WO2026007281A1 - 资源信息收集功能架构、处理方法、电子设备和存储介质 - Google Patents

资源信息收集功能架构、处理方法、电子设备和存储介质

Info

Publication number
WO2026007281A1
WO2026007281A1 PCT/CN2024/127412 CN2024127412W WO2026007281A1 WO 2026007281 A1 WO2026007281 A1 WO 2026007281A1 CN 2024127412 W CN2024127412 W CN 2024127412W WO 2026007281 A1 WO2026007281 A1 WO 2026007281A1
Authority
WO
WIPO (PCT)
Prior art keywords
resource information
resource
computing power
power network
component
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
PCT/CN2024/127412
Other languages
English (en)
French (fr)
Inventor
卫敏
赵倩颖
雷波
马晨昊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
China Telecom Corp Ltd Technology Innovation Center
China Telecom Corp Ltd
Original Assignee
China Telecom Corp Ltd Technology Innovation Center
China Telecom Corp Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by China Telecom Corp Ltd Technology Innovation Center, China Telecom Corp Ltd filed Critical China Telecom Corp Ltd Technology Innovation Center
Publication of WO2026007281A1 publication Critical patent/WO2026007281A1/zh
Pending legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/904—Browsing; Visualisation therefor
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]

Definitions

  • This application relates to the field of computing network technology, and in particular to a resource information collection functional architecture, processing method, electronic device, storage medium and computer program product.
  • the control layer collects resource information from the resource layer of the computing power network through methods such as computing power routing, and then sends this information to the service layer for further processing.
  • the computing network control layer needs to collect various resource information. Furthermore, the collection of resource information is a prerequisite for subsequent resource scheduling and utilization by the computing network; only after collecting various resource information can resources be monitored, managed, scheduled, and applied.
  • related technologies have not researched the methods for collecting resource information as proposed in the relevant standards, resulting in the computing network control layer being unable to dynamically grasp the resource information of the entire network.
  • this application provides a resource information collection functional architecture, deployed in the computing power network control layer, including:
  • the resource information sensing component is responsible for connecting to the computing power network resource layer and obtaining resource information from the computing power network resource layer.
  • a resource information announcement component is used to receive resource information from the resource information sensing component and announce the resource information.
  • the resource information integration component is used to receive resource information and generate an integrated resource information table
  • the resource view generation component is used to generate a visualized resource information table based on the integrated resource information table, and send the visualized resource information table to the computing power network service layer and/or the computing power network orchestration management layer.
  • the resource information sensing component periodically or as needed obtains resource information from the computing power network resource layer and sends the obtained resource information to the resource information announcement component.
  • the resource information obtained by the resource information sensing component from the computing power network resource layer includes at least one of the following: computing power resource information, network resource information, storage resource information, and algorithm resource information.
  • the resource information announcement component is used to receive resource information from the resource information sensing component and announce the resource information across the entire network.
  • the resource information integration component receives the announced resource information, including computing power resource information, storage resource information, algorithm resource information from various computing nodes across the network, as well as network resource information between each computing node and other devices, and generates an integrated resource information table containing all resource information.
  • the resource view generation component extracts some information based on the integrated resource information table to generate the visualized resource information table, and sends the visualized resource information table to the computing power network service layer and/or the computing power network orchestration management layer.
  • this application also provides a method for collecting and processing resource information.
  • the method includes:
  • the obtained resource information will be announced;
  • a visualized resource information table is generated, and the visualized resource information table is sent to the computing power network service layer and/or the computing power network orchestration management layer.
  • obtaining resource information from the computing power network resource layer and announcing the obtained resource information includes:
  • the resource information is transmitted from the computing power network resource layer to the resource information sensing component;
  • the resource information is sent from the resource information sensing component to the resource information announcement component;
  • the resource information announcement component receives resource information from the resource information perception component and announces the received resource information across the entire network.
  • the resource information transmitted from the computing power network resource layer to the resource information sensing component includes at least one of computing power resource information, network resource information, storage resource information, and algorithm resource information.
  • generating the integrated resource information table includes:
  • the resource information in the announcement is received by the resource information integration component, which then generates an integrated resource information table containing all resource information.
  • the resource information received and announced by the resource information integration component includes: computing power resource information, storage resource information, and algorithm resource information from all computing nodes across the network, as well as network resource information between each computing node and other devices.
  • the step of generating a visualized resource information table based on the integrated resource information table and sending the visualized resource information table to the computing power network service layer and/or the computing power network orchestration management layer includes:
  • the integrated resource information table is passed to the resource view generation component, which extracts some information to generate a visual resource information table.
  • the resource view generation component sends the visualized resource information table to the computing power network service layer and/or the computing power network orchestration management layer.
  • this application also provides an electronic device.
  • the electronic device includes a memory and a processor.
  • the memory stores a computer program
  • the processor executes the computer program to implement steps such as a resource information collection and processing method.
  • This method can be:
  • the obtained resource information will be announced;
  • a visualized resource information table is generated, and the visualized resource information table is sent to the computing power network service layer and/or the computing power network orchestration management layer.
  • the computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of a resource information collection and processing method, which may include:
  • the obtained resource information will be announced;
  • a visualized resource information table is generated, and the visualized resource information table is sent to the computing power network service layer and/or the computing power network orchestration management layer.
  • the computer program product includes a computer program that, when executed by a processor, implements steps such as those of a resource information collection and processing method, which can be:
  • the obtained resource information will be announced;
  • a visualized resource information table is generated, and the visualized resource information table is sent to the computing power network service layer and/or the computing power network orchestration management layer.
  • the aforementioned resource information collection functional architecture, processing method, electronic equipment, storage medium, and computer program products, within the resource information collection functional architecture of the computing power network control layer, include four functional components: a resource information perception component, a resource information announcement component, a resource information integration component, and a resource view generation component.
  • the resource information perception component connects to the computing power network resource layer, obtains resource information from the computing power network resource layer, and sends the resource information to the resource information announcement component.
  • the resource information announcement component receives the resource information and announces it.
  • the resource information integration component receives the announced resource information and generates an integrated resource information table containing all resource information.
  • the resource view generation component extracts the required information from the integrated resource information table and generates a visualized resource information table, which is then sent to the computing power network service layer and/or the computing power network orchestration management layer, thereby achieving visualization of resource monitoring and the transaction process.
  • the aforementioned resource information collection architecture can acquire resource information from the computing power network resource layer, announce the acquired resource information, and then integrate the various types of resource information acquired through the resource information integration component to obtain an integrated resource information table corresponding to the computing power network resource layer, thereby improving the comprehensiveness and standardization of the collected resource information.
  • the resource view generation component can extract information from the integrated resource information table and obtain a corresponding visualized resource information table, which is provided to the computing power network service layer and/or the computing power network orchestration management layer, thus ensuring that the computing power network service layer and/or the computing power network orchestration management layer have the visualized resource information required by them.
  • Figure 1 is an application environment diagram of the resource information collection function architecture in one embodiment
  • Figure 2 is a schematic diagram of the architecture of the computing power network control layer in one embodiment
  • Figure 3 is a flowchart illustrating a resource information collection and processing method in one embodiment
  • Figure 4 is a flowchart illustrating the resource information collection and processing method in another embodiment
  • Figure 5 is a flowchart illustrating the resource information collection and processing method in another embodiment
  • Figure 6 is a schematic diagram of resource information synchronization in a centralized architecture, hybrid architecture, and distributed architecture in one embodiment
  • Figure 7 is a schematic diagram of the resource information synchronization process in one embodiment
  • Figure 8 is a schematic diagram of the resource information synchronization process in another embodiment
  • Figure 9 is an internal structure diagram of an electronic device in one embodiment.
  • the resource information collection functional architecture provided in this application embodiment can be applied to the computing power network shown in Figure 1.
  • a computing power network is a technology that distributes computing power, algorithms, and storage information based on the network control plane. Combined with network information, this technology can achieve the distribution, association, transaction, and allocation of computing resources, network resources, and storage resources, centered on business needs, thereby achieving the goal of optimal allocation of network resources.
  • a computing network architecture comprises a computing network service layer, a computing network control layer, a computing network resource layer, and a computing network orchestration and management layer.
  • the computing network control layer has three basic functions: resource information collection, resource allocation, and network connection scheduling.
  • the resource information collection function is responsible for collecting information on the entire network's computing power and network resources.
  • the resource allocation function is responsible for configuring the optimal computing nodes according to business needs.
  • the network connection scheduling function is responsible for configuring network connections from the user's service side to the computing resource side based on business needs and the resource allocation module's resource configuration, to meet the user's requirements for bandwidth, latency, and QoS (Quality of Service).
  • the computing network control layer needs to collect various resource information. This collection is a prerequisite for the computing network to perform subsequent resource allocation and network connection scheduling functions. Only by dynamically grasping the resource information of the entire network can resources be monitored, managed, scheduled, and applied. Therefore, the computing network control layer is the core of computing network technology.
  • This disclosure provides an implementation architecture and processing procedure for resource information collection in the computing power network control layer. Specifically, it proposes a resource information collection functional architecture comprising four functional components: a resource information perception component, a resource information announcement component, a resource information integration component, and a resource view generation component.
  • the processing procedure for resource information collection includes: resource information being transmitted from the computing power network resource layer to the resource information perception component; resource information being sent from the resource information perception component to the resource information announcement component; the resource information announcement component receiving the resource information from the resource information perception component and announcing the received resource information across the entire network; the announced resource information being received by the resource information integration component, which generates an integrated resource information table containing all resource information; the integrated resource information table being passed to the resource view generation component, which extracts partial information to generate a visualized resource information table; and the visualized resource information table being sent to the computing power network service layer and/or the computing power network orchestration management layer through the resource view generation component to support visualization of resource transactions and monitoring processes.
  • the computing power network resource layer includes resource information such as computing power resource information, network resource information, storage resource information, and algorithm resource information.
  • the performance requirements for the computing network control layer include: requiring real-time sensing terminals (already ordered)
  • the requirements for the computing power network control layer will be described, including the location of the highly mobile terminal, network status (such as network jitter, network latency, and network bandwidth), and the processing capabilities of the computing nodes; as well as the requirement to support intelligent allocation and optimization.
  • the detailed requirements for the technical capabilities (network information collection capability, resource allocation capability, and resource allocation capability) required to support business needs will be specified.
  • the requirements for network information collection functions in the computing power network control layer include: supporting the dynamic acquisition of resource information from the computing power network resource layer, including but not limited to computing power resource information, network resource information, storage resource information, and algorithm resource information; supporting the announcement/publishing of resource information, synchronizing different resource information across the entire network so that users can grasp the overall resource status; and supporting the management of resource information, including not only global resource information but also the generation of simplified resource information entries according to user needs, to support the visualization functions of the computing power network service layer and/or the computing power network orchestration management layer.
  • a resource information collection functional architecture is provided. This is illustrated using the application of this architecture to the computing power network control layer in Figure 1 as an example.
  • the resource information collection function 200 includes four components: a resource information perception component 201, a resource information announcement component 202, a resource information integration component 203, and a resource view generation component 204. Each component is described below.
  • the resource information sensing component 201 is responsible for connecting with the computing power network resource layer and obtaining resource information from the computing power network resource layer.
  • the resource information sensing component 201 is responsible for connecting to the computing power network resource layer, obtaining resource information from the computing power network resource layer periodically or on demand, and sending the obtained resource information to the resource information announcement component.
  • the resource information obtained by the resource information sensing component 201 from the computing power network resource layer includes, but is not limited to, computing power resource information, network resource information, storage resource information, and algorithm resource information.
  • the resource information sensing component 201 can sense the resource information of each node in the computing power network through sensing technology.
