Forwarding method for computing power routing, and electronic device and computer-readable medium

Abstract

Provided in the embodiments of the present application are a forwarding method for computing power routing, and an electronic device and a computer-readable medium. The method comprises: a first edge device receiving a target message to be forwarded; determining, from among a plurality of forwarding information tables, a target forwarding information table which is associated with a first computing power SID carried by said target message; querying the target forwarding information table according to a service identifier carried by said target message, so as to obtain computing power routing information of the next hop; and according to the computing power routing information, replacing the first computing power SID in said target message with a second computing power SID, and forwarding said target message, wherein a plurality of computing power SIDs are stored in the first edge device, different computing power SIDs are associated with different forwarding information tables, different forwarding information tables are used for message forwarding at different network levels in a computing power network, and the first computing power SID belongs to the plurality of different computing power SIDs.

Description

Forwarding method for computing power routing, electronic device and computer-readable medium

[0001] Cross-references

[0002] This invention claims priority to a Chinese patent application filed with the Patent Office of China on June 30, 2023, with application number 202310802457.3 and titled “A forwarding method, electronic device and computer-readable medium for computing power routing”. The entire contents of this application are incorporated herein by reference. Technical Field

[0003] The present application relates to the field of communication technology, and in particular to a forwarding method for computing power routing, an electronic device, and a computer-readable medium. Background Art

[0004] In hierarchical computing power routing, since each edge device perceives the computing / network status changes at different times, resulting in different convergence orders and times, it is possible that the convergence status of each edge device on the forwarding path of the service message will be inconsistent, which will lead to forwarding loops and affect network performance.

[0005] Summary of the Invention

[0006] The purpose of the embodiments of the present application is to provide a forwarding method, electronic device and computer-readable medium for computing power routing.

[0007] To solve the above technical problems, the embodiments of the present application are implemented through the following aspects.

[0008] In a first aspect, an embodiment of the present application provides a forwarding method for computing power routing, which is applied to a first edge device, and the method includes: receiving a target message to be forwarded, the target message carrying a first computing power segment identifier SID and a service identifier of a requested specific service; determining a target forwarding information table associated with the first computing power SID in multiple forwarding information tables; querying the target forwarding information table according to the service identifier to obtain the computing power routing information of the next hop, the computing power routing information including a second computing power SID and a first output interface information; replacing the first computing power SID in the destination address field in the target message with the second computing power SID, and forwarding the replaced target message according to the first output interface information; wherein, the first edge device stores multiple computing power SIDs, different computing power SIDs are associated with different forwarding information tables, and different forwarding information tables are used for message forwarding at different network layers in the computing power network, and the first computing power SID belongs to the multiple different computing power SIDs.

[0009] In a second aspect, an embodiment of the present application provides an electronic device comprising: a memory, a processor, and computer-executable instructions stored on the memory and executable on the processor, wherein the computer-executable instructions, when executed by the processor, implement the steps of the method described in the first aspect.

[0010] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, which is used to store computer-executable instructions. When the computer-executable instructions are executed by a processor, the steps of the method described in the first aspect are implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments recorded in this application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0012] FIG1 shows a flow chart of a forwarding method for computing power routing provided in an embodiment of the present application.

[0013] FIG2 shows one of the schematic diagrams of the process of determining computing power status information provided in an embodiment of the present application.

[0014] FIG3 shows a second schematic diagram of the process of determining computing power status information provided in an embodiment of the present application.

[0015] FIG4 shows a schematic diagram of a hierarchical table entry generation process provided in an embodiment of the present application.

[0016] FIG5 shows one of the schematic diagrams of the aggregation and release process of computing power status information provided in an embodiment of the present application.

[0017] FIG6 shows a second schematic diagram of the aggregation and release process of computing power status information provided in an embodiment of the present application.

[0018] FIG7 shows one of the schematic diagrams of the hierarchical computing power routing process provided in an embodiment of the present application.

[0019] FIG8 a shows a second schematic diagram of the hierarchical computing power routing process provided in an embodiment of the present application.

[0020] FIG8b shows a schematic diagram of the computing power routing process in the related technology provided in an embodiment of the present application.

[0021] FIG9 shows a schematic structural diagram of a forwarding device for computing power routing provided in an embodiment of the present application.

[0022] FIG10 is a schematic diagram of the hardware structure of an electronic device that executes the forwarding method of computing power routing provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0024] 5G has driven edge computing to become a hot topic, accelerating the deployment of Mobile Edge Computing (MEC). For emerging services like the future Industrial Internet, network-side edge computing, and even on-site edge computing, will provide users with flexible, local access and deliver an exceptional user experience. Consequently, the trend is shifting from centralized computing power to decentralized computing, with distributed resources / computing power and distributed services becoming increasingly common. Furthermore, with the development of container technology and microservices architectures, applications are being deconstructed into functions and services, and networks are shifting from host-oriented addressing to service-oriented / computing power-oriented addressing. This has given rise to a new technological form and operational model, the computing power network.

[0025] Among them, for computing power networks, since the status of computing power resources is often highly dynamic and the number of computing power service instances is large, aggregating the status information of highly dynamic and fine-grained computing power service instances into the status information of relatively stable coarse-grained resource pools, and then announcing it in the computing power network as a hierarchical computing power perception and routing method is considered to be able to better adapt to the new scenario of computing power networks.

[0026] However, for hierarchical computing power routing, since each edge device perceives the computing / network status changes at different times, which leads to different convergence orders and times, it may result in inconsistent convergence states of each edge device on the forwarding path of business messages, which in turn leads to forwarding loops and affects network performance.

[0027] To this end, this application provides a forwarding solution for computing power routing to ensure the accuracy of path decisions, avoid forwarding loops, and improve network performance. The following describes the forwarding solution for computing power routing provided by this application in conjunction with the accompanying drawings and various implementation methods.

[0028] FIG1 illustrates a flow chart of a forwarding method 100 for computing power routing provided in an embodiment of the present application. This method 100 may be executed by a first edge device, such as software and / or hardware installed in the first edge device. In this embodiment, the first edge device may be a terminal device or a server device.

[0029] It is worth noting that in this embodiment, multiple computing power segment identifiers (Segment Identifier, SID) can be configured and stored in the first edge device. Different computing power SIDs are associated with different forwarding information tables (also called forwarding information base (FIB)), and different forwarding information tables are used for message forwarding at different network levels in the computing power network.

