage resources, achieving secure isolation between different tasks. The minimum unit of computing power / storage resources is a computing power / storage particle defined by the virtual layer. The specific number of particles is configured according to task requirements. 4) Result Verification Module: Verifies the task calculation results based on the result verification parameters issued by the task. If the verification is qualified, the pending computing task scheduling module retrieves the data. If the verification is unqualified, the task container module is instructed to recalculate. 5) Task orchestration module: manages all computing tasks running on computing power supply nodes and maintains a dynamically changing computing task table; 6) Status Perception Module: Real-time perception of the current status of the computing power supply node and reporting to the computing power open interface in real time; 7) Externally Open Interface: Responsible for interacting with external computing power scheduling nodes, taking on computing tasks, and feeding back computing results; interface interaction parameters include: computing power supply node status, local node computing power / storage granularity standard, security isolation capability, result verification capability, supported business types, local IP address, peer IP address, computing task parameters, and result data.

2. The computing power supply node open service architecture and interface structure according to claim 1, characterized in that: The specific process of the interface interaction is as follows: S1, the state perception of computing power supply nodes, and task scheduling, real-time synchronization update of computing power supply node status; S2, the computing power scheduling node initiates a status query request to the computing power supply node; S3, the external open interface of the computing power supply node transparently transmits the query request to the status perception; S4: Status perception feedback ("node open / closed", node computing power / storage granularity standard, computing power / storage margin, security isolation capability, result verification capability, supported business types, computing power supply node IP") to the external open interface; S5, the external open interface is transparently transmitted to the computing power scheduling node; S6, the computing power scheduling node matches computing tasks based on the current status and capabilities fed back by the computing power supply node; S7, when the computing task requirements and the current state and capabilities of the computing power supply node match successfully, the computing power scheduling node sends the computing task parameters to the open interface, including: computing parameters, computing power requirements, storage requirements, and result verification parameters; S8, open interface to forward computing task parameters to the task container; S9, the task container allocates container resources, computing power / storage resources for the new computing task and starts the computing task; S10, during the calculation process, the task container reports the task execution progress and current status to the task scheduler in real time according to the refresh interval defined by the task scheduler; S11, task scheduling, real-time maintenance and refresh of task table; S12, when the task calculation is completed, the task container pushes the calculation task information, result data and verification parameters to the result verification; S13, the result verification starts the task verification and informs the task scheduler, the task status changes from calculating to verifying; S14, when the result verification is completed, and the verification result meets the requirements, the result verification informs the task scheduler, and the task status becomes completed; S15, result verification pushes task information, calculation results and verification results to the open interface; S16, the externally open interface transparently transmits the above data information to the computing power scheduling node, and the interaction process is completed.

3. The computing power supply node open service architecture and interface structure according to claim 1, characterized in that: The security isolation between different tasks includes the security isolation of task parameters, computing power, storage, and intermediate data / results.

4. The computing power supply node open service architecture and interface structure according to claim 1, characterized in that: The real-time perception of the current status of the computing power supply node includes whether it undertakes computing tasks, the current amount of parallel computing tasks, the current computing power / storage idle amount and the externally open IP address.

5. The computing power supply node open service architecture and interface structure according to claim 1 is characterized by: The dynamically changing computing task table is maintained, including the task ID, the container ID to which it belongs, the task progress, and the amount of computing power / storage resources occupied.

6. The computing power supply node open service architecture and interface structure according to claim 2, characterized in that: The computing power scheduling node sends computing task parameters including computing parameters, computing power requirements, storage requirements and result verification parameters to the external open interface.

Images