  • the sensing technology can sense the resource information of the network resource layer according to a preset period or according to other sensing conditions set based on requirements.
  • the resource information sensing component 201 can be pre-configured with sensing conditions for obtaining resource information from the computing power network resource layer, including parameters such as triggering conditions, periodic conditions, resource requirements, and collection intervals.
  • the computing power network control layer can collect the resource information contained in the computing power network resource layer according to the above parameters and sensing technology. In each period, resource information that meets the parameters can be collected and used as the resource information for that period.
  • the resource information sensing component 201 can periodically perform multiple resource collections to obtain resource information for multiple time periods.
  • the computing network resource layer contains various resource information.
  • the resource information sensing component 201 can collect resource information corresponding to different time points. For example, the resource information sensing component 201 can determine that the trigger condition is a preset hourly time, such as 6:00 AM or 12:00 AM. When the time meets the trigger condition, the computing resource control layer collects resource information from the computing network resource layer through the resource information sensing component.
  • the resource information sensing component 201 can be configured with periodic conditions, such as setting one day as a period, meaning that the resource information acquired within 24 hours is the resource information corresponding to that period.
  • the resource information sensing component 201 can be configured with resource requirements, such as determining the resource information of the computing nodes included in the computing network, as well as the resource information between computing nodes and other devices.
  • the resource information sensing component 201 can be configured with periodic condition collection intervals, such as setting the collection interval within a one-day period. If the interval is 1 hour, then resource information can be collected once every hour.
  • the resource information announcement component 202 is used to receive resource information from the resource information sensing component 201 and announce the resource information.
  • the resource information announcement component 202 is used to receive resource information from the resource information sensing component 201 and announce the resource information across the entire network.
  • the resource information announcement component 202 can announce resource information across the entire network according to a preset announcement strategy.
  • the entire network can be understood as including all computing power gateways and/or centralized management and control platforms in the computing power network, so that each computing power gateway and each centralized management and control platform in the computing power network can obtain resource information from the entire network.
  • the resource information announcement component 202 can announce resource information within a control entity (e.g., a computing gateway or centralized management platform) in the computing power network, or it can announce resource information between different control entities (computing gateways or centralized management platforms).
  • a control entity e.g., a computing gateway or centralized management platform
  • the method can include function calls, message queues, pass-through, etc.
  • the method can include extended routing protocols, RESTful APIs, message queues, etc.
  • the resource information announcement component 202 can receive resource information acquired by the resource information sensing component 201. Based on the acquired resource information, the resource information announcement component 202 can determine the resource information to be announced, and the fields to be announced for each resource type, according to an announcement strategy, thereby determining the target announcement resource information. The resource information announcement component 202 announces the target announcement resource information across the entire network, including internal announcements within the current computing power gateway or the current centralized management platform, and external announcements to other computing power gateways or other centralized management platforms.
  • the announcement strategy could be to announce network resource information, specifically the first to tenth fields of that information.
  • the resource information announcement component 202 can announce the first to tenth fields of the network resource information contained in each node of the computing power network.
  • the resource information announcement component 202 can configure permissions for target announcement resource information, configuring different levels of permissions for different target announcement resource information, such as level 1 to level 10 permissions. The higher the permission level, the higher the level of confidentiality.
  • a query terminal queries target announcement resource information, it needs to determine whether it can view the target announcement resource information based on the query terminal's permission information. If the target announcement resource information has level 3 permissions, then the query terminal's permission information is level 3 to level 10.
  • the resource information integration component 203 is used to receive resource information and generate an integrated resource information table.
  • the resource information integration component 203 is used to receive the resource information from the announcement and generate an integrated resource information table.
  • the resource information integration component 203 can receive resource information announced by the resource information announcement component 202. Since the resource information announcement component 202 can be a resource information announcement component corresponding to other computing power gateways or other centralized management and control platforms, or it can be a resource information announcement component of the current computing power gateway or the current centralized management and control platform where the resource information integration component 203 is located, and each resource information announcement component announces the resource information it has obtained across the entire network, the resource information integration component 203 can receive resource information from the entire network, and the generated integrated resource information table can contain resource information from the entire network.
  • the resource information integration component 203 can integrate the received resource information from various computing nodes across the entire network into a single resource information table, resulting in an integrated resource information table. Since the resource information sensing component 201 can acquire multiple types of resource information within a given time period, it can also acquire... Based on the same type of resource information contained within the same time period, the resource information integration component 203 can integrate multiple types of resource information contained within multiple time periods into one information table by integrating the resource information table.
  • the resource view generation component 204 is used to generate a visualized resource information table based on the integrated resource information table, and send the visualized resource information table to the computing power network service layer and/or the computing power network orchestration management layer.
  • the resource view generation component 204 is used to extract partial information from the integrated resource information table to generate a visualized resource information table, and send the visualized resource information table to the computing power network service layer and/or the computing power network orchestration management layer.
  • the resource information contained in the visualized resource information table can be determined based on the specific business needs of the user side. Different business needs on the user side can generate different visualized resource information tables, which are then sent to the computing power network service layer and/or the computing power network orchestration management layer.
  • the resource view generation component 204 can generate multiple visualized resource information tables at once based on the integrated resource information table, and then send a portion of the multiple visualized resource information tables to the computing power network orchestration management layer and another portion to the computing power network service layer.
  • the resource view generation component 204 generates a single visualized resource information table at once based on the integrated resource information table, and then sends this visualized resource information table to either the computing power network orchestration management layer or the network service layer.
  • the resource view generation component 204 can extract partial information corresponding to specific business needs from the integrated resource information table according to the specific business needs of the user, and generate a visualized resource information table based on the extracted partial information.
  • the resource view generation component 204 sends the generated visualized resource information table to the computing power network service layer and/or the computing power network orchestration management layer.
  • the computing power network orchestration management layer and the computing power network service layer can respectively call the corresponding programs to render the visualized resource information table to obtain a visualized image, and monitor the data of the visualized resource information table through the visualized image to support the visualization of resource transactions and monitoring processes.
  • the aforementioned resource information collection functional architecture includes four functional components: a resource information perception component 201, a resource information announcement component 202, a resource information integration component 203, and a resource view generation component 204.
  • the resource information perception component 201 connects to the computing power network resource layer, obtains resource information from it, and sends the information to the resource information announcement component 202.
  • the resource information announcement component 202 then announces the received resource information.
  • the resource information integration component 203 receives the announced resource information and generates an integrated resource information table containing all resource information.
  • the resource view generation component 204 extracts the required information from the integrated resource information table and generates a visualized resource information table, which is then sent to the computing power network service layer and/or the computing power network orchestration management layer, thereby achieving visualization of resource monitoring and the transaction process.
  • the aforementioned resource information collection architecture can acquire resource information from the computing power network resource layer and announce the acquired resource information.
  • the resource information integration component 203 integrates the various types of resource information received from the announcements to obtain an integrated resource information table corresponding to the computing power network resource layer, thereby improving the comprehensiveness and standardization of the collected resource information.
  • the resource view generation component 204 can extract information from the integrated resource information table and obtain a corresponding visualized resource information table, which is provided to the computing power network service layer and/or the computing power network orchestration management layer, thus ensuring that the computing power network service layer and/or the computing power network orchestration management layer have the visualized resource information they need.
  • the resource information obtained by the resource information sensing component from the computing power network resource layer includes at least one of computing power resource information, network resource information, storage resource information, and algorithm resource information.
  • computing resource information can represent computing resource information such as the total computing power and currently available computing power of each computing node.
  • Network resource information can represent the amount of network resources of each computing node at a certain moment, such as the total bandwidth allocated to the computing node at 12:00, and the bandwidth already used.
  • Storage resource information can represent the computing node's... The total amount of storage resources, the amount of storable resources, and the amount of stored resources at a given moment.
  • Algorithm resources within a computing network refer to a collection of algorithms provided for data processing, analysis, and computation through distributed computing and network resources. These algorithm resources can be flexibly invoked on cloud computing platforms to achieve efficient and rapid computational processing.
  • Algorithm resources can be machine learning algorithms, such as support vector machines and deep neural networks.
  • the resource information sensing component can obtain various types of resource information from the computing power network resource layer.
  • the resource information may include computing power resource information, network resource information, storage resource information, and algorithm resource information. This can increase the types and amount of resource information obtained by the resource information sensing component, improve the data dimension of the obtained data, and enhance the comprehensiveness of the obtained resource information.
  • the resource information integration component receives the announced resource information, including computing power resource information, storage resource information, algorithm resource information from various computing nodes across the network, as well as network resource information between various computing nodes across the network and other devices, and generates an integrated resource information table containing all resource information.
  • other devices may be electronic devices that have a network connection with the computing node, such as personal terminals, servers, etc.
  • the computing network control layer obtains computing resource information, storage resource information, algorithm resource information, and network resource information between each computing node and other devices from the computing network resource layer through a resource information sensing component.
  • the resource information integration component then stores the current resource information of each computing node and the network resource information between each computing node and other devices in an integrated resource information table.
  • a resource information collection and processing method is provided, which is illustrated using the application of this method to the computing power network control layer in Figure 1 as an example.
  • the processing method includes:
  • obtaining resource information from the computing power network resource layer includes at least one of computing power resource information, network resource information, storage resource information, and algorithm resource information.
  • the aforementioned announcement of the acquired resource information includes both internal and external announcements.
  • An internal announcement is understood as an announcement of resource information within the current computing power gateway or centralized management platform, while an external announcement is understood as an announcement of resource information to other computing power gateways or other centralized management platforms.
  • the computing network control layer can obtain diverse resource information from the computing network resource layer and announce this diverse resource information across the entire network. This allows each computing gateway or centralized management platform in the computing network to receive diverse resource information from all computing nodes across the network, and then integrate this diverse resource information to obtain an integrated resource information table containing all resource information. Further, based on the resource information in the integrated resource information table, information extraction can be performed to obtain a visualized resource information table matching different business needs. This visualized resource information table is then sent to the computing network service layer and/or the computing network orchestration management layer, enabling the computing network service layer and/or the computing network orchestration management layer to visualize the visualized resource information table, obtaining a visualized version. Resource information.
  • the method described above can obtain resource information from the computing power network resource layer, announce the obtained resource information, and then integrate the various types of resource information to obtain an integrated resource information table, thereby improving the comprehensiveness and standardization of the collected resource information.
  • the resource view generation component can extract information from the integrated resource information table and obtain a corresponding visualized resource information table, which is provided to the computing power network service layer and/or the computing power network orchestration management layer, thereby ensuring that the computing power network service layer and/or the computing power network orchestration management layer have the visualized resource information required by them.
  • a resource information collection and processing method is provided. Taking the application of this method to the computing power network control layer in FIG1 as an example, the resource information collection and processing method includes steps S402 to S412, which are described in detail below:
  • step S402 resource information is transmitted from the computing power network resource layer to the resource information sensing component.
  • the resource information sensing component can sense the resource information of each node in the computing power network through sensing technology.
  • the sensing technology can sense the resource information according to a preset period or a preset resource type.
  • the computing power network control layer can pre-configure parameters such as triggering conditions, periodic conditions, resource requirements, and collection intervals for sensing resource information.