[0030] Based on this, in the process of forwarding computing power routing, the present application realizes the forwarding of the target message by matching the computing power SID carried in the message with the hierarchical table entry. For example, the first edge device judges the hit relationship between the computing power SID carried in the message and the locally defined computing power SID based on the hierarchical computing power SID stored in itself and the hierarchical multiple forwarding information tables corresponding to the hierarchical computing power SID, and makes a decision on the next-hop forwarding path for the target message to be forwarded (also called the business message) based on the hit relationship. In this way, the level of the forwarding information table used in the decision-making process can be consistent with the level indicated by the computing power SID in the target message, that is, the level of the forwarding information table used in the decision-making process is consistent with the network level of the edge device in the message forwarding process, which effectively ensures the accuracy of the path decision, avoids forwarding loops, and improves network performance.

[0031] In addition, the specific services or computing power services mentioned in the context of this application can be determined according to demand, and this application does not impose any restrictions on this.

[0032] Based on this, as shown in FIG1 , the forwarding method 100 for computing power routing provided in this embodiment may include the following steps.

[0033] S110: The first edge device receives a target message to be forwarded.

[0034] Among them, the target message carries the first computing power SID and the service identifier (Service ID) of the requested specific service. In this embodiment, the first computing power SID can be carried in the destination address (Destination Address) field in the target message for querying hierarchical table entries and hierarchical forwarding in the computing power routing process. In this embodiment, the first computing power SID is one of the multiple computing power SIDs stored in the first edge device, and is used by the previous hop edge device to forward the target message to the first edge device.

[0035] In one implementation, the process of the first edge device receiving the target message to be forwarded may include: after determining that the target message to be forwarded has been received, parsing a specified field in the message header of the target message to obtain the service identifier of the specific service. In other words, in this application, the computing power service (such as a specific service) is characterized by the service identifier.

[0036] In some embodiments, the designated field may be, but is not limited to, a Destination Options Header (DOH) field, a Hop-by-Hop Options Header (HBH) field, or a Segment Routing Header (SRH) field.

[0037] S120: Determine a target forwarding information table associated with the first computing power SID from multiple forwarding information tables.

[0038] S130, query the target forwarding information table according to the service identifier to obtain the computing power routing information of the next hop (Next Hop).

[0039] Among them, the first edge device queries the target forwarding information table with the service identifier as the index to obtain the computing power routing information of the next hop, such as the second computing power SID and the first outbound interface information, for forwarding the target message.

[0040] It should be noted that the computing power routing information may vary depending on the network layer corresponding to the target forwarding information table. For example, when the network layer corresponding to the target forwarding information table includes multiple different edge devices, the computing power routing information may include the SID and outbound interface information corresponding to a certain edge device in that network layer.

[0041] For another example, the network layer corresponding to the target forwarding information table includes multiple computing service instances (Instance) corresponding to the current edge device. Then, the computing routing information may include the SID and output interface information corresponding to a computing service instance under the network layer.

[0042] S140: Replace the first computing power SID in the destination address field of the target message with the second computing power SID, and forward the replaced target message according to the first outbound interface information.

[0043] In this embodiment, for the received target message to be forwarded, the edge device in the computing power network can make a decision on the next-hop forwarding path and forward the target message (also called a service message) based on the hierarchical SID stored in itself and the hierarchical forwarding information table associated with the SID. This effectively ensures the accuracy of the path decision, avoids forwarding loops, and improves network performance.

[0044] Based on this, for the multiple forwarding information tables stored in the aforementioned first edge device, there can be multiple implementation methods depending on the different network layers actually divided. For example, in one implementation method, the multiple forwarding information tables may include a global forwarding information table and a local forwarding information table. The global forwarding information table is used to indicate the message forwarding routing information of the target message between the edge devices in the computing power network. The local forwarding information table is used to indicate the message forwarding routing information of the target message between the computing power service instances corresponding to the first edge device.

[0045] Corresponding to the hierarchical forwarding information table, the present application may also determine the forwarding information table based on a hierarchical routing information table (also referred to as a routing information base (RIB)). For example, the global forwarding information table may be determined by the first edge device based on a global routing information table, wherein the global routing information table stores device information of each edge device in the computing power network and the capability of each edge device to provide the specific service.

[0046] For example, the local forwarding information table can also be determined by the first edge device based on the local routing information table, and the local routing information table stores the service instance information of each local computing service instance and the ability of each computing service instance to provide the specific service.

[0047] Regarding the generation of the aforementioned table entries, in this application, each edge device in the computing network can generate a routing information table through calculation on the control plane, and then send the generated routing information table to the forwarding plane to generate a forwarding information table. In this case, the following describes the generation process of each table entry using the first edge device as an example.

[0048] First, considering that different types of computing power services may be sensitive to different computing / network resource status metadata, such as the number of central processing unit (CPU) cores, CPU load, input / output interface (I / O), memory size, latency, etc., for a type of computing power service represented by a Service ID, the metadata of the related computing power service instances and the corresponding network forwarding paths that this type of computing power service is sensitive to can reflect their ability to provide this type of computing power service. And based on the evaluation method adapted to this type of computing power service, the metadata of the computing power service instances and the corresponding network forwarding paths can be converted into a metric value to reflect their ability to provide this type of computing power service through this metric value.

[0049] Based on this, the determination of the local routing information table and the global routing information table will be described below respectively.

[0050] In one implementation, the process of determining the global routing information table may include: the first edge device receiving first computing power status information sent by at least one second edge device, and then determining the global routing information table based on the second computing power status information and at least one of the first computing power status information. The second computing power status information is obtained by coarse-grained aggregation of local computing power status information perceived by the first edge device.

[0051] It can be understood that the first computing power status information indicates the ability of the second edge device to provide a specific service. It can be a set of metadata on the second edge device that is sensitive to the specific service, or a metric value obtained by converting the set of metadata on the second edge device that is sensitive to the specific service. Similarly, the second computing power status information indicates the ability of the first edge device to provide a specific service. It can be a set of metadata on the second edge device that is sensitive to the specific service, or a metric value obtained by converting the set of metadata on the second edge device that is sensitive to the specific service.

[0052] That is, each edge device in the computing power network can adopt the two different implementation methods mentioned above when determining the computing power status information. In this case, considering that the second edge device and the first device determine the computing power status information in a similar manner, this application will use the first edge device as an example to illustrate the process of determining the computing power status information.

[0053] For example, the first edge device aggregates the perceived local computing power status information to obtain the first computing power status information as shown in the following implementation method 1 and / or implementation method 2.

[0054] Implementation method 1: Aggregate multiple metric values ​​representing specific services provided by computing power service instances into one metric value, namely computing power status information. The implementation process can be shown in Figure 2 and the content is as follows.

[0055] S210, the first edge determines a computing network resource status metadata set corresponding to each computing power service instance.

[0056] Among them, considering that in the computing power network, there are multiple computing power service instances that can provide computing power services in the edge cloud or data center connected to each edge device, such as Instance 1, Instance 2,..., then each edge device can determine the computing network resource status metadata set corresponding to each computing power service instance based on the metadata set of different computing / network resource states of these computing power service instances connected to it at a certain moment, which can be recorded as Attr Set i, j.

