Centralized and distributed computing network decision-making method, device, equipment and medium

Abstract

The invention provides a centralized and distributed network calculation decision-making method and device, equipment and a medium. The method comprises the following steps: acquiring a time delay demand and a computing power service resource pool demand of a user according to a service request of the user; if the time delay demand indicates that the network time delay requirement of the service of the user is smaller than a first threshold value, selecting a target resource pool for the service of the user from a plurality of computing power service resource pools according to the computing power service resource pool demand; according to the target resource pool, obtaining a first service change time delay of the centralized architecture and a second service change time delay of the distributed architecture; and according to the time delay demand, the first business change time delay of the centralized architecture and the second business change time delay of the distributed architecture, confirming that the business of the user is provided with the centralized architecture or the distributed architecture. According to the method, the network calculation decision-making efficiency and the network performance can be improved.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to a method, device, equipment and medium for combining centralized and distributed computing network decision-making. Background Art

[0002] With the rapid development of 5G and the cloud era, thousands of industries have more stringent requirements for computing resources and network services. Computing power networks can realize on-demand scheduling of computing resources and provide intelligent network services.

[0003] In current research, computing network architectures are divided into centralized and distributed types: centralized is characterized by the advantages of global resource view, global orchestration, and centralized and unified control. The computing network information and computing power routing decisions are uniformly completed by the orchestration and control layer. Global resources are centralized and controllable, but the request response speed needs to be sacrificed; the distributed characteristic is that global resources cannot be controlled, but the routing convergence is fast and the service response speed is also faster. The computing power information notification and path calculation of the current computing network adopt a purely centralized or purely distributed method. Centralized and distributed have different characteristics and their own advantages and disadvantages. The computing network decision-making efficiency is low when used separately, and there is no good way to use them in a fusion manner, which leads to problems such as low computing network decision-making efficiency and poor network performance.

[0004] Therefore, the present application proposes a network decision-making method that can improve network decision-making efficiency and network performance. Summary of the invention

[0005] The present application provides a centralized and distributed network computing decision-making method, device, equipment and medium to solve the problem of low efficiency of network computing decision-making in the prior art.

[0006] In a first aspect, the present application provides a method for combining centralized and distributed computing network decision making, including:

[0007] According to the user's service request, the user's latency requirement and computing power service resource pool requirement are obtained, wherein the latency requirement is used to indicate the network latency requirement of the user's service, and the computing power service resource pool requirement is used to indicate the user's selection of the computing power service resource pool;

[0008] If the latency requirement indicates that the network latency requirement of the user's service is less than a first threshold, selecting a target resource pool for the user's service from the plurality of computing power service resource pools according to the computing power service resource pool requirement;

[0009] According to the target resource pool, obtaining a first service change delay of the centralized architecture and a second service change delay of the distributed architecture;

[0010] According to the delay requirement, the first service change delay of the centralized architecture and the second service change delay of the distributed architecture, it is determined whether a centralized architecture or a distributed architecture is provided for the user's service.

[0011] In a possible implementation, selecting a target resource pool for a user's business from a plurality of computing power service resource pools according to the computing power service resource pool requirement includes:

[0012] If the computing power service resource pool requirement indicates that the user has selected a candidate resource pool from among the multiple computing power service resource pools, the candidate resource pool is used as the target resource pool;

[0013] If the computing power service resource pool demand indicates that the user has not selected a candidate resource pool from the multiple computing power service resource pools, then the computing power service resource pool whose current resource utilization is lower than a second threshold is selected from the multiple computing power service resource pools as the target resource pool, wherein the second threshold includes multiple threshold numbers, and the number of the threshold numbers is the same as the number of the computing power service resource pools.

[0014] In a possible implementation, acquiring, according to the target resource pool, a first service change delay of the centralized architecture and a second service change delay of the distributed architecture includes:

[0015] Obtain the service initialization delay according to the delay from the user equipment to the first operator network access device, the delay from the first operator network access device to the second operator network access device, the delay from the second operator network access device to the resource pool gateway device, and the delay from the second operator network access device to the target resource pool;

[0016] Obtaining a first service change delay of the centralized architecture according to the service initialization delay, and the time for computing service selection, path calculation, and tunnel delivery at the orchestration layer after the service change;

[0017] According to the service initialization delay, as well as the time for selecting the computing service and the time for path calculation after the computing power information is changed, the second service change delay of the distributed architecture is obtained.

[0018] In a possible implementation, the step of confirming whether to provide a centralized architecture or a distributed architecture to a service of a user according to the delay requirement, the first service change delay of the centralized architecture, and the second service change delay of the distributed architecture includes:

[0019] If the first threshold is greater than or equal to the first service change delay, confirming that a centralized architecture is provided for the user's service;

[0020] If the first threshold is greater than or equal to the second service change delay and less than the first service change delay, it is confirmed that a distributed architecture is provided for the user's service, wherein the first service change delay is greater than the second service change delay.