  • the computing power network control layer can collect resource information contained in the computing power network resource layer based on these parameters and sensing technology. In each period, resource information that meets the parameters can be collected and used as the resource information for that period.
  • the computing power network control layer can periodically perform multiple resource collections to obtain resource information for multiple time periods.
  • the resource information transmitted from the computing power network resource layer to the resource information sensing component includes at least one of computing power resource information, network resource information, storage resource information, and algorithm resource information.
  • Step S404 Resource information is sent from the resource information perception component to the resource information announcement component.
  • the computing power network control layer can receive resource information obtained by the resource information sensing component through the resource information announcement component.
  • Step S406 The resource information announcement component receives resource information from the resource information perception component and announces the received resource information across the entire network.
  • the resource information announcement component can determine, according to the announcement strategy, the resource information corresponding to the resource type to be announced, and the announcement fields corresponding to the resource information of each resource type, and determine them as the target announcement resource information.
  • the resource information announcement component announces the target announcement resource information across the entire network.
  • step S408 the resource information in the announcement is received by the resource information integration component, which then generates an integrated resource information table containing all resource information.
  • the computing power network control layer can receive resource information of each node announced by the resource information announcement component through the resource information integration component.
  • the resource information integration component can integrate the received resource information into an integrated resource information table. Since the resource information sensing component can acquire multiple types of resource information within a single time period, and can also acquire the same type of resource information across multiple time periods, the resource information integration component can use the integrated resource information table to combine the various types of resource information across multiple time periods to obtain the integrated resource information table.
  • the resource information received by the resource information integration component includes: computing power resource information, storage resource information, and algorithm resource information from all computing nodes across the network, as well as information between computing nodes across the network and other devices.
  • Network resource information is generated by the resource information integration component into an integrated resource information table containing all resources.
  • the resource information sensing component can obtain various types of resource information from the computing power network resource layer.
  • the resource information may include computing power resource information, network resource information, storage resource information, and algorithm resource information. This can increase the types and amount of resource information obtained by the resource information sensing component, improve the data dimension of the obtained data, and enhance the comprehensiveness of the obtained resource information.
  • step S410 the integrated resource information table is passed to the resource view generation component, which extracts some information to generate a visual resource information table.
  • the resource view generation component can extract some information from the integrated resource information table according to the received resource filtering instructions, and convert all of the information into a visual resource information table.
  • Step S412 Send the visualized resource information table to the computing power network service layer and/or computing power network orchestration management layer through the resource view generation component to support the visualization of resource transaction and monitoring processes.
  • the resource view generation component sends the visualized resource information table corresponding to the resource filtering instruction to the computing power network service layer and/or the computing power network orchestration management layer.
  • the computing power network orchestration management layer and the computing power network service layer can respectively call the corresponding programs to perform visualization processing on the visualized resource information table, thereby obtaining a visualized image, and monitor the data of the visualized resource information table through the visualized image.
  • the above method can obtain resource information from the computing power network resource layer, announce the obtained resource information across the entire network, and then integrate the various types of resource information to obtain an integrated resource information table, thereby improving the comprehensiveness and standardization of the collected resource information.
  • the resource view generation component can extract some information from the integrated resource information table and obtain a visualized resource information table corresponding to some information, which is provided to the computing power network service layer and/or the computing power network orchestration management layer, thereby ensuring the visualization of key resource information required by the computing power network service layer and/or the computing power network orchestration management layer.
  • Step 51 The resource information sensing component obtains resource information from the computing power network resource layer, including but not limited to computing power resource information, network resource information, storage resource information, and algorithm resource information.
  • Step 52 The resource information sensing component sends the acquired resource information to the resource information announcement component;
  • Step 53 The resource information announcement component receives the resource information from the resource information perception component and announces the received information across the entire network according to certain rules.
  • Step 54 The resource information integration component receives computing power resource information, storage resource information, algorithm resource information, and network resource information between computing power nodes and other devices from each computing power node, and generates an integrated resource information table containing all resources.
  • Step 55 The resource view generation component obtains the resource information table, extracts some information, and generates a visual resource information table.
  • Step 56 The resource view generation component sends the visualized resource information table to the computing power network service layer and the orchestration management layer respectively to support the visualization of resource transactions and monitoring processes.
  • the resource information collection method in the defined computing power network scenario includes six steps, which define the entire process from resource information perception to resource view generation, improve the comprehensiveness of resource information acquisition, and ensure the accuracy of the visualized resource information required by the computing power network service layer and/or computing power network orchestration management layer.
  • the resource information collection function in the computing power network control layer may include a resource information perception component, a resource information announcement component, a resource information integration component, and a resource view generation component.
  • the computing power network package ... It includes multiple computing power gateways and multiple centralized management and control platforms (also known as multiple entities). Each computing power gateway and each centralized management and control platform contains the above four components and performs resource information collection and processing methods through the above four components.
  • the resource information sensing component corresponding to the current computing power gateway or the current centralized management and control platform can transmit the sensed resource information to the resource information integration component corresponding to the current computing power gateway or the current centralized management and control platform.
  • both computing power gateway 1 and computing power gateway 2 can obtain resource information through the resource information sensing component, and each of them transmits the acquired resource information to its respective resource information integration component.
  • the resource information sensing component corresponding to the current computing power gateway or the current centralized management and control platform can also transmit the sensed resource information to the resource information integration components corresponding to other computing power gateways or other centralized management and control platforms.
  • both computing power gateway 1 and computing power gateway 2 can obtain resource information through the resource information sensing component, computing power gateway 1 transmits the acquired resource information to the resource information integration component of computing power gateway 2, and computing power gateway 2 transmits the acquired resource information to its own resource information integration component.
  • the resource information of the computing power network can be synchronized through the computing power network control layer.
  • the resource information synchronization process of the computing power network can involve computing power gateways publishing their perceived resource information to a centralized management and control platform, or publishing it to adjacent computing power gateways, thereby achieving full network resource information synchronization.
  • the computing power network can include three deployment architectures: centralized architecture, hybrid architecture, and distributed architecture. The resource information synchronization mechanisms corresponding to these three deployment architectures share similarities.
  • the computing gateway sends the network information it perceives to the centralized management platform via a RESTful API.
  • computing gateways synchronize network information with adjacent gateways via extended routing protocols (such as CP-BGP), and the gateways then send the entire network resource information to the centralized management platform via RESTful APIs.
  • gateways synchronize resource information with each other via extended routing protocols (such as CP-BGP). It is evident that synchronizing information across different computing network architectures incurs significant network overhead, and existing resource information synchronization methods are not universally applicable to distributed, hybrid, and centralized computing networks.
  • the embodiments of this application can introduce a message queue mechanism, which can reduce the number of information announcements, reduce the consumption of network link resources, and can also uniformly complete the synchronization of resource information of the computing power network without distinguishing the specific deployment architecture of the computing power network (distributed architecture, hybrid architecture, centralized architecture).
  • the resource information synchronization method in a computing power network scenario includes a resource information announcement component and a resource information integration component in the resource information collection functional architecture. Both the computing power gateway and the centralized management and control platform integrate these two components. This method uses the broadcast-Fanout subscription pattern of the open-source general-purpose messaging library RabbitMQ for resource information synchronization.
  • RabbitMQ allows message producers to send messages to exchanges via a publish/subscribe pattern, and exchanges then distribute the messages to queues bound to those exchanges.
  • the broadcast-Fanout subscription pattern is a special form of publish/subscribe. In this pattern, the exchange distributes messages without considering routing keys, directly broadcasting the received messages to all queues bound to it. Because routing key matching is not required, Fanout exchanges can quickly distribute messages to all bound queues. Furthermore, the exchange can replicate each message, ensuring that all bound queues receive a copy.
  • the resource information announcement component can act as a message producer. First, it obtains resource information from the resource information sensing component. Then, it compares the currently sensed resource information with the resource information at the time of the last announcement. Finally, it announces resource information whose change exceeds a preset value, while not announcing resource information whose change does not exceed the preset value. For example, if the CPU utilization rate at the time of the last announcement was 20%, the currently sensed CPU utilization rate is 15%, and the preset value is 10%, then the change is 5%, which is less than 10%. Therefore, the currently sensed CPU utilization rate is not announced.
  • the resource information integration component can act as a message consumer, receiving resource information from resource information announcement components of other computing power gateways or centralized management platforms, and generating an integrated resource information table.
  • the Exchange and Queue entities in RabbitMQ can be deployed within the corresponding computing power gateway or centralized management platform where the producer resides.
  • Step 81 The resource information announcement component of computing power gateway 1 obtains the resource information perceived by computing power gateway 1 from the resource information perception component.
  • Step 82 Compare the currently perceived resource information with the value from the last announcement.
  • Step 83 Iterate through all resource information. If the change exceeds 10%, retain the information value; otherwise, delete it.
  • Step 84 If the change exceeds 10%, the resource information announcement component announces the current resource information.
  • Step 85 The resource information sharing component of the computing power gateway 2 or the centralized management and control platform receives the resource information from the computing power gateway 1.
  • computing power gateway 1 can synchronize information with computing power gateway 2/or centralized management and control platform
  • computing power gateway 2 or several management and control platforms can synchronize information with computing power gateway 1.
  • the computing power network resource information synchronization method based on RabbitMQ message queue can unify the resource information synchronization mechanism without distinguishing the computing power network deployment architecture, thereby improving the efficiency of resource information synchronization and reducing the network overhead of the information synchronization process.
  • steps in the flowcharts of the above embodiments are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the above embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.
  • this application also provides a resource information collection and processing apparatus for implementing the resource information collection and processing method described above.
  • the solution provided by this apparatus is similar to the solution described in the above method; therefore, the specific limitations in one or more resource information collection and processing apparatus embodiments provided below can be found in the limitations of the resource information collection and processing method described above, and will not be repeated here.
  • an electronic device which may be a server, and its internal structure diagram is shown in Figure 9.
  • the electronic device includes a processor, memory, input/output interfaces (I/O), and a communication interface.
  • the processor, memory, and I/O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I/O interfaces.
  • the processor of the electronic device provides computing and control capabilities.
  • the memory of the electronic device includes a non-volatile storage medium and internal memory.
  • the non-volatile storage medium stores a computer program.
  • the internal memory provides an environment for the execution of the computer program in the non-volatile storage medium.
  • the I/O interfaces of the electronic device are used for exchanging information between the processor and external devices.
  • the communication interface of the electronic device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a resource information collection and processing method.
  • FIG. 9 is only a portion of the structure relevant to the present application.
  • the block diagram is not intended to limit the electronic device to which the present application is applied.
  • the specific electronic device may include more or fewer components than those shown in the diagram, or may combine certain components, or may have different component arrangements.
  • an electronic device including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.
  • a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps in the above-described method embodiments.
  • a computer program product including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.
  • the user information including but not limited to user device information, user personal information, etc.
  • data including but not limited to data used for analysis, data stored, data displayed, etc.
  • Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc.
  • ROM read-only memory
  • MRAM magnetic random access memory
  • FRAM ferroelectric random access memory
  • PCM phase change memory
  • Volatile memory can include random access memory (RAM) or external cache memory, etc.
  • RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).
  • SRAM Static Random Access Memory
  • DRAM Dynamic Random Access Memory
  • the databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases.
  • the processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Power Sources (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本申请涉及一种资源信息收集功能架构、处理方法、电子设备和存储介质，涉及算力网络技术领域。所述功能架构包括：资源信息感知组件（201），负责与算力网络资源层连接，从算力网络资源层获取资源信息；资源信息公告组件（202），用于接收来自资源信息感知组件（201）的资源信息，并公告资源信息；资源信息整合组件（203），用于接收资源信息，并生成整合资源信息表；资源视图生成组件（204），用于基于整合资源信息表，生成可视化资源信息表，并将可视化资源信息表发送给算力网络服务层和/或算力网络编排管理层。该资源信息收集功能架构能够提升算力网络控制层的资源信息收集的全面性和规范性。