[0057] S220, the first edge device determines the first metric value corresponding to each computing power service instance, such as Metric i, based on the computing network resource status metadata set corresponding to each computing power service instance. The first metric value is used to characterize the ability of the computing power service instance to provide a specific service.

[0058] S230, the first edge device aggregates the first metric value corresponding to each of the computing power service instances to obtain a second metric value corresponding to the second computing power status information, where the second metric value is used to characterize the ability of the resource pool at the first edge device to provide the specific service.

[0059] In this implementation, there may be multiple ways for the first edge device to aggregate the first metric value corresponding to each computing service instance to obtain the second metric value, such as shown in (11)-(15), that is, the first edge device may aggregate the first metric value based on any one of the following (11)-(15).

[0060] (11) The weighted average of the first metric values ​​corresponding to each computing power service instance is used as the second metric value.

[0061] For example, assuming that the first edge device corresponds to four computing service instances, such as: Instance1, Instance2, Instance3, Instance4, and the first metric value corresponding to each is 15, 30, 10, and 20 respectively, then the second metric value can be (15*w1+30*w2+20*w3+15*w4) / (w1+w2+w3+w4), where w1, w2, w3, and w4 are weights, which can be set according to needs and are not limited here.

[0062] (12) The maximum value of the first metric values ​​corresponding to each computing power service instance is used as the second metric value.

[0063] For example, assuming that the first edge device corresponds to four computing service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding first metric value is 15, 30, 10, and 20 respectively, then the maximum value can be determined to be 30, and then the second metric value is determined to be 30.

[0064] (13) The minimum value of the first metric values ​​corresponding to each computing power service instance is used as the second metric value.

[0065] For example, assuming that the first edge device corresponds to four computing service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding first metric value is 15, 30, 10, and 20 respectively, then the minimum value can be determined to be 10, and then the second metric value is determined to be 10.

[0066] (14) The median of the first metric values ​​corresponding to each computing power service instance is used as the second metric value.

[0067] For example, assuming that the first edge device corresponds to four computing service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding first metric value is 15, 30, 10, and 20 respectively, then the median can be determined to be 15 and 20, and then the second metric value can be determined to be (20+15) / 2=17.5.

[0068] (15) The average value of the first metric value corresponding to each computing power service instance is used as the second metric value.

[0069] For example, assuming that the first edge device corresponds to four computing service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding first metric value is 15, 30, 10, and 20 respectively, then the second metric value can be determined to be (30+10+20+15) / 4=18.75.

[0070] Implementation method 2: Aggregate the metadata sets of multiple computing power service instances into an aggregated metadata set, namely, computing power status information. The implementation process can be shown in Figure 3, and the content is as follows.

[0071] S310, for each computing power service instance, respectively determine the first metadata set of computing network resource status metadata that the computing power service instance can provide for each computing power service, and determine the union of the first metadata sets corresponding to the computing power service instance to obtain a second metadata set.

[0072] For example, assuming that the metadata set sensitive to a computing power service is recorded as Attr Set, then the metadata set of computing network resource status sensitive to computing power service instance j at a certain moment can be recorded as Attr Set i, j (i.e., the first metadata set). Then, for multiple computing power services, the first metadata set corresponding to computing power service instance j can be Attr Set 1, j, Attr Set 2, j, Attr Set 2, j... . Then, computing power service instance j can take the union of the first metadata sets Attr Set 1, j, Attr Set 2, j, Attr Set 2, j... corresponding to multiple computing power services to obtain a second metadata set.

[0073] For example, assuming the first metadata set Attr Set 1, j is {30ms, 50G} and Attr Set 2, j is {30ms, 10c}, then the second metadata set determined by the union of Attr Set 1, j and Attr Set 2, j can be {30ms, 50G, 10c}, where "ms" represents latency, "G" represents memory, and "c" represents CPU cores.

[0074] S320: Aggregate the second metadata set corresponding to each of the computing power service instances to obtain a third metadata set corresponding to the second computing power status information.

[0075] In this embodiment, there may be multiple ways for the first edge device to aggregate the second metadata set corresponding to each computing power service instance, such as shown in (21)-(26), that is, the first edge device may aggregate the second metadata set based on any one of the following (21)-(26).

[0076] (21) The average value of the same type of elements in the second metadata set corresponding to each computing power service instance is used as the corresponding element in the third metadata set.

[0077] For example, assuming that the first edge device corresponds to four computing service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding second metadata set is {30ms, 50G, 10c}, {20ms, 60G, 12c}, {40ms, 100G, 12c}, {10ms, 40G, 6c}, then it can be determined that the third metadata set is {25ms, 62.5G, 10c}.

[0078] (22) The weighted average of the same type of elements in the second metadata set corresponding to each computing power service instance is used as the corresponding element in the third metadata set.

[0079] For example, assuming that the first edge device corresponds to four computing service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding second metadata set is {30ms, 50G, 10c}, {20ms, 60G, 12c}, {40ms, 100G, 12c}, {10ms, 40G, 6c}, then the third metadata set can be determined to be {a ms, bG, cc}, where a = (30*a1+20*a2+40*a3+10*a4) / (a1+a2+a3+a4), b = (50*b1+60*b2+100*b3+40*b4), c = (10*c1+12*c2+12c3+6*c4), and the size of the aforementioned weights can be set according to needs and is not limited here.

[0080] (23) The maximum value of the same type of elements in the second metadata set corresponding to each computing power service instance is used as the corresponding element in the third metadata set.

[0081] For example, assuming that the first edge device corresponds to four computing service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding second metadata set is {30ms, 50G, 10c}, {20ms, 60G, 12c}, {40ms, 100G, 12c}, {10ms, 40G, 6c}, then it can be determined that the third metadata set is {40ms, 100G, 12c}.

[0082] (24) The minimum value of the same type of elements in the second metadata set corresponding to each computing power service instance is used as the corresponding element in the third metadata set.

[0083] For example, assuming that the first edge device corresponds to four computing service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding second metadata set is {30ms, 50G, 10c}, {20ms, 60G, 12c}, {40ms, 100G, 12c}, {10ms, 40G, 6c}, then it can be determined that the third metadata set is {10ms, 40G, 6c}.

[0084] (25) The median of the same type of elements in the second metadata set corresponding to each computing power service instance is used as the corresponding element in the third metadata set.

[0085] For example, assuming that the first edge device corresponds to four computing service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding second metadata set is {30ms, 50G, 10c}, {20ms, 60G, 12c}, {40ms, 100G, 12c}, {10ms, 40G, 6c}, then it can be determined that the third metadata set is {25ms, 55G, 15c}.