[0021] In a possible implementation, if the latency requirement indicates that the user's service has no network latency requirement, and the computing power service resource pool requirement indicates that the user has not selected a candidate resource pool, the method further includes:

[0022] Acquiring whether the user distinguishes business service categories according to the business request, wherein the business service categories are divided according to whether the business complies with a centralized architecture or a distributed architecture;

[0023] If the user does not distinguish the business service category, the computing power service resource pool whose current resource utilization rate is lower than a third threshold is selected from the multiple computing power service resource pools as the target resource pool, wherein the third threshold includes multiple threshold numbers, and the number of the threshold numbers is the same as the number of the computing power service resource pools;

[0024] The centralized architecture or the distributed architecture provided to the user's service is determined according to the positional relationship between the user's service access point and the target resource pool.

[0025] In a possible implementation, the determining, based on the positional relationship between the service access point of the user and the target resource pool, whether to provide the user with a centralized architecture or a distributed architecture includes:

[0026] If the service access point and the target resource pool belong to the same region, confirming that the service provided to the user is provided with a distributed architecture, the same region includes a provincial region or a prefecture-level region;

[0027] If the service access point and the target resource pool do not belong to the same area, and the number of the target resource pools within a preset distance range from the service access point is greater than or equal to a preset number value, confirming that a distributed architecture is provided for the user's service;

[0028] If the service access point and the target resource pool do not belong to the same area, the number of the target resource pools within a preset distance range from the service access point is less than a preset number value, and the service access point is located in a delay-sensitive area, it is confirmed that a distributed architecture is provided for the user's service.

[0029] In a possible implementation manner, if the user distinguishes the business service category, the method further includes:

[0030] If the business service category selected by the user is global control type, it is confirmed that a centralized architecture is provided for the user's business;

[0031] If the business service category selected by the user is delay-sensitive, it is confirmed that a distributed architecture is provided for the user's business.

[0032] In a second aspect, the present application provides a computing network decision-making device combining centralized and distributed computing networks, including:

[0033] A first acquisition module is used to acquire the user's latency requirement and computing power service resource pool requirement according to the user's service request, wherein the latency requirement is used to indicate the network latency requirement of the user's service, and the computing power service resource pool requirement is used to indicate the user's selection of the computing power service resource pool;

[0034] A determination module, configured to select a target resource pool for the user's service from a plurality of computing power service resource pools according to the computing power service resource pool requirement if the latency requirement indicates that the network latency requirement of the user's service is less than a first threshold;

[0035] A second acquisition module, configured to acquire, according to the target resource pool, a first service change delay of the centralized architecture and a second service change delay of the distributed architecture;

[0036] The allocation module is used to confirm whether to provide a centralized architecture or a distributed architecture for the user's service according to the delay requirement, the first service change delay of the centralized architecture and the second service change delay of the distributed architecture.

[0037] In a third aspect, the present application provides a computing network decision device combining centralized and distributed computing, including: at least one processor and a memory;

[0038] The memory stores computer-executable instructions;

[0039] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the above-mentioned method for combining centralized and distributed computing network decision-making.

[0040] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-mentioned method for combining centralized and distributed computing network decision-making.

[0041] The present application provides a method, apparatus, device and medium for combining centralized and distributed computing network decision-making, which obtains the user's latency requirement and computing power service resource pool requirement according to the user's business request, wherein the latency requirement is used to indicate the network latency requirement of the user's business, and the computing power service resource pool requirement is used to indicate the user's selection of the computing power service resource pool; if the latency requirement indicates that the network latency requirement of the user's business is less than a first threshold, then according to the computing power service resource pool requirement, a target resource pool is selected for the user's business from multiple computing power service resource pools; according to the target resource pool, the first business change delay of the centralized architecture and the second business change delay of the distributed architecture are obtained; according to the latency requirement, the first business change delay of the centralized architecture and the second business change delay of the distributed architecture, it is confirmed whether a centralized architecture or a distributed architecture is provided for the user's business.

[0042] In the above method, it is first confirmed whether the user has made a demand, including latency demand and computing power service resource pool demand. If there is a latency demand, it is further confirmed whether the user has specified a computing power service resource pool based on the user's computing power service resource pool demand, so as to confirm the target resource pool that provides services to the user among multiple computing power service resource pools. After the target resource pool is determined, the first service change delay of the centralized architecture and the second service change delay of the distributed architecture are confirmed based on the participation of the target resource pool in the service. The user's latency demand is compared with the first service change delay and the second service change delay, and the appropriate architecture is confirmed in the centralized architecture and the distributed architecture; the centralized architecture and the distributed architecture are reasonably allocated in consideration of the user's needs and the ability of the resource pool to provide services, thereby improving the efficiency of computing network decision-making, while ensuring the reasonable allocation of network resources and improving network performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0044] Figure 1 A schematic diagram of a scenario combining centralized and distributed computing network decision-making provided in an embodiment of the present application;

[0045] Figure 2 A schematic diagram of a process of combining centralized and distributed network computing decision-making method provided in an embodiment of the present application Figure 1 ;

[0046] Figure 3A schematic diagram of a process of combining centralized and distributed network computing decision-making method provided in an embodiment of the present application Figure 2 ;