Description

资源信息收集功能架构、处理方法、电子设备和存储介质
本申请要求于2024年7月2日提交中国专利局，申请号为2024108804620，申请名称为“资源信息收集功能架构、处理方法、电子设备和存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及算力网络技术领域，特别是涉及一种资源信息收集功能架构、处理方法、电子设备、存储介质和计算机程序产品。
背景技术
相关标准中定义了算力网络的功能架构，包括资源层、控制层、服务层和编排管理层。其中，控制层通过算力路由等方式对算力网络资源层的资源信息进行收集，并将资源信息发送到算力网络服务层进行进一步处理。
根据相关标准的定义，算力网络控制层需要收集各种资源信息。并且，资源信息的收集是算力网络进行后续资源调度使用的前提，在收集各种资源信息之后，才能对各资源监控、管理、调度和应用。然而，相关技术中都未对相关标准所提的资源信息收集的方法进行研究，导致算力网络控制层不能动态掌握全网资源信息。
发明内容
基于此，有必要针对上述技术问题，提供一种能够提升算力网络控制层的资源信息收集效果的资源信息收集功能架构、处理方法、电子设备、计算机可读存储介质和计算机程序产品。
第一方面，本申请提供了一种资源信息收集功能架构，部署于算力网络控制层，包括：
资源信息感知组件，负责与算力网络资源层连接，从所述算力网络资源层获取资源信息；
资源信息公告组件，用于接收来自所述资源信息感知组件的资源信息，并公告所述资源信息；
资源信息整合组件，用于接收资源信息，并生成整合资源信息表；
资源视图生成组件，用于基于所述整合资源信息表，生成可视化资源信息表，并将所述可视化资源信息表发送给算力网络服务层和/或算力网络编排管理层。
在其中一个实施例中，其中：
所述资源信息感知组件定期或按需从所述算力网络资源层获取资源信息，并将所获取的资源信息发送给所述资源信息公告组件。
在其中一个实施例中，其中：
所述资源信息感知组件从所述算力网络资源层获取的资源信息包括：算力资源信息、网络资源信息、存储资源信息和算法资源信息中至少一种。
在其中一个实施例中，其中：
所述资源信息公告组件，用于接收来自所述资源信息感知组件的资源信息，并在全网公告所述资源信息。
在其中一个实施例中，其中：
所述资源信息整合组件接收公告的资源信息，包括来自全网各个计算节点的算力资源信息、存储资源信息、算法资源信息，以及各个计算节点与其他设备之间的网络资源信息，并生成包含所有资源信息的整合资源信息表。
在其中一个实施例中，其中：
所述资源视图生成组件基于所述整合资源信息表，提取部分信息以生成所述可视化资源信息表，并将所述可视化资源信息表发送给算力网络服务层和/或算力网络编排管理层。
第二方面，本申请还提供了一种资源信息收集处理方法。所述方法包括：
从算力网络资源层获取资源信息；
对获取到的所述资源信息进行公告；
生成整合资源信息表；
基于所述整合资源信息表，生成可视化资源信息表，并将所述可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层。
在其中一个实施例中，所述从算力网络资源层获取资源信息，对获取到的所述资源信息进行公告，包括：
所述资源信息从算力网络资源层被传输到资源信息感知组件；
所述资源信息从所述资源信息感知组件发送到资源信息公告组件；
资源信息公告组件从资源信息感知组件接收资源信息，并在全网公告所接收到的资源信息。
在其中一个实施例中，从算力网络资源层被传输到资源信息感知组件的所述资源信息包括：算力资源信息、网络资源信息、存储资源信息和算法资源信息中至少一种。
在其中一个实施例中，所述生成整合资源信息表，包括：
所述公告的资源信息被资源信息整合组件接收，由所述资源信息整合组件生成一个包含所有资源信息的整合资源信息表；
被所述资源信息整合组件接收公告的资源信息包括：来自全网各计算节点的算力资源信息、存储资源信息和算法资源信息，以及所述各个计算节点与其他设备之间的网络资源信息。
在其中一个实施例中，所述基于所述整合资源信息表，生成可视化资源信息表，并将所述可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层，包括：
所述整合资源信息表被传递给资源视图生成组件，通过所述资源视图生成组件提取部分信息生成可视化资源信息表；
通过所述资源视图生成组件将所述可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层。
第三方面，本申请还提供了一种电子设备。所述电子设备包括存储器和处理器，所述存储器存储有计算机程序，所述处理器执行所述计算机程序时实现如资源信息收集处理方法的步骤，该方法可以为：
从算力网络资源层获取资源信息；
对获取到的所述资源信息进行公告；
生成整合资源信息表；
基于所述整合资源信息表，生成可视化资源信息表，并将所述可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层。
第四方面，本申请还提供了一种计算机可读存储介质。所述计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现如资源信息收集处理方法的步骤，该方法可以为：
从算力网络资源层获取资源信息；
对获取到的所述资源信息进行公告；
生成整合资源信息表；
基于所述整合资源信息表，生成可视化资源信息表，并将所述可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层。
第五方面，本申请还提供了一种计算机程序产品。所述计算机程序产品，包括计算机程序，该计算机程序被处理器执行时实现如资源信息收集处理方法的步骤，该方法可以为：
从算力网络资源层获取资源信息；
对获取到的所述资源信息进行公告；
生成整合资源信息表；
基于所述整合资源信息表，生成可视化资源信息表，并将所述可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层。
上述资源信息收集功能架构、处理方法、电子设备、存储介质和计算机程序产品，在算力网络控制层的资源信息收集功能架构中，包括资源信息感知组件、资源信息公告组件、资源信息整合组件、资源视图生成组件四个功能组件，通过资源信息感知组件与算力网络资源层连接，并从算力网络资源层获取资源信息，并将资源信息发送至资源信息公告组件。资源信息公告组件接收资源信息后公告资源信息。资源信息整合组件可以接收公告的资源信息，并生成包含所有资源信息的整合资源信息表。资源视图生成组件可以根据整合资源信息表，从中提取所需的信息并生成可视化资源信息表，并将可视化资源信息表发送至算力网络服务层和/或算力网络编排管理层，从而实现资源监控和交易过程的可视化。上述资源信息收集功能架构可以从算力网络资源层中获取资源信息，并对获取到的资源信息进行公告，再通过资源信息整合组件将获取到的多种类型的资源信息进行整合，得到算力网络资源层对应的整合资源信息表，从而提升所收集的资源信息的全面性和规范性。资源视图生成组件可以对整合资源信息表中的信息进行提取，并得到对应的可视化资源信息表提供给算力网络服务层和/或算力网络编排管理层，进而保障了算力网络服务层和/或算力网络编排管理层所需可视化的资源信息。
附图说明
为了更清楚地说明本申请实施例或相关技术中的技术方案，下面将对实施例或相关技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为一个实施例中资源信息收集功能架构的应用环境图；
图2为一个实施例中算力网络控制层的架构示意图；
图3为一个实施例中资源信息收集处理方法的流程示意图；
图4为另一个实施例中资源信息收集处理方法的流程示意图；
图5为再一个实施例中资源信息收集处理方法的流程示意图；
图6为一个实施例中集中式架构、混合架构、分布式架构的资源信息同步的示意图；
图7为一个实施例中资源信息同步的流程示意图；
图8为另一个实施例中资源信息同步的流程示意图；
图9为一个实施例中电子设备的内部结构图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
本申请实施例提供的资源信息收集功能架构，可以应用于如图1所示的算力网络中。算力网络是一种基于网络控制面分发算力、算法、存储信息的技术，该技术再结合网络信息，可以实现以业务需求为核心，提供计算资源、网络资源、存储资源等资源的分发、关联、交易和调配，从而达到整网资源的最优化配置的目标。
根据相关标准的定义，算力网络架构包含算力网络服务层、算力网络控制层、算力网络资源层和算力网络编排管理层。其中，算力网络控制层有三个基本功能，包括资源信息收集功能、资源分配功能和网络连接调度功能。资源信息收集功能负责收集全网算力和网络资源信息，资源分配功能负责根据业务需求配置最佳算力节点，网络连接调度功能负责根据业务需求和资源分配功能模块的资源配置情况，配置从用户业务侧到算力资源侧的网络连接，以满足用户对于带宽、时延、QoS(Quality of Service，服务质量)等指标要求。