[0086] (26) The second metadata set corresponding to a specific computing power service instance in each of the computing power service instances is used as the third metadata set.

[0087] For example, assuming that the first edge device corresponds to four computing power service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding second metadata set is {30ms, 50G, 10c}, {20ms, 60G, 12c}, {40ms, 100G, 12c}, {10ms, 40G, 6c}, and Instance2 is a specific computing power service instance, then, it can be determined that the third metadata set is {20ms, 60G, 12c}.

[0088] It is understandable that the specific computing service instance mentioned in (26) can be determined based on demand, such as computing resource stability, etc., and no restrictions are imposed here.

[0089] It is worth noting that after the first edge device determines the second computing power status information based on the aforementioned implementation method 1 and / or implementation method 2, it can publish or notify the second computing power status information, such as the second metric value or the third metadata set, to each of the second edge devices in the computing power network. In some embodiments, when the first edge device publishes or notifies the second computing power status information, the extensible Border Gateway Protocol (BGP) and Interior Gateway Protocol (IGP) protocols are used for notification.

[0090] Similarly, after determining the first computing power status information, each second edge device in the computing power network can also publish and notify the second computing power status information to the computing power network.

[0091] Then, in this case, if the first edge device receives the first computing power status information sent by each second edge device, a global routing information table can be determined based on each of the first computing power status information and the locally determined second computing power status information.

[0092] For example, assuming that the computing power status information is Metric (i.e., the second Metric value), and for the computing power service represented by Service ID 1, the second Metric value determined by the first edge device PE1 is 17.5, and the second Metric values ​​published by the second edge devices PE2, PE3, and PE4 are 15, 25, and 30 respectively, then the global routing information table determined by the first edge device can be as shown in Table 1.

[0093] For another example, assuming that the computing power status information is a meta-information set (i.e., a third meta-information set), and for the computing power service represented by Service ID 1, the third meta-information set determined by the first edge device PE1 is {25ms, 55G, 15c}, and the third meta-information sets published by the second edge devices PE2, PE3, and PE4 are {30ms, 50G, 15c}, {40ms, 10G, 10c}, and {10ms, 80G, 8c} respectively. Then, the global routing information table determined by the first edge device can be as shown in Table 2.

[0094] Table 1

[0095] Table 2

[0096] It is worth noting that, based on the above content, the second computing power status information method provided in the above implementation method 1 and implementation method 2 is determined based on the granularity of the resource pool, and the resource pool includes resources corresponding to at least one computing power service instance.

[0097] Table 3

[0098] In addition, in the process of determining the aforementioned global routing table, each edge device can generate a local routing information table based on the computing power status information corresponding to each computing power service instance. For example, assuming that the computing power status information is Metric (i.e., the first Metric value), and for the computing power service represented by Service ID 1, the first edge device corresponds to four computing power service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding first Metric value is 15, 30, 10, and 20, then the local routing information table can be as shown in Table 3.

[0099] For another example, assuming that the computing power status information is a meta-information set (i.e., the second meta-information set), and for the computing power service represented by Service ID 1, the first edge device PE1 corresponds to four computing power service instances, such as: Instance1, Instance2, Instance3, Instance4, and each corresponding second meta-information set is {30ms, 50G, 10c}, {20ms, 60G, 12c}, {40ms, 100G, 12c}, {10ms, 40G, 6c}, then the local routing information table can be as shown in Table 4.

[0100] Table 4

[0101] In this embodiment, the aforementioned aggregation method of computing power status information is adopted to maintain the granularity of computing power service instances, so that the computing power status information has semantics comparable to the local detailed service instance information, thereby ensuring the reliability of the routing information table determined based on the computing power status information, and the basic correctness of the computing power routing control plane entries when making decisions, avoiding the problem of the aggregate value losing the semantics of the granularity of a single service instance due to the application of the accumulation method for aggregation in related technologies, resulting in the occurrence of computing power routing decision errors.

[0102] In this case, the generation of the global forwarding information table is described below using the method for determining the computing power status information provided in Implementation 1. Assuming that, based on the aforementioned collection, aggregation, and notification of computing power status information, each edge device in the computing power network has collected information about local computing power service instances, as well as computing power status information at the aggregated resource pool granularity, it can generate table entries at the global and local levels on the forwarding plane, such as the global forwarding information table and the local forwarding information table.

[0103] Here, we take the first edge device and the aggregation of computing power status information based on the metric value as an example. As shown in Figure 4, assuming that the first edge device (such as PE4) collects information about each local computing power service instance (such as Instance4,1, Instance4,2, Instance4,3, Instance4,4) and receives the first computing power status information sent by other edge devices (such as PE1, PE2, PE3), then the first edge device PE4 can determine the local routing information table as shown in Table 5 and the global routing information table as shown in Table 6.

[0104] Table 5

[0105] Table 6

[0106] There are many ways to determine the local forwarding information table based on the local routing information table and the global forwarding information table based on the global routing information table. In this embodiment, the entry with the smallest metric value is selected as the best forwarding entry in a preferential manner to generate the local forwarding information table shown in Table 7 and the global forwarding information table shown in Table 8.

[0107] Table 7

[0108] Table 8

[0109] It is understood that the aforementioned generation process of the global forwarding information table and the local forwarding information table is described using the metric value as an example (the generation process can also be shown in FIG4 ). In addition, considering that the process of generating the global forwarding information table and the local forwarding information table based on the meta-information set is similar to the process of generating the global forwarding information table and the local forwarding information table based on the metric value, to avoid repetition, this application will not elaborate on this process.

[0110] In one implementation, in order to effectively support the aforementioned hierarchical computing power routing forwarding process, the present application is configured with corresponding message features (i.e., the aforementioned computing power SID) to match the aforementioned hierarchical table entries (such as the aforementioned global forwarding information table and local forwarding information table), thereby realizing the query of next-hop routing information.

[0111] However, it should be noted that, corresponding to the aforementioned forwarding information table including the global forwarding information table and the local forwarding information table, the multiple computing power SIDs may also include a global computing power SID (Computing Global, END.CG) and a local computing power SID (Computing Local), and the global computing power SID is associated with the global forwarding information table, and the local computing power SID is associated with the local forwarding information table.

[0112] Then, in this case, the semantics of the local computing power SID (which can be recorded as END.CL) can be: for the received business message to be forwarded, if the computing power SID that can be carried in the business message is the local computing power SID, then the service identifier encapsulated in the DOH, HBH or SRH position of the business message can be parsed first, and then the local forwarding information table associated with the local computing power SID can be queried using the service identifier as an index, and finally the business message is forwarded to the next hop according to the queried computing power route.