[0047] Figure 4 A schematic diagram of a process of combining centralized and distributed network computing decision-making method provided in an embodiment of the present application Figure 3 ;

[0048] Figure 5 A schematic diagram of a process of combining centralized and distributed network computing decision-making method provided in an embodiment of the present application Figure 4 ;

[0049] Figure 6 A diagram of a computing network decision device combining centralized and distributed computing networks provided by an embodiment of the present invention;

[0050] Figure 7 A hardware schematic diagram of a centralized and distributed computing network decision-making device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] Computing network is a new type of information infrastructure that allocates and flexibly schedules computing resources, storage resources, and network resources between clouds, networks, and edges according to business needs. According to the working mode of the network, the computing network architecture can be divided into centralized architecture and distributed architecture. The centralized architecture is a computer network composed of a powerful host device connected to remote terminal devices through a communication system to serve remote terminals in a time-sharing manner; the distributed architecture is a computer network composed of interconnected computer systems with independent working functions.

[0053] Therefore, the characteristic of a centralized architecture is that it can grasp the overall situation, collect all required resources and allocate appropriate resource pools to users, while the characteristic of a distributed architecture is that it can quickly allocate appropriate resource pools based on user needs; the existing way of allocating resource pools generally separates centralized architecture and distributed architecture, and there is no computing network decision-making method that combines centralized architecture and distributed architecture, resulting in poor network decision-making efficiency.

[0054] Therefore, this application proposes a network computing decision-making method that can improve the efficiency of network computing decision-making based on the architectural characteristics and specific business scenario requirements.

[0055] The following describes the implementation process of a centralized and distributed computing network decision-making method proposed in this application in conjunction with the accompanying drawings and specific embodiments.

[0056] Figure 1 A schematic diagram of a scenario combining centralized and distributed computing network decision-making provided in an embodiment of the present application. Figure 1 As shown, the system includes: user equipment 101, network node 102, computing node 103, orchestration layer 104 and business system 105; wherein, network node 102 includes first operator network access device 1021, second operator network access device 1022 and central network device 1023, computing node 103 includes resource pool gateway device 1031, and each device has time delay during communication;

[0057] The user can issue a service request to the orchestration layer 104 through the service system 105, including mandatory information and non-mandatory information. The mandatory information includes the province where the user access point is located and the attributes of the user's requested service. The non-mandatory information includes latency requirements and computing power service resource pool requirements. The user access point is the interface between the user device 101 and the first operator's network access device 1021. The attributes of the user's requested service include, for example, requesting video services, application services, and hard disk storage services.

[0058] Alternatively, when the user device 101 accesses, the corresponding service request is transmitted to the computing power node 103 through the first operator network access device 1021, the central network device 1023, the second operator network access device 1022, and the resource pool gateway device 1031 in the form of information flow. The computing power node 103 also includes a computing power service resource pool that provides corresponding services to the user; for example, if the user requests a hard disk storage service, then the corresponding computing power service resource pool that has spare capacity to provide services to the user will provide the user with hard disk storage services;

[0059] The second operator network access device 1022 in each region corresponds to the resource pool gateway device 1031 in the region, and is connected to the appropriate computing power service resource pool through the resource pool gateway device 1031;

[0060] In the centralized architecture, the orchestration layer 104 has a global view, collects all information from network nodes 102, computing nodes 103, and business systems 105, and manages and allocates computing service resource pools to users;

[0061] In the distributed architecture, after collecting the service request of the user device 101 through the first operator network access device 1021 and collecting the relevant computing power information of the computing power node 103 synchronized by the orchestration layer 104, the computing power service resource pool can be allocated to the user management;

[0062] The service request includes latency requirements and computing power service resource pool requirements; after the orchestration layer 104 confirms that the user has a latency requirement, it confirms whether the user has a computing power service resource pool requirement, and selects a target resource pool for the user's service from multiple computing power service resource pools based on the computing power service resource pool requirement; the target resource pool can be selected by the user or can be selected based on the resource situation of the computing power service resource pool; after the target resource pool is confirmed, the latency in the path of providing services to the user device 101 is confirmed through the target resource pool, including the first service change latency of the centralized architecture and the second service change latency of the distributed architecture; based on the latency requirement, the first service change latency, and the second service change latency, a comparison is made as to whether it is suitable to provide the user with a centralized architecture or a distributed architecture, so as to facilitate subsequent services through the corresponding architecture.

[0063] Figure 2 A schematic diagram of a process of combining centralized and distributed network computing decision-making method provided in an embodiment of the present application Figure 1 .like Figure 2 As shown, the method includes:

[0064] S201. According to the user's business request, obtain the user's latency requirement and computing power service resource pool requirement, wherein the latency requirement is used to indicate the network latency requirement of the user's business, and the computing power service resource pool requirement is used to indicate the user's selection of the computing power service resource pool.

[0065] The user's service request is received through the service system or service access point, which is the interface between the user device and the first operator's network access device; the service request includes optional latency requirements and computing power service resource pool requirements. If the user does not fill in this information, it is confirmed that the user has no network latency requirements and has not selected a computing power service resource pool.