算力网络控制层需要收集各种资源信息，收集资源信息是算力网络进行后续资源分配功能、网络连接调度功能的前提条件，在动态掌握全网资源信息的情况下，才可以对各资源监控、管理、调度和应用。基于此，算力网络控制层是算力网络技术的核心，对于算力网络资源层获取资源信息并提供给算力网络服务层和/或算力网络编排管理层的过程，目前尚不存在成熟有效的解决办法。本公开方案提供了一种算力网络控制层的资源信息收集功能的实现架构和处理过程，具体提出了一种包含资源信息感知组件、资源信息公告组件、资源信息整合组件、资源视图生成组件四个功能组件的资源信息收集功能架构，同时提出了一种资源信息收集的处理过程，包括资源信息从算力网络资源层被传输到资源信息感知组件；资源信息从资源信息感知组件发送到资源信息公告组件；资源信息公告组件从资源信息感知组件接收资源信息，并在全网公告所接收到的资源信息；公告的资源信息被资源信息整合组件接收，由资源信息整合组件生成一个包含所有资源信息的整合资源信息表；整合资源信息表被传递给资源视图生成组件，通过资源视图生成组件提取部分信息生成可视化资源信息表；通过资源视图生成组件将可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层，以支持资源交易和监控过程的可视化。
其中，算力网络资源层包含的资源信息包括：算力资源信息、网络资源信息、存储资源信息和算法资源信息等。
本申请实施例中，对于算力网络控制层的性能要求包括：要求实时感知终端(已被订 阅为高移动性终端)的位置、网络状态(如网络抖动、网络延迟、网络带宽)以及计算节点的处理能力；以及要求支持智能分配和优化，将描述算力网络控制层的能力需求，明确支持业务需求所需的技术能力(组网信息收集能力、资源分配能力、资源分配能力)的详细要求。
对于算力网络控制层中的网络信息收集功能的要求包括：要求支持从算力网络资源层动态获取资源信息，包括但不限于算力资源信息、网络资源信息、存储资源信息和算法资源信息；要求支持资源信息的公告/发布，同步整个网络中不同的资源信息使用户能够掌握全局的资源情况；要求支持资源信息的管理，不仅包括全局资源信息，还可以根据用户需求生成简化的资源信息表项，以支持算力网络服务层和/或算力网络编排管理层的可视化功能。
下面以具体地实施例对本申请的技术方案以及本申请的技术方案如何解决上述技术问题进行详细说明。下面这几个具体的实施例可以相互结合，对于相同或相似的概念或过程可能在某些实施例中不再赘述。下面将结合附图，对本申请的实施例进行描述。
在一个示例性的实施例中，如图2所示，提供了一种资源信息收集功能架构，以该架构应用于图1中的算力网络控制层为例进行说明。其中，资源信息收集功能200包括资源信息感知组件201，资源信息公告组件202，资源信息整合组件203，资源视图生成组件204四个组件，下面分别对各组件进行说明。
其中，资源信息感知组件201负责与算力网络资源层连接，从算力网络资源层获取资源信息。
在一个实施例中，资源信息感知组件201负责与算力网络资源层连接，从算力网络资源层周期性或按需获取资源信息，并将所获取的资源信息发送给资源信息公告组件。
其中，资源信息感知组件201从算力网络资源层获得的资源信息包含但不限于算力资源信息、网络资源信息、存储资源信息和算法资源信息等。
其中，资源信息感知组件201可以通过通感技术感知算力网络中各节点的资源信息，通感技术可以按照预设周期或者根据其他基于需求设定的感知条件对网络资源层的资源信息进行感知。
在一些实施例中，资源信息感知组件201可以预先配置从算力网络资源层获取资源信息的感知条件，包括触发条件、周期条件、资源需求、收集间隔等参数，算力网络控制层可以根据上述参数和通感技术对算力网络资源层包含的资源信息进行收集，在每一次周期中可以收集得到满足参数的资源信息，并作为该周期内的资源信息。资源信息感知组件201可以周期性地进行多次资源收集，得到多个时间周期下的资源信息。
在一个示例中，算力网络资源层包含多种资源信息，资源信息感知组件201可以通过资源信息感知组件，对不同时间点对应的资源信息进行收集。例如，资源信息感知组件201可以确定触发条件为预设的整点时间，如早上6点整、晚上24点整等，当时间满足触发条件时，算力资源控制层通过资源信息感知组件，从算力网络资源层收集资源信息。资源信息感知组件201可以配置周期条件，例如设置一日为一个周期，即在24小时内获取的资源信息为该周期内对应的资源信息。资源信息感知组件201可以配置资源需求，例如，确定算力网络中包含的计算节点的资源信息，以及计算节点与其他设备之间的资源信息。资源信息感知组件201可以配置周期条件收集间隔，例如在一天的周期内，设置收集 间隔为1小时，则每隔一小时可以进行一次资源信息的收集。
其中，资源信息公告组件202用于接收来自资源信息感知组件201的资源信息，并公告资源信息。
在一个实施例中，资源信息公告组件202用于接收来自资源信息感知组件201的资源信息，并在全网公告资源信息。
在一些实施例中，资源信息公告组件202可以依据预设的公告策略，在全网公告资源信息。全网可以理解为包含算力网络中全部算力网关和/或集中管控平台的范围，使得算力网络中各算力网关、各集中管控平台能够获取到全网的资源信息。
在一个实施例中，资源信息公告组件202可以在算力网络中的控制实体(例如算力网关或集中管控平台)内部进行资源信息公告，也可以是在不同控制实体(算力网关或集中管控平台)之间进行资源信息公告。在控制实体内部进行资源信息公告的情况下，资源信息公告方法可以包括函数调用、消息队列、透传等，在控制实体之间的资源信息公告方法包括扩展路由协议、restful API、消息队列等方式。
在一些实施例中，资源信息公告组件202可以接收资源信息感知组件201所获取到的资源信息。针对所获取的资源信息，资源信息公告组件202可以根据公告策略，确定所需公告的资源信息，以及每种资源类型的资源信息对应的待公告字段，由此确定目标公告资源信息。资源信息公告组件202对目标公告资源信息在全网进行公告，包括在当前算力网关或当前集中管控平台内部公告，和向其他算力网关或其他集中管控平台进行外部公告。
在一个示例中，公告策略可以是针对网络资源信息，以及网络资源信息的第一至第十字段进行公告。资源信息公告组件202根据公告策略，可以将算力网络中各节点包含的网络资源信息的第一至第十字段进行公告。
在一些实施例中，资源信息公告组件202可以为目标公告资源信息配置权限，针对不同的目标公告资源信息配置不同级别的权限，例如第一级别权限至第十级别权限，权要级别越高，则保密程度越高。当查询终端查询目标公告资源信息时，需要根据查询终端的权限信息，确定是否能够查看目标公告资源信息。若目标公告资源信息为第三级别权限，则查询终端的权要信息为第三级别至第十级别。
其中，资源信息整合组件203用于接收资源信息，并生成整合资源信息表。
在一个实施例中，资源信息整合组件203用于接收公告的资源信息，并生成一个整合资源信息表。
其中，资源信息整合组件203可以接收资源信息公告组件202所公告的资源信息。由于资源信息公告组件202可以是其他算力网关或其他集中管控平台对应的资源信息公告组件，也可以是资源信息整合组件203所在的当前算力网关或当前集中管控平台的资源信息公告组件，且各资源信息公告组件均在全网公告其获得的资源信息，因此资源信息整合组件203可以接收到全网的资源信息，生成的整合资源信息表可以包含全网的资源信息。
在一些实施例中，资源信息整合组件203可以将接收到的来自全网各个计算节点的资源信息整合到一个资源信息表中，得到整合资源信息表。由于资源信息感知组件201可以获取一个时间周期内包含的多种类型的资源信息，资源信息感知组件201也可以获取不 同时间周期内包含的同一种类型的资源信息，基于此，资源信息整合组件203可以通过整合资源信息表将多个时间周期内包含的多种类型的资源信息整合到一个信息表。
在一些实施例中，资源视图生成组件204用于基于整合资源信息表，生成可视化资源信息表，并将可视化资源信息表发送给算力网络服务层和/或算力网络编排管理层。
在一个实施例中，资源视图生成组件204用于基于整合资源信息表，提取部分信息生成可视化资源信息表，并将可视化资源信息表发送给算力网络服务层和/或算力网络编排管理层。
在一些实施例中，可视化资源信息表包含的资源信息可以基于用户侧的具体业务需求确定，用户侧的不同业务需求，可以生成不同的可视化资源信息表，将不同的可视化资源信息表发送给算力网络服务层和/或算力网络编排管理层。在一个示例中，资源视图生成组件204，基于整合资源信息表可以一次性生成多个可视化资源信息表，然后将多个可视化资源信息表中的一部分发送给算力网络编排管理层，另一部分发送给算力网络服务层。在另一个示例中，资源视图生成组件204，基于整合资源信息表一次性生成一个可视化资源信息表，然后将该可视化资源信息表发送给算力网络编排管理层、网络服务层中的一者。
在一些实施例中，资源视图生成组件204可以根据用户侧的具体业务需求，从整合资源信息表中提取与该具体业务需求对应的部分信息，并基于提取出的部分信息生成可视化资源信息表。资源视图生成组件204将生成的可视化资源信息表，发送至算力网络服务层和/或算力网络编排管理层。算力网络编排管理层和算力网络服务层可以分别调用对应的程序对可视化资源信息表进行渲染处理，从而得到可视化图像，并通过可视化图像对可视化资源信息表的数据进行监控，以支持资源交易和监控过程的可视化。
上述资源信息收集功能架构中，包括资源信息感知组件201、资源信息公告组件202、资源信息整合组件203、资源视图生成组件204四个功能组件，通过资源信息感知组件201与算力网络资源层连接，并从算力网络资源层获取资源信息，并将资源信息发送至资源信息公告组件202。资源信息公告组件202接收资源信息后公告资源信息。资源信息整合组件203可以接收公告的资源信息，并生成包含所有资源信息的整合资源信息表。资源视图生成组件204可以根据整合资源信息表，从中提取所需的信息并生成可视化资源信息表，并将可视化资源信息表发送至算力网络服务层和/或算力网络编排管理层，从而实现资源监控和交易过程的可视化。上述资源信息收集功能架构可以从算力网络资源层中获取资源信息，并对获取到的资源信息进行公告，再通过资源信息整合组件203将接收到的公告的多种类型的资源信息进行整合，得到算力网络资源层对应的整合资源信息表，从而提升所收集的资源信息的全面性和规范性。资源视图生成组件204可以对整合资源信息表中的信息进行提取，并得到对应的可视化资源信息表提供给算力网络服务层和/或算力网络编排管理层，进而保障了算力网络服务层和/或算力网络编排管理层所需可视化的资源信息。