[0113] Similarly, the semantics of the global computing power SID (denoted as END.CG) can be: if the computing power SID that can be carried in the business message is the global computing power SID, then the service identifier encapsulated in the DOH, HBH or SRH position of the business message can be parsed first, and then the global forwarding information table associated with the global computing power SID can be queried using the service identifier as an index, and finally the business message is forwarded to the next hop according to the queried computing power route.

[0114] It is worth noting that in the computing power network, the global computing power SID and the local computing power SID can be stored on each edge device respectively, and associated with the global forwarding information table and the local forwarding information table respectively. In this way, by separating the computing power resource information perceived at different levels, and then forming table entries at different levels, and explicitly specifying different table entries to be queried at different edge devices through the computing power SID, the micro-loop problem in hierarchical computing power routing can be effectively solved.

[0115] Of course, in order to realize the forwarding of business messages, after determining the computing power SID, each edge device can also publish or announce the computing power SID in the computing power network through the extended IGP or BGP, so that in the case of hierarchical computing power routing, the business message can carry the computing power SID, and the edge device that receives the business message can complete the hierarchical table entry query and hierarchical forwarding in the computing power routing process according to the corresponding SID semantics.

[0116] In one embodiment, based on the design of the aforementioned hierarchical table entries and SIDs, when the first edge device realizes routing forwarding based on the aforementioned S110-S140, if it senses that the forwarding path corresponding to the computing power routing information of the next hop is abnormal (such as interruption, etc.), then the target forwarding information table can be refreshed, and then the refreshed target forwarding information table is queried according to the service identifier to obtain the refreshed computing power routing information of the next hop, the target forwarding information table is associated with the first computing power SID, and the refreshed computing power routing information includes the third computing power SID and the second output interface information; the first computing power SID in the target message is replaced with the third computing power SID, and the replaced target message is forwarded according to the second output interface information, thereby achieving the correctness of routing forwarding. It can be understood that after the first edge device refreshes the target forwarding information table, it can announce the refreshed computing power status information to ensure the correctness of subsequent message forwarding and avoid forwarding loops.

[0117] In summary, the forwarding scheme of the aforementioned computing power routing proposed in the embodiments of this application mainly involves the following points (31)-(33).

[0118] (31) An aggregation method that integrates fine-grained detailed computing power status information into coarse-grained computing power status information is proposed, which is suitable for hierarchical routing decisions. It includes both the aggregation of computing power status metadata sets and the aggregation of metric values ​​that represent computing power service instances. It can solve the loop problem caused by inappropriate aggregation methods in hierarchical computing power perception and routing solutions.

[0119] (32) Different from the unified RIB table and FIB table of traditional IP networks, this application proposes a hierarchical table design based on the perception level, which may include a global RIB table, a global FIB table, and a local RIB table, a local FIB table.

[0120] (33) A hierarchical computing power routing message forwarding method is proposed by associating hierarchical table entries with computing power SID.

[0121] Among them, for the aforementioned (32)-(33), this application distinguishes the information perceived at different levels through a hierarchical table design, and then associates different table entries with different computing power SIDs. Finally, by displaying the specified SID, it can solve the micro-loop problem caused by convergence order and time in sudden and interruption situations.

[0122] That is, the forwarding solution for computing power routing provided by this application can supplement and improve the hierarchical computing power perception, notification, and routing solutions. Hierarchical computing power perception and notification reduce the number of control plane table entries in the network, significantly reducing the frequency of control plane calculations and refreshing FIB tables to the forwarding plane, and adapting to the application scenarios and heterogeneous development trends of computing power networks. At the same time, through the design of computing power status information aggregation methods, the semantic design of computing power SIDs, and the hierarchical table design, the loop and micro-loop problems under multi-point decision-making in hierarchical routing are solved.

[0123] Based on this, it can be understood that the forwarding solution of the computing power routing provided in this application can be applied to but not limited to IP-based transport networks (IP Radio Access Network, IPRAN) / encrypted virtual networks (Secret Private Network, SPN), metropolitan area networks / backbones, etc.

[0124] Based on the forwarding method of computing power routing provided in the aforementioned embodiment, for ease of understanding, the implementation process of this application is exemplarily described below with reference to different examples.

[0125] Example 1

[0126] Assume that edge devices PE1, PE2, and PE3 aggregate the metric values ​​corresponding to multiple computing service instances by taking the median of the metric values ​​corresponding to all computing service instances providing that service, and then publish and announce the unique metric value corresponding to a service ID to the network. The process of aggregating and publishing computing status information can be shown in Figure 5.

[0127] It can be understood that taking the aggregation process of computing power status information on PE1 shown in Figure 5 as an example, for the computing power service instances Instance1, Instance2, Instance3, and Instance4 corresponding to PE1, the ability of Instance1 to provide the computing power service represented by Service ID 1 (i.e., the first metric value) is 15, and the ability to provide the computing power service represented by Service ID 2 (i.e., the first metric value) is 20.

[0128] The ability of Instance2 to provide the computing service represented by Service ID 1 (i.e., the first metric value) is 30, and the ability to provide the computing service represented by Service ID 2 (i.e., the first metric value) is 15.

[0129] The capability of Instance3 to provide the computing service represented by Service ID 1 (i.e., the first metric value) is 10, and the capability of Instance3 to provide the computing service represented by Service ID 2 (i.e., the first metric value) is 25.

[0130] The capability of Instance4 to provide the computing service represented by Service ID 1 (i.e., the first metric value) is 20, and the capability of Instance4 to provide the computing service represented by Service ID 2 (i.e., the first metric value) is 30.

[0131] Based on this, the ability of PE1 to provide the computing service represented by Service ID 1 (i.e., the second metric value) is 17.5, and the ability to provide the computing service represented by Service ID 2 (i.e., the second metric value) is 22.5, as shown in Figure 5<Service ID1 PE1 Metric 17.5> 、<Service ID2 PE1 Metric 22.5> .

[0132] The process for PE2, PE3, and PE4 to determine their ability to provide the computing power service represented by Service ID 1 is similar to that of PE1 and will not be repeated here.

[0133] Then, PE4 acts as the source PE, and PE1, PE2, and PE3 can publish and announce their determined computing power status information (i.e., the ability to provide the computing power service represented by Service ID 1 and the ability to provide the computing power service represented by Service ID 2) in the computing power network. Then, PE4 can generate hierarchical table entries based on the received computing power status information corresponding to PE1, PE2, and PE3 and its own computing power status information, such as the local routing table, the global routing information table, the local forwarding information table, and the global forwarding information table. The generation process of the hierarchical table entries will not be repeated here.

[0134] It is worth noting that when aggregating computing power status information, the differentiated evaluation method for each type of computing power service can be degenerated into a unified or service-class-based evaluation method based on actual conditions. At this time, the metric values ​​corresponding to different Service IDs are the same, or a group of Service IDs corresponds to one metric, etc., and there is no restriction here.

