CN116346938B

CN116346938B - Calculation power access method and device, electronic equipment and storage medium

Info

Publication number: CN116346938B
Application number: CN202310602809.0A
Authority: CN
Inventors: 付志华
Original assignee: New H3C Technologies Co Ltd
Current assignee: New H3C Technologies Co Ltd
Priority date: 2023-05-25
Filing date: 2023-05-25
Publication date: 2023-08-18
Anticipated expiration: 2043-05-25
Also published as: CN116346938A

Abstract

The embodiment of the application provides a computing power access method, a computing power access device, electronic equipment and a storage medium, and relates to the technical field of communication, wherein the computing power access method comprises the following steps: acquiring target path information from a first computing gateway to a second computing gateway, wherein the first computing gateway is a computing gateway connected with computing request equipment, and the second computing gateway is a computing gateway connected with target equipment for providing target computing resources; filling the target path information in a first preset message header of an application layer to obtain a first message header; encapsulating a first message header for an original message to obtain a calculation power access message; and sending the computational power access message to the first computational power gateway so that the first computational power gateway forwards the computational power access message to the target equipment according to the target path information. By applying the technical scheme provided by the embodiment of the application, the large-scale evolution of the power calculation network can be promoted, and the updating speed of the power calculation network characteristic can be improved.

Description

Calculation power access method and device, electronic equipment and storage medium

Technical Field

The present application relates to the field of communications technologies, and in particular, to a method and apparatus for accessing computing power, an electronic device, and a storage medium.

Background

The computing power network is a novel information infrastructure and provides integrated services for clients by sensing network states and cloud computing power states. Currently, each of the power networks uses a SRv6 (Segment Routing over IPv, IPv 6-based segment routing) network as a data plane, each node in the power network selects SRv a path based on network information and power information, obtains power services through the selected SRv path, and provides the power services to users.

Currently, most networks are also IPv4 networks, the coverage of the IPv6 network is smaller, and SRv is realized based on the IPv6 network, which results in that the computing network can only be deployed under a specific private network and cannot be evolved on a large scale. When SRv is used as a data plane to transfer network information and computing power information, the network information and computing power information are carried in an optional TLV (Type Length Value ) field of SRH (Segments Route Header, segment routing header), the SRH optional TLV field needs to be processed in a Linux kernel, and the influence of changing an kernel checking operation system is large, so that the computing power network characteristics cannot be updated rapidly.

Disclosure of Invention

The embodiment of the application aims to provide a computing power access method, a computing power access device, electronic equipment and a storage medium, which are used for promoting the large-scale evolution of a computing power network and improving the updating speed of the characteristics of the computing power network. The specific technical scheme is as follows:

In a first aspect, the present application provides a computing power access method applied to a computing power request device, the method comprising:

acquiring target path information from a first computing gateway to a second computing gateway, wherein the first computing gateway is the computing gateway connected with the computing request equipment, and the second computing gateway is the computing gateway connected with target equipment for providing target computing resources;

filling the target path information in a first preset message header of an application layer to obtain a first message header;

encapsulating the first message header for the original message to obtain a calculation power access message;

and sending the power access message to the first power gateway so that the first power gateway forwards the power access message to the target equipment according to the target path information.

In some embodiments, the step of obtaining target path information for the first computing force gateway to the second computing force gateway comprises:

acquiring target computing power information and target network information;

filling the target calculation force information and the target network information in a second preset message header of an application layer to obtain a second message header;

encapsulating the second message header for the original message to obtain a calculation force request message;

Sending the power calculation request message to the first power calculation gateway so that the first power calculation gateway extracts the target power calculation information and the target network information from a second message header carried by the power calculation request message, determines target equipment conforming to the target power calculation information and the target network information, and determines target path information from the first power calculation gateway to a second power calculation gateway connected with the target equipment;

when a computing force response message sent by the first computing force gateway is received, acquiring the target path information carried by the computing force response message from the computing force response message.

In some embodiments, the second header includes a second message type field, a power calculation region identification field, a bitmap field, a length field of a power calculation service name, a power calculation service name field, a transaction identification field, a power calculation and network constraint field;

the second message type field is used for filling the type identifier of the calculation request message;

the power calculation region identification field is used for filling the identification of the power calculation region of the request;

the bitmap field is used for indicating whether the computing power and network constraint field carries computing power information and network information;

The length field of the service name is used for filling the length of the service name;

the power calculating service name field is used for filling the power calculating service name;

the transaction identification field is used for filling the transaction identification of the current calculation force request;

the computing power and network constraint field is used to populate the target computing power information and target network information.