[0066] S202. If the latency requirement indicates that the network latency requirement of the user's service is less than a first threshold, a target resource pool is selected for the user's service from a plurality of computing power service resource pools according to the computing power service resource pool requirement.

[0067] If the user's latency requirement indicates that the network latency that the user needs the computing power service resource pool to provide services is less than a first threshold, then it is necessary to further confirm which computing power service resource pools provide services to the user, that is, which target resource pools are, so that corresponding services can be provided according to the latency of the devices along the service path provided by the network, where the first threshold is a value set by the user, and the devices along the path include the target resource pool.

[0068] The method of selecting a computing service resource pool can be determined according to the requirements of the computing service resource pool:

[0069] For example, if the computing power service resource pool requirement indicates that the user has selected a candidate resource pool from among the multiple computing power service resource pools, the candidate resource pool is used as the target resource pool.

[0070] If the computing power service resource pool requirement filled in by the user specifies that the user needs the computing power service resource pool in a certain area as the computing power service resource pool that the user expects to use (i.e., the candidate resource pool), the computing power service resource pool in the area selected by the user will be directly used by the user (i.e., the candidate resource pool will be used as the target resource pool), where one area has multiple computing power service resource pools.

[0071] If the computing power service resource pool demand indicates that the user has not selected a candidate resource pool from the multiple computing power service resource pools, then the computing power service resource pool whose current resource utilization is lower than a second threshold is selected from the multiple computing power service resource pools as the target resource pool, wherein the second threshold includes multiple threshold numbers, and the number of the threshold numbers is the same as the number of the computing power service resource pools.

[0072] If the user does not fill in the computing power service resource pool requirements, that is, the user has no requirements on which computing power service resource pool(s) to use, the orchestration layer collects the current resource utilization of each computing power service resource pool; if any computing power service resource pool has a current resource utilization rate less than its corresponding second threshold, the computing power service resource pool is confirmed as the target resource pool, wherein computing power service resource pools with the same resource type have the same threshold number of the second threshold corresponding to them;

[0073] For example, the resource types in the computing power service resource pool include infrastructure computing power resources, function computing power resources, etc.; infrastructure computing power resources include central processing units (CPU), graphics processing units (GPU), field programmable gate arrays (FPGA), etc.; function computing power resources include virtual machines, containers, microkernels, etc.; the threshold number of the second threshold corresponding to the infrastructure computing power resources can be set to 80% (that is, only the infrastructure computing power resources are allowed to be occupied by 80%), and the threshold number of the second threshold corresponding to the function computing power resources can be set to 60%. If a computing power resource (such as GPU) corresponding to multiple computing power service resource pools has been occupied by 78%, 60%, 84%, etc., respectively, the computing power service resource pool corresponding to the GPU occupancy of 78% and 60% will be used as the target resource pool.

[0074] S203: Acquire, according to the target resource pool, a first service change delay of the centralized architecture and a second service change delay of the distributed architecture.

[0075] The target resource pool corresponds to the resource pool gateway device in its area. It goes through the resource pool gateway device all the way back to the service access point to provide services to users. There will be a delay in this process, and the delays of the centralized architecture and the distributed architecture are different. The first service change delay and the second service change delay can be obtained respectively.

[0076] S204: Determine whether to provide a centralized architecture or a distributed architecture for the user's service based on the delay requirement, the first service change delay of the centralized architecture, and the second service change delay of the distributed architecture.

[0077] Compare the latency requirement with the first service change latency and the second service change latency to see which one better meets the latency requirement, and then provide the corresponding architecture to the user and use the architecture to serve the user.

[0078] In the embodiment of the present application, it is first confirmed whether the user has made a demand, including latency demand and computing power service resource pool demand. If there is a latency demand, it is further confirmed whether the user has specified a computing power service resource pool based on the user's computing power service resource pool demand, so as to confirm the target resource pool that provides services to the user from multiple computing power service resource pools. After the target resource pool is determined, the first service change delay of the centralized architecture and the second service change delay of the distributed architecture are confirmed based on the participation of the target resource pool in the service, and the user's latency demand is compared with the first service change delay and the second service change delay, and the appropriate architecture is confirmed in the centralized architecture and the distributed architecture; while considering the user's needs and the ability of the resource pool to provide services, the centralized architecture and the distributed architecture are reasonably allocated, thereby improving the efficiency of computing network decision-making, while ensuring the reasonable allocation of network resources and improving network performance.

[0079] Figure 3 A schematic diagram of a process of combining centralized and distributed network computing decision-making method provided in an embodiment of the present application Figure 2 .like Figure 3 As shown, the method includes:

[0080] S301. Obtain the service initialization delay based on the delay from the user equipment to the first operator network access device, the delay from the first operator network access device to the second operator network access device, the delay from the second operator network access device to the resource pool gateway device, and the delay from the second operator network access device to the target resource pool.