在一个示例性的实施例中，资源信息感知组件从算力网络资源层获取的资源信息包括：算力资源信息、网络资源信息、存储资源信息和算法资源信息中至少一种。
在一些实施例中，算力资源信息可以表示各计算节点的总算力、当前可用算力等算力资源信息。网络资源信息可以表示各计算节点在某个时刻的网络资源量，例如12点整时，该计算节点所分配的总带宽，以及已使用带宽等数据。存储资源信息可以表示各计算节点 在某个时刻的总存储资源量、可存储资源量和已存储资源量等。算力网络下的算法资源指的是通过分布式计算和网络资源，提供用于数据处理、分析和计算的算法集合，这些算法资源可以在云计算平台上灵活调用，实现高效、快速的计算处理。算法资源可以为机器学习算法，如支持向量机和深度神经网络。
本实施例中，资源信息感知组件可以向算力网络资源层获取多种类型的资源信息，资源信息可以包括算力资源信息、网络资源信息、存储资源信息和算法资源信息，可以增加资源信息感知组件获取到的资源信息的种类和资源量，提升所获取数据的数据维度，并提升所获取资源信息的全面性。
在一个实施例中，资源信息整合组件接收公告的资源信息，包括来自全网各个计算节点的算力资源信息、存储资源信息、算法资源信息，以及全网各个计算节点与其他设备之间的网络资源信息，并生成包含所有资源信息的整合资源信息表。
在一些实施例中，其他设备可以为与计算节点存在网络连接的各电子设备，例如个人终端、服务器等设备。
在一些实施例中，算力网络控制层通过资源信息感知组件从算力网络资源层中获取各计算节点的算力资源信息、存储资源信息、算法资源信息，以及各计算节点与其他设备之间的网络资源信息。通过资源信息整合组件将各计算节点的当前资源信息，以及各计算节点与其他设备之间的网络资源信息，存储在一个整合资源信息表中。
本实施例中，通过将全网各计算节点的算力资源信息、存储资源信息、算法资源信息，以及全网各计算节点与其他设备之间的网络资源信息整合至整合资源信息表，可以得到全网完整的资源信息，便于后续基于用户侧的业务需求生成多种多样的可视化资源信息表，以满足多种多样的业务需求。
在一个实施例中，如图3所示，提供了一种资源信息收集处理方法，以该方法应用于图1中的算力网络控制层为例进行说明。该处理方法包括：
从算力网络资源层获取资源信息；对获取到的资源信息进行公告；生成整合资源信息表；基于整合资源信息表，生成可视化资源信息表，并将可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层。
在一个实施例中，从算力网络资源层获取资源信息包括：算力资源信息、网络资源信息、存储资源信息和算法资源信息中的至少一种。
在一个实施例中，上述的对获取到的资源信息进行公告，包含对获取到的资源信息进行内部公告和外部公告。内部公告理解为在当前算力网关或当前集中管控平台内部进行资源信息公告，外部公告可以理解为是向其他算力网关或其他集中管控平台进行资源信息公告。
在一些实施例中，算力网络控制层可以从算力网络资源层获取多种多样的资源信息，并将所获取的多种多样的资源信息在全网范围内进行公告，由此算力网络中的各算力网关或集中管控平台可以接收到全网各计算节点的多种多样的资源信息，进而可以将多种多样的资源信息行整合，得到包含所有资源信息的整合资源信息表。进一步地，基于整合资源信息表中的资源信息进行信息提取，可以得到匹配于不同业务需求的可视化资源信息表，并将可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层，从而使得算力网络服务层和/或算力网络编排管理层对可视化资源信息表进行可视化，得到可视化后的 资源信息。
本实施例中上述方法可以从算力网络资源层中获取资源信息，并对获取到的资源信息进行公告，再将获取到的多种类型的资源信息进行整合，得到整合资源信息表，从而提升所收集的资源信息的全面性和规范性。资源视图生成组件可以对整合资源信息表中的信息进行提取，并得到对应的可视化资源信息表，提供给算力网络服务层和/或算力网络编排管理层，进而保障了算力网络服务层和/或算力网络编排管理层所需可视化的资源信息。
在一个实施例中，如图4所示，提供了一种资源信息收集处理方法，以该方法应用于图1中的算力网络控制层为例进行说明，该资源信息收集处理方法包括步骤S402～S412，具体说明如下：
步骤S402，资源信息从算力网络资源层被传输到资源信息感知组件。
其中，资源信息感知组件可以通过通感技术感知算力网络中各节点的资源信息，通感技术可以按照预设周期或者预设资源类型对资源信息进行感知。
在一些实施例中，算力网络控制层可以预先配置感知资源信息的触发条件、周期条件、资源需求、收集间隔等参数，算力网络控制层可以根据上述参数和通感技术对算力网络资源层包含的资源信息进行收集，在每一次周期中可以收集得到满足参数的资源信息，并作为该周期内的资源信息。算力网络控制层可以周期性地进行多次资源收集，得到多个时间周期下的资源信息。
在一个实施例中，从算力网络资源层被传输到资源信息感知组件的资源信息，包括：算力资源信息、网络资源信息、存储资源信息和算法资源信息中的至少一种。
步骤S404，资源信息从资源信息感知组件发送到资源信息公告组件。
在一些实施例中，算力网络控制层可以通过资源信息公告组件接收资源信息感知组件所获取到的资源信息。
步骤S406，资源信息公告组件从资源信息感知组件接收资源信息，并在全网公告所接收到的资源信息。
在一些实施例中，针对每一周期所获取的资源信息，资源信息公告组件可以根据公告策略，确定所需公告的资源类型对应的资源信息，以及每种资源类型的资源信息对应的公告字段，并确定为目标公告资源信息。资源信息公告组件对目标公告资源信息在全网进行公告。
步骤S408，公告的资源信息被资源信息整合组件接收，由资源信息整合组件生成一个包含所有资源信息的整合资源信息表。
在一些实施例中，算力网络控制层可以通过资源信息整合组件接收资源信息公告组件公告的各节点的资源内信息。
资源信息整合组件可以将接收到的各资源信息整合到整合资源信息表中。由于资源信息感知组件可以获取一个时间周期内包含的多种类型的资源信息，资源信息感知组件也可以获取多个时间周期内包含的同一种类型的资源信息，基于此，资源信息整合组件可以通过整合资源信息表将多个时间周期内包含的多种类型的资源信息一一整合，得到整合资源信息表。
在一个实施例中，被资源信息整合组件接收的资源信息，包括：来自全网各计算节点的算力资源信息、存储资源信息和算法资源信息，以及全网各计算节点与其他设备之间的 网络资源信息，由资源信息整合组件生成一个包含所有资源的整合资源信息表。
本实施例中，资源信息感知组件可以向算力网络资源层获取多种类型的资源信息，资源信息可以包括算力资源信息、网络资源信息、存储资源信息和算法资源信息，可以增加资源信息感知组件获取到的资源信息的种类和资源量，提升所获取数据的数据维度，并提升所获取资源信息的全面性。
步骤S410，整合资源信息表被传递给资源视图生成组件，通过资源视图生成组件提取部分信息生成可视化资源信息表。
在一些实施例中，资源视图生成组件可以根据接收到的资源筛选指令，从整合资源信息表中提取部分信息，并将部分信息全部转换为可视化资源信息表。
步骤S412，通过资源视图生成组件将可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层，以支持资源交易和监控过程的可视化。
在一些实施例中，资源视图生成组件将资源筛选指令对应的可视化资源信息表，发送至算力网络服务层和/或算力网络编排管理层。算力网络编排管理层和算力网络服务层可以分别调用对应的程序对可视化资源信息表进行可视化处理，从而得到可视化图像，并通过可视化图像对可视化资源信息表的数据进行监控。
本实施例中，上述方法可以从算力网络资源层中获取资源信息，并对获取到的资源信息进行全网公告，再将获取到的多种类型的资源信息进行整合，得到整合资源信息表，从而提升所收集的资源信息的全面性和规范性。资源视图生成组件可以对整合资源信息表中的部分信息进行提取，并得到部分信息对应的可视化资源信息表，提供给算力网络服务层和/或算力网络编排管理层，进而保障了算力网络服务层和/或算力网络编排管理层所需可视化的关键资源信息。
在另一个实施例中，如图5所示，详细说明一种资源信息收集处理方法的具体实现过程，包括：
步骤51，资源信息感知组件从算力网络资源层获取资源信息，包括但不限于算力资源信息，网络资源信息，存储资源信息和算法资源信息等。
步骤52，资源信息感知组件将所获取的资源信息发送给资源信息公告组件；
步骤53，资源信息公告组件收到来自资源信息感知组件的资源信息，按照一定规则将接收的信息在全网进行公告。
步骤54，资源信息整合组件接收来自各算力节点的算力资源信息、存储资源信息、算法资源信息，以及算力节点与其它设备间的网络资源信息，生成包含所有资源的整合资源信息表。
步骤55，资源视图生成组件获取资源信息表，提取部分信息生成可视化资源信息表。
步骤56，资源视图生成组件将可视化资源信息表分别发送给算力网络服务层和编排管理层，用于支撑资源交易和监控过程的可视化。
本实施例中，通过定义的算力网络场景下的资源信息收集方法所包含的6个步骤，定义了从资源信息感知到资源视图生成的全过程，提升获取资源信息的全面性，以及保障了算力网络服务层和/或算力网络编排管理层所需可视化的资源信息的准确性。
在一个实施例中，算力网络控制层中资源信息收集功能可以包括资源信息感知组件、资源信息公告组件、资源信息整合组件和资源视图生成组件。在一个示例中，算力网络包 含多个算力网关和多个集中管控平台(也称为多个实体)，每个算力网关以及每个集中管控平台均包含上述四个组件，并通过上述四个组件执行资源信息收集处理方法。