In some embodiments, the power response message includes a third header including a third message type field, a power region identification field, a segment length field, a power service address length field, a transaction identification field, a power service address field, and a Duan Liebiao field;

the third message type field is used for filling the type identifier of the calculation force response message;

the computing power area identification field is used for filling the identification of the computing power area of the computing power resource pool where the target equipment is located;

the segment length field is used for filling the length of the target path information;

the power-calculation service address length field is used for filling the length of the address of the target device;

the transaction identification field is used for filling the transaction identification of the current computing force response;

the power-calculation service address field is used for filling the address of the target device;

The segment list field includes one or more elements for populating the target path information.

In some embodiments of the present invention, in some embodiments,

the method further comprises the steps of:

receiving a first message sent by the first computing gateway;

and when the destination port of the application layer carried by the first message is a preset port, determining that the first message is a calculation force response message.

In some embodiments, the first header includes a first message type field, a flow identification length field, a flow identification field, a segment type field, a segment routing header length field, a segment list field, a current segment index field, and a first segment index field;

the first message type field is used for filling the type identifier of the computing power access message;

the flow identification length field is used for filling the length of the identification of the flow to which the computing power access message belongs;

the flow identification field is used for filling the identification of the flow to which the computing power access message belongs;

the segment routing header length field is used to populate the length of the segment list;

the segment list field includes one or more elements for populating the target path information;

the current segment index field is used for filling the sequence number of the current element to be searched in the segment list;

The first segment index field is used for filling a sequence number of a first element of the segment list;

the segment type field is used to populate the type of element in the segment list.

In some embodiments, the first header further includes: a source port field, a source address field, and a source address length field;

the source port field is used for filling a source port of the flow indicated by the flow identification field;

the source address field is used for filling the source address of the flow indicated by the flow identification field;

the source address length field is used to populate a length of a source address indicated by the source address field.

In a second aspect, the present application provides a method for accessing a computing power, which is applied to a target node in a computing power network, where the target node is any node on a path from a first computing power gateway to a second computing power gateway in the computing power network, the first computing power gateway is a computing power gateway connected with a computing power request device, and the second computing power gateway is a computing power gateway connected with a target device that provides a target computing power resource; the method comprises the following steps:

receiving a calculation power access message;

extracting a first message header of an application layer carried by the computing power access message, wherein the first message header comprises target path information from the first computing power gateway to the second computing power gateway;

And forwarding the power calculation access message to the target equipment according to the target path information.

In some embodiments, the method further comprises:

receiving a second message sent by upstream equipment;

and when the destination port of the application layer carried by the second message is a preset port, determining the second message as a calculation access message.

In some embodiments, the step of forwarding the power access packet to the target device according to the target path information includes:

when the target node is determined to be a first computing gateway or an intermediate node according to the target path information, acquiring the address of a next-hop node from the target path information; updating the destination address of the calculation power access message into the acquired address to obtain an updated calculation power access message; forwarding the updated calculation power access message;

when the target node is determined to be a second computing gateway according to the target path information, stripping the first message header to obtain an original message; and forwarding the original message.

In some embodiments, the first header further includes a source port field, a source address field, and a source address length field; the source port field is used for filling a source port of a flow to which the computing power access message belongs; the source address field is used for filling the source address of the flow to which the computing power access message belongs; the source address length field is used for filling the length of the source address indicated by the source address field;

The method further comprises the steps of:

when the target node is determined to be a first computing gateway according to the target path information, acquiring a source address and a source port of the computing access message; filling the acquired source address and source port into a source port field and a source address field of the first message header;

the step of stripping the first message header to obtain an original message comprises the following steps: copying the source address of the source address field of the first message header to the source address field of the IP header of the power-calculation access message, copying the source port of the source port field of the first message header to the source port field of the UDP header of the power-calculation access message, and stripping the first message header to obtain an original message.

In some embodiments, when the target node is a first computing gateway, the method further comprises:

receiving an calculation force request message sent by the calculation force request equipment;

extracting the target calculation power information and target network information from a second message header carried by the calculation power request message;

determining target equipment conforming to the target computing power information and the target network information;

determining target path information of a second computing gateway connected from the first computing gateway to the target device;

Filling the target path information in a third preset message header of an application layer to obtain a third message header;

encapsulating the third message header for the original message to obtain a calculation force response message;

and sending the calculation response message to the calculation request terminal so that the calculation request terminal obtains the target path information.