[0081] Whether it is the service initialization of a centralized architecture or the service initialization of a distributed architecture, the device resources in the path involve the user equipment, the first operator network access device, the second operator network access device, the resource pool gateway device and the target resource pool; therefore, the delay from the user equipment to the first operator network access device, the delay from the first operator network access device to the second operator network access device, the delay from the second operator network access device to the resource pool gateway device, and the delay from the resource pool gateway device to the target resource pool are obtained, and the delays obtained above are summed to obtain the service initialization delay;

[0082] Among them, the delay from the first operator's network access equipment to the second operator's network access equipment belongs to the delay of the inter-city link between the AZ ends of the service, and the delay from the second operator's network access equipment to the target resource pool belongs to the delay within the resource pool.

[0083] S302. Obtain a first service change delay of the centralized architecture according to the service initialization delay, and the time for computing service selection, path calculation, and tunnel delivery at the orchestration layer after the service change.

[0084] For a centralized architecture, after a service change, the orchestration layer also needs to obtain the time for the centralized architecture to select computing services, the time for path calculation, and the time for tunnel delivery, and add these times to the service initialization delay to obtain the first service change delay of the centralized architecture; wherein, computing service selection includes the process of selecting a computing service resource pool, path calculation includes calculating a suitable data interaction path, and tunnel delivery includes sending service feedback from the orchestration layer to the first operator's network access device.

[0085] S303. Obtain a second service change delay of the distributed architecture according to the service initialization delay, the time of computing service selection and the time of path calculation after the computing power information is changed.

[0086] For the distributed architecture, after the service change, the difference from the centralized architecture is that there is no need to send tunnels, so it is only necessary to obtain the time for computing service selection and path calculation time in the first operator's network access device, add these times and the service initialization delay, and get the second service change delay of the distributed architecture. It can be seen that the second service change delay is less than the first service change delay.

[0087] In an embodiment of the present application, the service initialization delay is obtained through the equipment passed through the path during the service initialization process, and the equipment passed through the path includes user equipment, first operator network access equipment, second operator network access equipment, resource pool gateway equipment and target resource pool; and the first service change delay of the centralized architecture and the second service change delay of the distributed architecture are obtained through the working characteristics of the centralized architecture and the distributed architecture, fully considering the delay of the computing power network of the service provided to the user, so as to facilitate subsequent comparison to see whether it can meet the user's delay requirements.

[0088] Figure 4 A schematic diagram of a process of combining centralized and distributed network computing decision-making method provided in an embodiment of the present application Figure 3 .like Figure 4 As shown, the method includes:

[0089] S401: If the first threshold is greater than or equal to the first service change delay, confirm that a centralized architecture is provided for the user's service.

[0090] If the user's delay requirement indicates that the user's expected delay is less than a first threshold, after the user fills in the first threshold, the first threshold is compared to see whether it is greater than or equal to the first service change delay. If so, it is confirmed that the centralized architecture meets the delay requirement and the centralized architecture has global control capabilities, providing a centralized architecture for the user's business.

[0091] S402: If the first threshold is greater than or equal to the second service change delay and less than the first service change delay, confirm that a distributed architecture is provided for the user's service, wherein the first service change delay is greater than the second service change delay.

[0092] Based on the fact that the first service change delay is greater than the second service change delay, if the first threshold is less than the first service change delay and greater than or equal to the second service change delay, it is confirmed that the centralized architecture cannot meet the delay requirement, and it is confirmed to provide the user with a distributed architecture.

[0093] If none of the above conditions are met, that is, the first threshold is less than the second service change delay, the user has too high a delay requirement, and the corresponding service cannot be provided to the user, and corresponding prompt information can be fed back to the user.

[0094] In the embodiment of the present application, by comparing the first threshold filled in the delay requirement with the first service change delay and the second service change delay, a suitable architecture is selected to achieve reasonable allocation.

[0095] The above embodiment takes into account the situation of latency requirements. If the latency requirements indicate that the user's service has no network latency requirements, and the computing power service resource pool requirements indicate that the user selects a candidate resource pool, a distributed architecture is provided to the user.

[0096] If the latency requirement indicates that the user's service has no network latency requirement, and the computing power service resource pool requirement indicates that the user has not selected a candidate resource pool, further decision is made through the following embodiments:

[0097] Figure 5 A schematic diagram of a process of combining centralized and distributed network computing decision-making method provided in an embodiment of the present application Figure 4 .like Figure 5 As shown, the method includes:

[0098] S501. Obtain, according to the service request, whether the user differentiates service categories, wherein the service categories are divided according to whether the service complies with a centralized architecture or a distributed architecture.

[0099] The business request also includes a business service category requirement, which is used to indicate whether the user has selected a business service category. The business service category is divided according to whether the business complies with a centralized architecture or a distributed architecture:

[0100] For example, if the user distinguishes the business service category, if the business service category selected by the user is a global control type, it is confirmed that a centralized architecture is provided for the user's business;

[0101] If the business service category selected by the user is delay-sensitive, it is confirmed that a distributed architecture is provided for the user's business.

[0102] Centralized path calculation for globally controlled services can achieve the best global resource utilization and find the optimal solution for global resource utilization. Globally controlled services correspond to centralized architectures. Delay-sensitive services use distributed architectures for local path calculation. When services are changed or after a failure, fast local tuning and convergence can be achieved. Delay-sensitive services correspond to distributed architectures.