在一个实施例中，当前算力网关或当前集中管控平台对应的资源信息感知组件，可以将感知到的资源信息传递给当前算力网关或当前集中管控平台对应的资源信息整合组件。例如，算力网关1和算力网关2均可以通过资源信息感知组件获取资源信息，算力网关1和算力网关2分别将各自获取的资源信息传递给各自的资源信息整合组件。同时，当前算力网关或当前集中管控平台对应的资源信息感知组件，还可以将感知到的资源信息传递给其他算力网关或其他集中管控平台对应的资源信息整合组件。例如，算力网关1和算力网关2均可以通过资源信息感知组件获取资源信息，算力网关1将获取的资源信息传递给算力网关2的资源信息整合组件，算力网关2将获取的资源信息传递给算力网关2的资源信息整合组件。
在一个示例性的实施例中，算力网络的资源信息可以通过算力网络控制层进行同步。相关技术中，算力网络的资源信息同步过程，可以是算力网关将各自感知到的资源信息发布到集中管控平台，或者发布至相邻的算力网关，从而实现全网资源信息同步。如图6所示，算力网络可以包含三种部署架构，具体为集中式架构、混合式架构、分布式架构，上述三种部署架构对应的资源信息同步机制存在同。
在集中式架构的算力网络中，算力网关通过restful API将所感知到的算网信息发送给集中管控平台；在混合式架构的算力网络中，算力网关之间通过扩展路由协议(例如CP-BGP)将算网信息同步给相邻的算力网关，算力网关再通过restful API将全网资源信息发送给集中管控平台；在分布式架构的算力网络中，网关间通过扩展路由协议(例如CP-BGP)同步资源信息。可知，在通过不同架构的算力网络进行信息同步时，网络开销大，现有资源信息同步方法无法通用适配分布式架构、混合式架构、集中式架构算力网络。
基于此，本申请实施例可以引入消息队列机制，可以减少信息公告数量、减少对网络链路资源的消耗，还可以在不区分算力网络具体的部署架构(分布式架构、混合式架构、集中式架构)的情况下，使用该方法统一完成算力网络的资源信息同步。
如图7所示，算力网络场景下的资源信息同步方法，包含资源信息收集功能架构中的资源信息公告组件和资源信息整合组件。算力网关和集中管控平台均集成了这两个组件。该方法采用开源通用消息传递库RabbitMQ的广播-Fanout订阅模式进行资源信息的同步。
其中，RabbitMQ可以通过发布/订阅模式使得消息生产者发送消息至交换机(Exchange)，并通过交换机(Exchange)将消息分发到绑定至交换机的队列(Queue)中。广播-Fanout订阅模式是发布/订阅模式的一种特殊形式，广播-Fanout订阅模式下，交换机在分发消息时不考虑路由键，直接将接收到的消息广播到所有绑定到它的队列。由于不需要进行路由键匹配，Fanout交换机可以快速地将消息分发到所有绑定的队列。并且，交换机可以复制每条消息，确保所有绑定的队列都能接收到一份消息副本。
具体地，资源信息公告组件可以作为消息生产者(Producer)，首先，从资源信息感知组件获取资源信息；然后，对比当前感知到的资源信息和上一次公告时的资源信息；最后，公告变动度超过预设值的资源信息，对于变动度不超过预设值的资源信息不进行公告。例如，上一次公告时的CPU占有率为20％，当前感知到的CPU占有率为15％，并且预设值为10％，则可知变动度为5％，且小于10％，因此，不公告当前感知到的CPU占有率。
资源信息整合组件可以作为消息消费者(Consumer)，接收其它的算力网关或集中管控平台的资源信息公告组件发出的资源信息，生成整合资源信息表。RabbitMQ中的(交换机Exchange)和队列(queue)实体可以部署在相应的生产者(Producer)所在的算力网关或者集中管控平台中。
在一个实施例中，如图8所示，将采用该资源信息同步机制的流程介绍如下：
步骤81，算力网关1的资源信息公告组件从资源信息感知组件获取算力网关1所感知到的资源信息。
步骤82，对比当前感知到的资源信息和上一次公告时的值。
步骤83，遍历所有资源信息，如果变动超过10％，则保留该信息值，否则删除。
步骤84，在变动超过10％的情况下，资源信息公告组件公告当前资源信息。
步骤85，算力网关2或集中管控平台的资源信息共享组件接收算力网关1的资源信息。
可选地，不同的算力网关和集中管控平台之间可以互相进行信息同步，即算力网关1可以向算力网关2/或集中管控平台进行信息同步，算力网关2或几种管控平台可以和向算力网关1进行信息同步。
本实施例中，通过基于消息队列RabbitMQ的算力网络资源信息同步方法，可以在不区分算力网络部署架构的情况下，完成资源信息同步机制的统一，从而提高资源信息同步效率，降低信息同步过程的网络开销。
应该理解的是，虽然如上的各实施例所涉及的流程图中的各个步骤按照箭头的指示依次显示，但是这些步骤并不是必然按照箭头指示的顺序依次执行。除非本文中有明确的说明，这些步骤的执行并没有严格的顺序限制，这些步骤可以以其它的顺序执行。而且，如上的各实施例所涉及的流程图中的至少一部分步骤可以包括多个步骤或者多个阶段，这些步骤或者阶段并不必然是在同一时刻执行完成，而是可以在不同的时刻执行，这些步骤或者阶段的执行顺序也不必然是依次进行，而是可以与其它步骤或者其它步骤中的步骤或者阶段的至少一部分轮流或者交替地执行。
基于同样的发明构思，本申请实施例还提供了一种用于实现上述所涉及的资源信息收集处理方法的资源信息收集处理装置。该装置所提供的解决问题的实现方案与上述方法中所记载的实现方案相似，故下面所提供的一个或多个资源信息收集处理装置实施例中的具体限定可以参见上文中对于资源信息收集处理方法的限定，在此不再赘述。
在一个示例性的实施例中，提供了一种电子设备，该电子设备可以是服务器，其内部结构图可以如图9所示。该电子设备包括处理器、存储器、输入/输出接口(Input/Output，简称I/O)和通信接口。其中，处理器、存储器和输入/输出接口通过系统总线连接，通信接口通过输入/输出接口连接到系统总线。其中，该电子设备的处理器用于提供计算和控制能力。该电子设备的存储器包括非易失性存储介质和内存储器。该非易失性存储介质存储有计算机程序。该内存储器为非易失性存储介质中计算机程序的运行提供环境。该电子设备的输入/输出接口用于处理器与外部设备之间交换信息。该电子设备的通信接口用于与外部的终端通过网络连接通信。该计算机程序被处理器执行时以实现一种资源信息收集处理方法。
本领域技术人员可以理解，图9中示出的结构，仅仅是与本申请方案相关的部分结构 的框图，并不构成对本申请方案所应用于其上的电子设备的限定，具体的电子设备可以包括比图中所示更多或更少的部件，或者组合某些部件，或者具有不同的部件布置。
在一个示例性的实施例中，还提供了一种电子设备，包括存储器和处理器，存储器中存储有计算机程序，该处理器执行计算机程序时实现上述各方法实施例中的步骤。
在一个示例性的实施例中，提供了一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器执行时实现上述各方法实施例中的步骤。
在一个示例性的实施例中，提供了一种计算机程序产品，包括计算机程序，该计算机程序被处理器执行时实现上述各方法实施例中的步骤。
需要说明的是，本申请所涉及的用户信息(包括但不限于用户设备信息、用户个人信息等)和数据(包括但不限于用于分析的数据、存储的数据、展示的数据等)，均为经用户授权或者经过各方充分授权的信息和数据，且相关数据的收集、使用和处理需要符合相关规定。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程，是可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一非易失性计算机可读取存储介质中，该计算机程序在执行时，可包括如上述各方法的实施例的流程。其中，本申请所提供的各实施例中所使用的对存储器、数据库或其它介质的任何引用，均可包括非易失性和易失性存储器中的至少一种。非易失性存储器可包括只读存储器(Read-Only Memory，ROM)、磁带、软盘、闪存、光存储器、高密度嵌入式非易失性存储器、阻变存储器(ReRAM)、磁变存储器(Magnetoresistive Random Access Memory，MRAM)、铁电存储器(Ferroelectric Random Access Memory，FRAM)、相变存储器(Phase Change Memory，PCM)、石墨烯存储器等。易失性存储器可包括随机存取存储器(Random Access Memory，RAM)或外部高速缓冲存储器等。作为说明而非局限，RAM可以是多种形式，比如静态随机存取存储器(Static Random Access Memory，SRAM)或动态随机存取存储器(Dynamic Random Access Memory，DRAM)等。本申请所提供的各实施例中所涉及的数据库可包括关系型数据库和非关系型数据库中至少一种。非关系型数据库可包括基于区块链的分布式数据库等，不限于此。本申请所提供的各实施例中所涉及的处理器可为通用处理器、中央处理器、图形处理器、数字信号处理器、可编程逻辑器、基于量子计算的数据处理逻辑器等，不限于此。
以上实施例的各技术特征可以进行任意的组合，为使描述简洁，未对上述实施例中的各个技术特征所有可能的组合都进行描述，然而，只要这些技术特征的组合不存在矛盾，都应当认为是本说明书记载的范围。
以上所述实施例仅表达了本申请的几种实施方式，其描述较为具体和详细，但并不能因此而理解为对本申请专利范围的限制。应当指出的是，对于本领域的普通技术人员来说，在不脱离本申请构思的前提下，还可以做出若干变形和改进，这些都属于本申请的保护范围。因此，本申请的保护范围应以所附权利要求为准。