[0135] Example 2

[0136] Assume that edge devices PE1, PE2, and PE3 aggregate the second metadata sets corresponding to multiple computing service instances as the average value of similar elements, and then publish and announce the unique metadata set corresponding to a Service ID to the computing network. The process of aggregating and publishing computing status information for each edge device can be shown in Figure 6.

[0137] It can be understood that, taking the aggregation process of computing power status information on PE1 shown in Figure 6 as an example, for the computing power service instances Instance1, Instance2, Instance3, and Instance4 corresponding to PE1, the capability of Instance1 to provide the computing power service represented by Service ID 1 (i.e., the first metadata set) is {30ms, 50G}, and the capability of Instance1 to provide the computing power service represented by Service ID 2 is {30ms, 10c}. Then, the union of the two first metadata sets is taken to obtain the second metadata set {30ms, 50G, 10c}.

[0138] Instance2's ability to provide the computing service represented by Service ID 1 (i.e., the first metadata set) is {20ms, 60G}, and its ability to provide the computing service represented by Service ID 2 is {20ms, 12c}. Then, the union of the two first metadata sets is taken to obtain the second metadata set {20ms, 60G, 12c}.

[0139] Instance 3's ability to provide the computing service represented by Service ID 1 (i.e., the first metadata set) is {40ms, 100G}, and its ability to provide the computing service represented by Service ID 2 is {40ms, 12c}. Then, the union of the two first metadata sets is taken to obtain the second metadata set {40ms, 100G, 12c}.

[0140] Instance 4's capability to provide the computing service represented by Service ID 1 (i.e., the first metadata set) is {10ms, 40G}, and its capability to provide the computing service represented by Service ID 2 is {10ms, 6c}. Therefore, the union of the two first metadata sets yields the second metadata set {10ms, 40G, 60c}.

[0141] Based on this, if the average value of the same type of elements is taken to aggregate the second metadata set corresponding to each computing power service instance, then the ability of PE1 to provide the computing power service represented by Service ID 1 and the computing power service represented by Service ID 2 (that is, the third metadata set (Cohesive Metadata Set1 as shown in Figure 6)) is {25ms, 63G, 10c}.

[0142] The process for PE2, PE3, and PE4 to determine their ability to provide the computing power service represented by Service ID 1 and the computing power service represented by Service ID 2 is similar to that of PE1 and will not be repeated here.

[0143] Then, PE4 acts as the source PE. When PE1, PE2, and PE3 publish and announce their determined computing power status information in the computing power network, PE4 can generate hierarchical table entries based on the received computing power status information corresponding to PE1, PE2, and PE3 and its own computing power status information, such as the local routing table, the global routing information table, the local forwarding information table, and the global forwarding information table. The generation process of the hierarchical table entries will not be repeated here.

[0144] Example 3

[0145] As shown in Figure 7, it is assumed that each edge device PE can be configured with a global computing power SID and a local computing power SID, as well as the corresponding global forwarding information table and local forwarding information table, and the computing power SID is flooded throughout the network by extending the IGP.

[0146] In this case, as shown in FIG7 , the hierarchical computing power routing and forwarding process provided by this application is as follows.

[0147] (a) The client sends a service message, and the global computing power SID of the connected edge device PE1 is filled in the Destination Address field of the service message, as shown in END.CG(PE 1) in Figure 7.

[0148] (a) When PE 1 receives a service message, it can parse the Service ID (which can be placed in a DOH, HBH, etc.) carried in the message header, i.e., Service ID 1, based on the global computing power SID in the Destination Address field, i.e., END.CG (PE 1), and then query the global forwarding information table associated with the global computing power SID based on Service ID 1. If the next hop is END.CL (PE 4) based on the query of Service ID 1, the END.CG (PE 1) in the Destination Address field is replaced with END.CL (PE 4), and the corresponding outbound interface information is used. In the case of Traffic Engineering (TE), the outbound interface information is Policy 1 shown in Figure 7, which encapsulates the outer message header of the original message, including SRH, source address, and destination address (take the first SID in the segment list of the policy).

[0149] (a) PE1 continues forwarding the service packet. Along the way, the service packet is forwarded along the explicit path according to the Segment List in Policy 1 until it reaches PE 4.

[0150] (a) After receiving the service message, PE4 can decapsulate the outer message header of the service message to reveal the inner message. Based on the local computing power SID in the Destination Address field, namely END.CL(PE 4), it parses the Service ID carried in the message header (which can be placed in a position such as DOH or HBH), namely Service ID 1, and then queries the associated local forwarding information table based on Service ID 1. If the next hop found based on Service ID 1 is Instance 4, 2, the computing power SID in the Destination Address field can be replaced based on the queried routing information, and the service message can be forwarded based on the corresponding outbound interface information, ultimately reaching Instance 4, 2.

[0151] It can be understood that the implementation process of this example 3 can refer to the relevant description in the aforementioned method embodiment, which will not be repeated here. Of course, the implementation process of this example 3 is only an example. In the actual forwarding process, it may have more or fewer steps than the aforementioned ones, and there is no limitation here.

[0152] Example 4

[0153] As shown in Figure 8a, PE 1 to PE 4 respectively generate their own global and local control plane table entries, namely the global routing information table and the local routing information table, by taking the minimum value of the corresponding metric values ​​of all service instances providing computing power services as the aggregation method. At this time, Instance 4,4 located at PE 4 has the smallest metric value, so it is selected at PE 4 as the instance providing the computing power service represented by Service ID 1 locally and is sent down as the local forwarding information table entry of the forwarding plane; similarly, at PE 1, PE 4 connected to Instance 4,4 is selected as the forwarding next hop of the instance providing the computing power service represented by Service ID 1 and is sent down as the global forwarding information table entry of the forwarding plane.

[0154] Based on this, if a computing power service user accesses from PE1, PE1 forwards the service packet to PE 4 according to the global computing power SID semantics and forwarding rules. If Instance 4,4 experiences an anomaly or failure, causing Instance 4,4 to become unavailable, PE 4 will detect that Instance 4,4 is unreachable and refresh its global and local control plane entries. For example, in the local routing information table, Instance 4,1, because its metric value is lower than all other entries, is issued as the new forwarding plane local forwarding information table entry, thus refreshing the local forwarding information table.

[0155] Next, PE 4 queries the new local forwarding information table based on the local computing power SID semantics and forwarding rules in the received service message and forwards the message to Instance 4,1.