[0103] Provide users with global control type and delay-sensitive type options. Users can choose one or neither. The selection result can be fed back to the orchestration layer through business requests, and the orchestration layer makes architectural decisions. If the user chooses the global control type, a centralized architecture is provided to the user. If the user chooses the delay-sensitive type, a distributed architecture is provided to the user. If the user does not distinguish between business service categories, that is, does not select a specific business service category, further decisions are made according to the following steps.

[0104] S502. If the user does not distinguish the business service category, then select the computing power service resource pool whose current resource utilization is lower than a third threshold from the multiple computing power service resource pools as the target resource pool, wherein the third threshold includes multiple threshold numbers, and the number of the threshold numbers is the same as the number of the computing power service resource pools.

[0105] When the user has no business service category requirements, the resource status of each computing power service resource pool is calculated. The calculation method is the same as the calculation method for setting the second threshold, which will not be repeated here.

[0106] S503: If the service access point and the target resource pool belong to the same area, confirm that a distributed architecture is provided for the user's service, and the same area includes a provincial area or a prefecture-level city area.

[0107] After selecting a suitable target resource pool from multiple computing power service resource pools, confirm the location relationship between the target resource pool and the business access point, including whether the business access point and the target resource pool belong to the same area. If so, confirm that a distributed architecture is provided for the user's business. The same area can be divided according to provinces, cities, etc. The computing power service resource pools at the same province / city level are related. If the user moves, the initial business access point is switched to other computing power service resource pools in the same area to provide services to the user. The switch is convenient and quick.

[0108] S504: If the service access point and the target resource pool do not belong to the same area, and the number of the target resource pools within a preset distance range from the service access point is greater than or equal to a preset number value, confirm that a distributed architecture is provided for the user's service.

[0109] If the target resource pool that can provide services to the user does not belong to the same area as the service access point (it may be that the current resource utilization rate of the computing service resource pool in the area has reached the threshold and cannot provide services), then it can be confirmed whether there are enough target resource pools within the preset distance range centered on the service access point to provide services to the user. If so, it is confirmed that a distributed architecture is provided for the user's business; for example, the preset distance is 500km, the preset number value is 3, and there are 5 target resource pools. If there are 3 or more target resource pools within 500km centered on the service access point, then it is confirmed that a distributed architecture is provided for the user's business; if there are only 2 or fewer target resource pools within 500km centered on the service access point, then other judgments can be further made.

[0110] S505: If the service access point and the target resource pool do not belong to the same area, the number of the target resource pools within a preset distance range from the service access point is less than a preset number value, and the service access point is located in a delay-sensitive area, confirm that a distributed architecture is provided for the user's service.

[0111] If the service access point and the target resource pool do not belong to the same area, and the number of target resource pools within a preset distance range from the service access point is less than a preset number value (see the previous step for a specific example), then it can also be confirmed whether the user's service access point is located in a delay-sensitive area. If so, it is confirmed that a distributed architecture is provided for the user's service; wherein, the delay-sensitive area includes areas where industries such as Internet of Vehicles, smart medical care, and digital twins are located. These industries are sensitive to network delay requirements, and providing a distributed architecture can better meet user needs.

[0112] Otherwise, that is, if the service access point and the target resource pool do not belong to the same area, and the number of target resource pools within the preset distance range from the service access point is less than the preset number value, and the service access point is not located in a delay-sensitive area, then it is confirmed that the centralized architecture is provided for the user's service.

[0113] In the embodiment of the present application, when the user has no latency requirement and does not specify a computing power service resource pool, an architecture is allocated to the user based on whether the user distinguishes between business service categories; if the user distinguishes between business service categories, an architecture is allocated based on the characteristics of the business service categories; if the user does not distinguish between business service categories, an allocation architecture is confirmed based on whether the business access point and the target resource pool belong to the same area, whether the number of target resource pools within a preset distance range from the business access point is less than a preset number value, and whether the business access point is located in a latency-sensitive area, and the most suitable architecture is provided to the user, thereby ensuring computing network decision-making efficiency and network performance when serving users.

[0114] Figure 6 A diagram of a network decision-making device combining centralized and distributed computing is provided in an embodiment of the present invention, such as Figure 6 As shown, the device includes: a first acquisition module 601, a determination module 602, a second acquisition module 603 and an allocation module 604;

[0115] The first acquisition module 601 is used to obtain the user's latency requirement and computing power service resource pool requirement according to the user's business request, wherein the latency requirement is used to indicate the network latency requirement of the user's business, and the computing power service resource pool requirement is used to indicate the user's selection of the computing power service resource pool.

[0116] The determination module 602 is used to select a target resource pool for the user's business from multiple computing power service resource pools according to the computing power service resource pool requirement if the latency requirement indicates that the network latency requirement of the user's business is less than a first threshold.