Claims (16)

  1. 一种资源信息收集功能架构，部署于算力网络控制层，包括：
    资源信息感知组件，负责与算力网络资源层连接，从所述算力网络资源层获取资源信息；
    资源信息公告组件，用于接收来自所述资源信息感知组件的资源信息，并公告所述资源信息；
    资源信息整合组件，用于接收资源信息，并生成整合资源信息表；
    资源视图生成组件，用于基于所述整合资源信息表，生成可视化资源信息表，并将所述可视化资源信息表发送给算力网络服务层和/或算力网络编排管理层。
  2. 如权利要求1所述的资源信息收集功能架构，其中，
    所述资源信息感知组件定期或按需从所述算力网络资源层获取资源信息，并将所获取的资源信息发送给所述资源信息公告组件。
  3. 如权利要求1所述的资源信息收集功能架构，其中，
    所述资源信息感知组件从所述算力网络资源层获取的资源信息包括：算力资源信息、网络资源信息、存储资源信息和算法资源信息中至少一种。
  4. 如权利要求1所述的资源信息收集功能架构，其中，
    所述资源信息公告组件，用于接收来自所述资源信息感知组件的资源信息，并在全网公告所述资源信息。
  5. 如权利要求1所述的资源信息收集功能架构，其中，
    所述资源信息整合组件接收公告的资源信息，包括来自全网各个计算节点的算力资源信息、存储资源信息、算法资源信息，以及所述各个计算节点与其他设备之间的网络资源信息，并生成包含所有资源信息的整合资源信息表。
  6. 如权利要求1所述的资源信息收集功能架构，其中，
    所述资源视图生成组件基于所述整合资源信息表，提取部分信息以生成所述可视化资源信息表，并将所述可视化资源信息表发送给算力网络服务层和/或算力网络编排管理层。
  7. 如权利要求1所述的资源信息收集功能架构，其中，
    所述资源信息公告组件还用于为所要公告的资源信息配置权限，针对不同的所要公告的资源信息配置不同级别的权限。
  8. 一种资源信息收集处理方法，包括：
    从算力网络资源层获取资源信息；
    对获取到的所述资源信息进行公告；
    生成整合资源信息表；
    基于所述整合资源信息表，生成可视化资源信息表，并将所述可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层。
  9. 如权利要求8所述的资源信息收集处理方法，其中，所述从算力网络资源层获取资源信息，对获取到的所述资源信息进行公告，包括：
    所述资源信息从算力网络资源层被传输到资源信息感知组件；
    所述资源信息从所述资源信息感知组件发送到资源信息公告组件；
    资源信息公告组件从资源信息感知组件接收资源信息，并在全网公告所接收到的资源 信息。
  10. 如权利要求8所述的资源信息收集处理方法，其中，
    从算力网络资源层被传输到资源信息感知组件的所述资源信息包括：算力资源信息、网络资源信息、存储资源信息和算法资源信息中至少一种。
  11. 如权利要求9所述的资源信息收集处理方法，其中，所述生成整合资源信息表，包括：
    所述公告的资源信息被资源信息整合组件接收，由所述资源信息整合组件生成一个包含所有资源信息的整合资源信息表；
    被所述资源信息整合组件接收公告的资源信息包括：来自全网各个计算节点的算力资源信息、存储资源信息和算法资源信息，以及所述各个计算节点与其他设备之间的网络资源信息。
  12. 如权利要求8或11所述的资源信息收集处理方法，其中，所述基于所述整合资源信息表，生成可视化资源信息表，并将所述可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层，包括：
    所述整合资源信息表被传递给资源视图生成组件，通过所述资源视图生成组件提取部分信息生成可视化资源信息表；
    通过所述资源视图生成组件将所述可视化资源信息表发送到算力网络服务层和/或算力网络编排管理层。
  13. 如权利要求8所述的资源信息收集处理方法，其中，所述对获取到的所述资源信息进行公告，包括：
    在当前算力网关或当前集中管控平台内部进行所述资源信息公告；
    向其他算力网关或其他集中管控平台进行所述资源信息公告。
  14. 一种电子设备，包括存储器和处理器，所述存储器存储有计算机程序，其中，所述处理器执行所述计算机程序时实现权利要求8至13中任一项所述的资源信息收集处理方法的步骤。
  15. 一种计算机可读存储介质，其上存储有计算机程序，其中，所述计算机程序被处理器执行时实现权利要求8至13中任一项所述的资源信息收集处理方法的步骤。
  16. 一种计算机程序产品，包括计算机程序，其中，该计算机程序被处理器执行时实现权利要求8至13中任一项所述的资源信息收集处理方法的步骤。
PCT/CN2024/127412 2024-07-02 2024-10-25 资源信息收集功能架构、处理方法、电子设备和存储介质 Pending WO2026007281A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202410880462.0A CN121277661A (zh) 2024-07-02 2024-07-02 资源信息收集功能架构、处理方法、电子设备和存储介质
CN202410880462.0 2024-07-02

Publications (1)

Publication Number Publication Date
WO2026007281A1 true WO2026007281A1 (zh) 2026-01-08

Family

ID=98241623

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2024/127412 Pending WO2026007281A1 (zh) 2024-07-02 2024-10-25 资源信息收集功能架构、处理方法、电子设备和存储介质

Country Status (2)

Country Link
CN (1) CN121277661A (zh)
WO (1) WO2026007281A1 (zh)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110035391A1 (en) * 2009-08-06 2011-02-10 Sap Ag Uniform Data Model and API for Representation and Processing of Semantic Data
CN115002681A (zh) * 2021-03-02 2022-09-02 中国移动通信有限公司研究院 一种算力感知网络及其使用方法、存储介质

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110035391A1 (en) * 2009-08-06 2011-02-10 Sap Ag Uniform Data Model and API for Representation and Processing of Semantic Data
CN115002681A (zh) * 2021-03-02 2022-09-02 中国移动通信有限公司研究院 一种算力感知网络及其使用方法、存储介质

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
ANONYMOUS: "Computing Awareness Network (CAN) Technical White Paper (2021 Edition))", CHINA MOBILE COMMUNICATION LTD., RESEARCH INSTITUTE, 1 January 2021 (2021-01-01), XP093386061, Retrieved from the Internet <URL:https://13115299.s21i.faiusr.com/61/1/ABUIABA9GAAgqJSijQYoxc_jqQM.pdf> *
LIU JUNLIN, SUN YUKUN, SU JUNQI, LI ZHAOJIANG, ZHANG XING, LEI BO, WANG WENBO: "Computing Power Network: A Testbed and Applications with Edge Intelligence", IEEE INFOCOM 2022 - IEEE CONFERENCE ON COMPUTER COMMUNICATIONS WORKSHOPS (INFOCOM WKSHPS), IEEE, 2 May 2022 (2022-05-02) - 2022-05-05, pages 1 - 2, XP093386059, DOI: 10.1109/infocomwkshps54753.2022.9798112 *

Also Published As

Publication number Publication date
CN121277661A (zh) 2026-01-06

Similar Documents

Publication Publication Date Title
CN111754216B (zh) 一种支付链路自动切换方法
CN110019240B (zh) 一种业务数据交互方法、装置及系统
US11321139B2 (en) Streaming traffic pattern for public cloud auto scaling
Naas et al. IoT data replication and consistency management in fog computing
CN109995669B (zh) 分布式限流方法、装置、设备及可读存储介质
CN109639775A (zh) 全局单调递增id生成方法、装置、系统及设备
Rossigneux et al. A generic and extensible framework for monitoring energy consumption of OpenStack clouds
US10505863B1 (en) Multi-framework distributed computation
CN114710571A (zh) 数据包处理系统
CN115134373A (zh) 数据同步方法、装置、存储介质及电子设备
CN116775420A (zh) 基于Flink流计算的信创云平台资源展示和预警方法及系统
CN114610765B (zh) 流计算方法、装置、设备及存储介质
CN119576574A (zh) 跨地域算力协同系统及方法
Pham et al. Towards an Elastic Fog‐Computing Framework for IoT Big Data Analytics Applications
EP2674876A1 (en) Streaming analytics processing node and network topology aware streaming analytics system
CN117763020A (zh) 数据传输方法、装置、设备和介质
CN109525590A (zh) 数据包的传输方法及装置
WO2026007281A1 (zh) 资源信息收集功能架构、处理方法、电子设备和存储介质
CN108111630A (zh) 一种Zookeeper集群系统及其连接方法和系统
CN116743558B (zh) 并发流量监控方法、装置、终端设备以及存储介质
Zhao et al. A real-time processing system for massive traffic sensor data
Kotov et al. Layered Architecture for RSDP V3. 0: Modular Distributed Consensus and Coordination
Di Stefano et al. Scalable and configurable monitoring system for cloud environments
CN116346663B (zh) 针对容器集群的指标采集方法和装置
CN121239684A (zh) 一种‌基于算网条件编排容器资源的方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 24946147

Country of ref document: EP

Kind code of ref document: A1