[0156] It is worth noting that, please refer to Figure 8b. For the scenario of computing power routing based on a unified FIB table or a unified FIB table in the related technology, if PE4 senses that Instance 4 is unreachable and refreshes the control plane table entry and forwarding plane (i.e., computing power resource status information), the refreshed computing power resource status information will be notified to the remote PE 1. However, since it takes time for PE 4 to notify the new computing power resource status information to the remote PE 1, it may cause PE1 to be selected as the next hop routing information again before the table entry of PE 1 is refreshed, thereby forming a loop between PE1 and PE4. As for the loop problem in the related art, the present application adopts a hierarchical routing method based on the association of hierarchical computing power SID and hierarchical table entries to implement a loop-free forwarding path, which can be: as shown in the aforementioned Figure 8a, since the message reaching PE4 carries the computing power SID associated with the local forwarding information table at PE4, PE4 searches the local forwarding information table according to the semantics of the hit computing power SID and forwards the message to Instance 4, 1, instead of routing the message to PE1 again as shown in the related art.

[0157] At the same time, when PE 4 notifies the remote PE 1 of the new computing resource status information, PE 1 will refresh the corresponding table entries and forward the newly accessed services based on the refreshed new table entries to ensure the smooth forwarding of the newly accessed services.

[0158] 9 shows a schematic structural diagram of a forwarding device 900 for computing power routing provided in an embodiment of the present application. The device 900 includes: a receiving module 910, configured to receive a target message to be forwarded, the target message carrying a first computing power segment identifier SID and a service identifier of a requested specific service; a query module 920, configured to determine a target forwarding information table associated with the first computing power SID from multiple forwarding information tables, and query the target forwarding information table based on the service identifier to obtain computing power routing information for the next hop, the computing power routing information including a second computing power SID and first outbound interface information; a forwarding module 930, configured to replace the first computing power SID in the destination address field of the target message with the second computing power SID, and forward the replaced target message based on the first outbound interface information; wherein the first edge device stores multiple computing power SIDs, different computing power SIDs are associated with different forwarding information tables, and different forwarding information tables are used for message forwarding at different network layers in the computing power network, and the first computing power SID belongs to the multiple different computing power SIDs.

[0159] In some embodiments, the multiple forwarding information tables include a global forwarding information table and a local forwarding information table; the global forwarding information table is used to indicate the message forwarding routing information of the target message between the edge devices in the computing power network; the local forwarding information table is used to indicate the message forwarding routing information of the target message between the computing power service instances corresponding to the first edge device.

[0160] In some embodiments, the global forwarding information table is determined based on the global routing information table, which stores the device information of each edge device in the computing power network and the ability of each edge device to provide the specific service; the local forwarding information table is determined based on the local routing information table, which stores the service instance information of each local computing power service instance and the ability of each computing power service instance to provide the specific service.

[0161] In some embodiments, the process of determining the global routing information table includes: receiving first computing power status information sent by at least one second edge device; determining the global routing information table based on the second computing power status information and at least one of the first computing power status information, wherein the second computing power status information is obtained by aggregating the local computing power status information perceived by the first edge device.

[0162] In some embodiments, the first computing power status information and the second computing power status information are determined based on a resource pool granularity, and the resource pool includes resources corresponding to at least one computing power service instance.

[0163] In some embodiments, the perceived local computing power status information is aggregated to obtain the second computing power status information, including: respectively determining the computing network resource status metadata set corresponding to each of the computing power service instances; determining the first metric value corresponding to each of the computing power service instances based on the computing network resource status metadata set corresponding to each of the computing power service instances, the first metric value being used to characterize the ability of the computing power service instance to provide a specific service; aggregating the first metric value corresponding to each of the computing power service instances to obtain a second metric value corresponding to the second computing power status information, the second metric value being used to characterize the ability of the resource pool at the first edge device to provide the specific service.

[0164] In some embodiments, the perceived local computing power status information is aggregated to obtain the second computing power status information, including: for each computing power service instance, determining a first metadata set of computing network resource status metadata that the computing power service instance can provide for each computing power service, and determining the union of each of the first metadata sets corresponding to the computing power service instance to obtain a second metadata set; aggregating the second metadata sets corresponding to each of the computing power service instances to obtain a third metadata set corresponding to the second computing power status information.

[0165] In some embodiments, the receiving module 910 receives a target message to be forwarded, including: determining that the target message to be forwarded is received; parsing a specified field in the message header of the target message to obtain a service identifier of the specific service; wherein the specified field is a destination option extended message header DOH field, a hop-by-hop option extended message header HBH field, or a segment routing extended message header SRH field.

[0166] In some embodiments, the plurality of computing power SIDs include a global computing power SID and a local computing power SID, the global computing power SID is associated with the global forwarding information table, and the local computing power SID is associated with the local forwarding information table.

[0167] In some embodiments, the device 900 also includes a module for at least one of the following: notifying each third edge device of the second computing power status information, wherein the second computing power status information includes the second metric value or the third metadata set; notifying each third edge device of the second computing power SID.

[0168] In some embodiments, the second computing power SID and / or the second computing power status information is announced based on an extended Border Gateway Protocol (BGP) and an Interior Gateway Protocol (IGP).

[0169] In some embodiments, the query module 920 queries the target forwarding information table according to the service identifier, obtains the computing power routing information of the next hop, and then refreshes the target forwarding information table when it senses that the forwarding path corresponding to the computing power routing information of the next hop is abnormal; queries the refreshed target forwarding information table according to the service identifier to obtain the refreshed computing power routing information of the next hop, the target forwarding information table is associated with the first computing power SID, and the refreshed computing power routing information includes the third computing power SID and the second output interface information; and replaces the first computing power SID in the target message with the third computing power SID through the forwarding module, and forwards the replaced target message according to the second output interface information.

[0170] The device 900 provided in the embodiment of the present application can execute the various methods described in the above method embodiments and realize the functions and beneficial effects of the various methods described in the above method embodiments, which will not be repeated here.

[0171] FIG9 shows a schematic diagram of the hardware structure of an electronic device that implements an embodiment of the present application. Referring to the figure, at the hardware level, the electronic device includes a processor, and in some embodiments, includes an internal bus, a network interface, and a memory. Among them, the memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage. Of course, the electronic device may also include hardware required for other services.

[0172] The processor, network interface, and memory can be interconnected via an internal bus, such as an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. These buses can be classified as address buses, data buses, and control buses. For ease of illustration, the figure uses only one bidirectional arrow, but this does not imply that there is only one bus or only one type of bus.

[0173] A memory is used to store programs. The programs may include program code, which includes computer operating instructions. The memory may include internal memory and non-volatile memory, and provides instructions and data to the processor.

[0174] The processor reads the corresponding computer program from the non-volatile memory into the internal memory and then runs it, logically forming a device for locating a specific user. The processor executes the program stored in the memory and is used to perform the methods disclosed in the embodiments shown in Figures 1-3 and achieve the functions and benefits of the various methods described in the method embodiments above, which will not be repeated here.