[0117] The determination module 602 is further configured to use the candidate resource pool as the target resource pool if the computing power service resource pool requirement indicates that the user has selected a candidate resource pool from among the plurality of computing power service resource pools;

[0118] If the computing power service resource pool demand indicates that the user has not selected a candidate resource pool from the multiple computing power service resource pools, then the computing power service resource pool whose current resource utilization is lower than a second threshold is selected from the multiple computing power service resource pools as the target resource pool, wherein the second threshold includes multiple threshold numbers, and the number of the threshold numbers is the same as the number of the computing power service resource pools.

[0119] The second acquisition module 603 is used to acquire the first service change delay of the centralized architecture and the second service change delay of the distributed architecture according to the target resource pool.

[0120] The second acquisition module 603 is further used to obtain the service initialization delay according to the delay from the user equipment to the first operator network access device, the delay from the first operator network access device to the second operator network access device, the delay from the second operator network access device to the resource pool gateway device, and the delay from the second operator network access device to the target resource pool;

[0121] Obtaining a first service change delay of the centralized architecture according to the service initialization delay, and the time for computing service selection, path calculation, and tunnel delivery at the orchestration layer after the service change;

[0122] According to the service initialization delay, as well as the time for selecting the computing service and the time for path calculation after the computing power information is changed, the second service change delay of the distributed architecture is obtained.

[0123] The allocation module 604 is used to determine whether to provide a centralized architecture or a distributed architecture for the user's service according to the delay requirement, the first service change delay of the centralized architecture, and the second service change delay of the distributed architecture.

[0124] The allocation module 604 is further configured to confirm that a centralized architecture is provided for the user's service if the first threshold is greater than or equal to the first service change delay;

[0125] If the first threshold is greater than or equal to the second service change delay and less than the first service change delay, it is confirmed that a distributed architecture is provided for the user's service, wherein the first service change delay is greater than the second service change delay.

[0126] The allocation module 604 is further configured to: if the latency requirement indicates that the user's service has no network latency requirement, and the computing power service resource pool requirement indicates that the user has not selected a candidate resource pool, then the method further includes:

[0127] Acquiring whether the user distinguishes business service categories according to the business request, wherein the business service categories are divided according to whether the business complies with a centralized architecture or a distributed architecture;

[0128] If the user does not distinguish the business service category, the computing power service resource pool whose current resource utilization rate is lower than a third threshold is selected from the multiple computing power service resource pools as the target resource pool, wherein the third threshold includes multiple threshold numbers, and the number of the threshold numbers is the same as the number of the computing power service resource pools;

[0129] The centralized architecture or the distributed architecture provided to the user's service is determined according to the positional relationship between the user's service access point and the target resource pool.

[0130] The present application also provides a computing network decision device combining centralized and distributed computing networks, including: at least one processor and a memory;

[0131] The memory stores computer-executable instructions;

[0132] The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor performs a computing network decision-making method that combines centralized and distributed computing.

[0133] Figure 7 The hardware diagram of the centralized and distributed computing network decision-making device provided by the embodiment of the present invention is shown in FIG. Figure 7 As shown, the centralized and distributed network decision-making device 70 provided in this embodiment includes: at least one processor 701 and a memory 702. The device 70 also includes a communication component 703. The processor 701, the memory 702 and the communication component 703 are connected via a bus 704.

[0134] In the specific implementation process, at least one processor 701 executes the computer execution instructions stored in the memory 702, so that at least one processor 701 executes the above-mentioned centralized and distributed computing network decision-making method.

[0135] The specific implementation process of the processor 701 can be found in the above method embodiment, and its implementation principle and technical effect are similar, so this embodiment will not be repeated here.

[0136] In the above Figure 7In the illustrated embodiment, it should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the invention may be directly implemented as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0137] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk storage.

[0138] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.

[0139] The present application also provides a computer-readable storage medium, in which computer-executable instructions are stored. When a processor executes the computer-executable instructions, the above-mentioned method for combining centralized and distributed computing network decision-making is implemented.

[0140] The computer-readable storage medium mentioned above can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The readable storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0141] An exemplary readable storage medium is coupled to a processor so that the processor can read information from the readable storage medium and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can be located in an application specific integrated circuit (Application Specific Integrated Circuits, referred to as: ASIC). Of course, the processor and the readable storage medium can also exist in the device as discrete components.

[0142] The division of the units is only a logical function division, and there may be other divisions in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, which may be electrical, mechanical or other forms.

[0143] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0144] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0145] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0146] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0147] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention, which follow the general principles of the present invention and include common knowledge or customary technical means in the art not disclosed by the present invention, are not limited to the precise structure described above and shown in the drawings, and may be modified and changed in various ways without departing from the scope thereof. The scope of the present invention is limited only by the appended claims.

Claims

1. A method for combining centralized and distributed network computing decision-making, characterized in that: include: According to the user's service request, the user's latency requirement and computing power service resource pool requirement are obtained, wherein the latency requirement is used to indicate the network latency requirement of the user's service, and the computing power service resource pool requirement is used to indicate the user's selection of the computing power service resource pool; If the latency requirement indicates that the network latency requirement of the user's service is less than a first threshold, selecting a target resource pool for the user's service from the plurality of computing power service resource pools according to the computing power service resource pool requirement; According to the target resource pool, obtaining a first service change delay of the centralized architecture and a second service change delay of the distributed architecture; According to the delay requirement, the first service change delay of the centralized architecture and the second service change delay of the distributed architecture, it is determined whether a centralized architecture or a distributed architecture is provided for the user's service.

2. The method according to claim 1, characterized in that: The step of selecting a target resource pool for a user's business from a plurality of computing power service resource pools according to the computing power service resource pool requirements includes: If the computing power service resource pool requirement indicates that the user has selected a candidate resource pool from among the multiple computing power service resource pools, the candidate resource pool is used as the target resource pool; If the computing power service resource pool demand indicates that the user has not selected a candidate resource pool from the multiple computing power service resource pools, then the computing power service resource pool whose current resource utilization is lower than a second threshold is selected from the multiple computing power service resource pools as the target resource pool, wherein the second threshold includes multiple threshold numbers, and the number of the threshold numbers is the same as the number of the computing power service resource pools.

3. The method according to claim 1, characterized in that The acquiring, according to the target resource pool, a first service change delay of the centralized architecture and a second service change delay of the distributed architecture includes: Obtain the service initialization delay according to the delay from the user equipment to the first operator network access device, the delay from the first operator network access device to the second operator network access device, the delay from the second operator network access device to the resource pool gateway device, and the delay from the second operator network access device to the target resource pool; Obtaining a first service change delay of the centralized architecture according to the service initialization delay, and the time for computing service selection, path calculation, and tunnel delivery at the orchestration layer after the service change; According to the service initialization delay, as well as the time for selecting the computing service and the time for path calculation after the computing power information is changed, the second service change delay of the distributed architecture is obtained.

4. The method according to claim 1, characterized in that: The step of confirming whether to provide a centralized architecture or a distributed architecture for a service of a user according to the delay requirement, the first service change delay of the centralized architecture, and the second service change delay of the distributed architecture includes: If the first threshold is greater than or equal to the first service change delay, confirming that a centralized architecture is provided for the user's service; If the first threshold is greater than or equal to the second service change delay and less than the first service change delay, it is confirmed that a distributed architecture is provided for the user's service, wherein the first service change delay is greater than the second service change delay.

5. The method according to claim 1, characterized in that If the latency requirement indicates that the user's service has no network latency requirement, and the computing power service resource pool requirement indicates that the user has not selected a candidate resource pool, the method further includes: Acquiring whether the user distinguishes business service categories according to the business request, wherein the business service categories are divided according to whether the business complies with a centralized architecture or a distributed architecture; If the user does not distinguish the business service category, the computing power service resource pool whose current resource utilization rate is lower than a third threshold is selected from the multiple computing power service resource pools as the target resource pool, wherein the third threshold includes multiple threshold numbers, and the number of the threshold numbers is the same as the number of the computing power service resource pools; The centralized architecture or the distributed architecture provided to the user's service is determined according to the positional relationship between the user's service access point and the target resource pool.

6. The method according to claim 5, characterized in that The method of determining whether to provide a centralized architecture or a distributed architecture for a service to a user based on the positional relationship between the user's service access point and the target resource pool includes: If the service access point and the target resource pool belong to the same region, confirming that the service provided to the user is provided with a distributed architecture, the same region includes a provincial region or a prefecture-level region; If the service access point and the target resource pool do not belong to the same area, and the number of the target resource pools within a preset distance range from the service access point is greater than or equal to a preset number value, confirming that a distributed architecture is provided for the user's service; If the service access point and the target resource pool do not belong to the same area, the number of the target resource pools within a preset distance range from the service access point is less than a preset number value, and the service access point is located in a delay-sensitive area, it is confirmed that a distributed architecture is provided for the user's service.

7. The method according to claim 5, characterized in that If the user distinguishes the business service category, the method further includes: If the business service category selected by the user is global control type, it is confirmed that a centralized architecture is provided for the user's business; If the business service category selected by the user is delay-sensitive, it is confirmed that a distributed architecture is provided for the user's business.

8. A network computing decision-making device combining centralized and distributed computing, characterized in that: include: A first acquisition module is used to acquire the user's latency requirement and computing power service resource pool requirement according to the user's service request, wherein the latency requirement is used to indicate the network latency requirement of the user's service, and the computing power service resource pool requirement is used to indicate the user's selection of the computing power service resource pool; A determination module, configured to select a target resource pool for the user's service from a plurality of computing power service resource pools according to the computing power service resource pool requirement if the latency requirement indicates that the network latency requirement of the user's service is less than a first threshold; A second acquisition module, configured to acquire, according to the target resource pool, a first service change delay of the centralized architecture and a second service change delay of the distributed architecture; The allocation module is used to confirm whether to provide a centralized architecture or a distributed architecture for the user's service according to the delay requirement, the first service change delay of the centralized architecture and the second service change delay of the distributed architecture.

9. A network computing decision-making device combining centralized and distributed computing, characterized in that: include: at least one processor and memory; The memory stores computer-executable instructions; The at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the centralized and distributed computing network decision-making method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for combining centralized and distributed computing network decision-making as described in any one of claims 1 to 7 are implemented.

Images