[0175] The methods disclosed in the embodiments shown in Figures 1 to 3 of the present application can be applied to a processor or implemented by a processor. The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method can be completed by hardware integrated logic circuits in the processor or software instructions. The above processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The methods, steps, and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of the present application can be directly implemented as being executed by a hardware decoding processor, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and, in conjunction with its hardware, completes the steps of the above method.

[0176] The electronic device can also execute the methods described in the above method embodiments and realize the functions and beneficial effects of the methods described in the above method embodiments, which will not be described in detail here.

[0177] Of course, in addition to software implementation, the electronic device of this application does not exclude other implementation methods, such as logic devices or a combination of software and hardware, etc. That is to say, the execution subject of the following processing flow is not limited to each logic unit, but can also be hardware or logic devices.

[0178] An embodiment of the present application also proposes a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple applications, the electronic device executes the method disclosed in the embodiments shown in Figures 1 to 3 and realizes the functions and beneficial effects of the various methods described in the previous method embodiments, which will not be repeated here.

[0179] The computer-readable storage medium includes a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0180] In one embodiment, the embodiment of the present application also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the following process is implemented: the method disclosed in the embodiments shown in Figures 1 to 3 and the functions and beneficial effects of the various methods described in the previous method embodiments are implemented, which will not be repeated here.

[0181] In short, the above description is only a preferred embodiment of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0182] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. The computer may be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or a combination of any of these devices.

[0183] Computer-readable media includes permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media (transitory media), such as modulated data signals and carrier waves.

[0184] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0185] The various embodiments in this specification are described in a progressive manner. Similar parts between the various embodiments can be referred to in conjunction with each other. Each embodiment focuses on the differences between the other embodiments. In particular, the system embodiments are generally similar to the method embodiments, so the description is relatively simple. For relevant parts, refer to the description of the method embodiments.

Claims

1. A forwarding method for computing power routing, characterized in that: Applied to a first edge device, the method includes: Receive a target message to be forwarded, wherein the target message carries a first computing segment identifier SID and a service identifier of a requested specific service; Determine a target forwarding information table associated with the first computing power SID from multiple forwarding information tables; Query the target forwarding information table according to the service identifier to obtain the computing power routing information of the next hop, where the computing power routing information includes the second computing power SID and the first outbound interface information; Replacing the first computing power SID in the destination address field of the target message with the second computing power SID, and forwarding the replaced target message according to the first outbound interface information; Among them, the first edge device stores multiple computing power SIDs, different computing power SIDs are associated with different forwarding information tables, different forwarding information tables are used for message forwarding at different network layers in the computing power network, and the first computing power SID belongs to the multiple different computing power SIDs.

2. The method according to claim 1, characterized in that The multiple forwarding information tables include a global forwarding information table and a local forwarding information table; The global forwarding information table is used to indicate the message forwarding routing information of the target message between the edge devices in the computing power network; The local forwarding information table is used to indicate the message forwarding routing information of the target message between the computing power service instances corresponding to the first edge device.

3. The method according to claim 1, characterized in that The global forwarding information table is determined according to the global routing information table, wherein the global routing information table stores the device information of each edge device in the computing power network and the capability of each edge device to provide the specific service; The local forwarding information table is determined according to the local routing information table, and the local routing information table stores service instance information of each local computing power service instance and the capability of each computing power service instance to provide the specific service.

4. The method according to claim 3, characterized in that The determination process of the global routing information table includes: Receiving first computing power status information sent by at least one second edge device; The global routing information table is determined according to second computing power status information and at least one of the first computing power status information, where the second computing power status information is obtained by aggregating the local computing power status information perceived by the first edge device.

5. The method according to claim 4, characterized in that The first computing power status information and the second computing power status information are determined based on a resource pool as a granularity, and the resource pool includes resources corresponding to at least one computing power service instance.

6. The method according to claim 4, characterized in that Aggregating the perceived local computing power state information to obtain the second computing power state information includes: Determine the computing network resource status metadata set corresponding to each computing service instance; Determine a first metric value corresponding to each computing service instance according to a computing network resource state metadata set corresponding to each computing service instance, wherein the first metric value is used to characterize the ability of the computing service instance to provide a specific service; The first metric value corresponding to each computing power service instance is aggregated to obtain a second metric value corresponding to the second computing power status information, where the second metric value is used to characterize the ability of the resource pool at the first edge device to provide the specific service.

7. The method according to claim 3, characterized in that Aggregating the perceived local computing power state information to obtain the second computing power state information includes: For each computing service instance, determine the first metadata set of computing network resource status metadata that the computing service instance can provide for each computing service, and determine the first metadata set of computing network resource status metadata that the computing service instance can provide for each computing service. a union of the first metadata sets corresponding to the service instances to obtain a second metadata set; The second metadata set corresponding to each of the computing power service instances is aggregated to obtain a third metadata set corresponding to the second computing power status information.

8. The method according to any one of claims 1 to 7, characterized in that The receiving a target message to be forwarded includes: Determining that the target message to be forwarded is received; Parsing a specified field in a message header of the target message to obtain a service identifier of the specific service; The designated field is a destination option extended message header DOH field, a hop-by-hop option extended message header HBH field, or a segment routing extended message header SRH field.

9. The method according to any one of claims 2 to 7, characterized in that: The multiple computing power SIDs include a global computing power SID and a local computing power SID. The global computing power SID is associated with the global forwarding information table, and the local computing power SID is associated with the local forwarding information table.

10. The method according to any one of claims 4 to 7, characterized in that: The method further comprises at least one of the following: Notify each third edge device of the second computing power status information, wherein the second computing power status information includes the second metric value or the third metadata set; Notify each third edge device of the second computing power SID.

11. The method according to claim 10, characterized in that The second computing power SID and / or the second computing power status information is announced based on an extended Border Gateway Protocol BGP and an Interior Gateway Protocol IGP protocol.

12. The method according to any one of claims 1 to 11, characterized in that After querying the target forwarding information table according to the service identifier to obtain the computing power routing information of the next hop, the method further includes: When it is detected that the forwarding path corresponding to the computing power routing information of the next hop is abnormal, refreshing the target forwarding information table; According to the service identifier, query the refreshed target forwarding information table to obtain the refreshed next-hop computing power routing information, wherein the target forwarding information table is associated with the first computing power SID, and the refreshed computing power routing information includes the third computing power SID and the second outbound interface information; The first computing power SID in the target message is replaced with the third computing power SID, and the replaced target message is forwarded according to the second outbound interface information.

13. An electronic device, characterized in that: include: processor; as well as A memory arranged to store computer executable instructions which, when executed, cause the processor to perform the steps of the method of any one of claims 1 to 12.

14. A computer readable medium, characterized in that The computer-readable medium stores one or more programs, which, when executed by an electronic device including a plurality of application programs, enable the electronic device to perform the steps of the method according to any one of claims 1 to 12.

Images