In some embodiments, the third header includes a third message type field, a power calculation region identification field, a segment length field, a power calculation service address length field, a transaction identification field, a power calculation service address field, and a Duan Liebiao field;

the computing power area identification field is used for filling a computing power area of a computing power resource pool where the target equipment is located;

In some embodiments, the first header includes a first message type field, a source address length field, a flow identification length field, a source port field, a source address field, a flow identification field, a segment type field, a segment routing header length field, a segment list field, a current segment index field, and a first segment index field;

the source address length field is used for filling the length of the source address indicated by the source address field;

In a third aspect, the present application provides a computing power access apparatus for use in a computing power requesting device, the apparatus comprising:

The system comprises an acquisition module, a calculation module and a calculation module, wherein the acquisition module is used for acquiring target path information from a first calculation gateway to a second calculation gateway, the first calculation gateway is a calculation gateway connected with the calculation request equipment, and the second calculation gateway is a calculation gateway connected with target equipment for providing target calculation resources;

the filling module is used for filling the target path information into a first preset message header of an application layer to obtain the first message header;

the encapsulation module is used for encapsulating the first message header for the original message to obtain a calculation power access message;

and the sending module is used for sending the power calculation access message to the first power calculation gateway so that the first power calculation gateway forwards the power calculation access message to the target equipment according to the target path information.

In some embodiments, the obtaining module is specifically configured to:

acquiring target computing power information and target network information;

In some embodiments, the computing power access device may further include:

the receiving module is used for receiving the first message sent by the first computing gateway;

and the determining module is used for determining the first message as a calculation response message when the destination port of the application layer carried by the first message is a preset port.

In a fourth aspect, an embodiment of the present application provides a computing power access apparatus, applied to a target node in a computing power network, where the target node is any node on a path from a first computing power gateway to a second computing power gateway in the computing power network, the first computing power gateway is a computing power gateway connected to a computing power request device, and the second computing power gateway is a computing power gateway connected to a target device that provides a target computing power resource, and the computing power access apparatus includes:

the first receiving module is used for receiving the computing power access message;

the extraction module is used for extracting a first message header of an application layer carried by the computing power access message, wherein the first message header comprises target path information from the first computing power gateway to the second computing power gateway;

and the forwarding module is used for forwarding the computing power access message to the target equipment according to the target path information.

In some embodiments, the computing power access device may further include:

the second receiving module is used for receiving a second message sent by the upstream equipment;

and the first determining module is used for determining the second message as a calculation power access message when the destination port of the application layer carried by the second message is a preset port.

In some embodiments, the forwarding module is specifically configured to:

in this case, the above-described computing force access device may further include:

the first filling module is used for acquiring a source address and a source port of the computing power access message when the target node is determined to be a first computing power gateway according to the target path information; filling the acquired source address and source port into a source port field and a source address field of the first message header;

The forwarding module is specifically configured to: copying the source address of the source address field of the first message header to the source address field of the IP header of the power-calculation access message, copying the source port of the source port field of the first message header to the source port field of the UDP header of the power-calculation access message, and stripping the first message header to obtain an original message.

In some embodiments, the computing power access device may further include:

the third receiving module is used for receiving the calculation force request message sent by the calculation force request device when the target node is the first calculation force gateway;

the extraction module is used for extracting the target calculation power information and the target network information from a second message header carried by the calculation power request message;

a second determining module, configured to determine a target device that conforms to the target computing power information and the target network information;

a third determining module, configured to determine target path information of a second computing gateway connected from the first computing gateway to the target device;

the second filling module is used for filling the target path information into a third preset message header of the application layer to obtain a third message header;

the encapsulation module is used for encapsulating the third message header for the original message to obtain a calculation force response message;

And the sending module is used for sending the calculation response message to the calculation request terminal so as to enable the calculation request terminal to acquire the target path information.

In a fifth aspect, the present application provides an electronic device comprising a processor and a machine-readable storage medium storing machine-executable instructions executable by the processor, the processor being caused by the machine-executable instructions to: any computational power access method applied to the computational power request equipment or any computational power access method applied to the target node is realized.

In a sixth aspect, the present application provides a computer readable storage medium having stored therein a computer program which, when executed by a processor, implements any one of the computing power access methods applied to the computing power requesting device, or implements any one of the computing power access methods applied to the target node.

In a seventh aspect, embodiments of the present application also provide a computer program product comprising instructions that, when run on a computer, cause the computer to perform any of the above-described computational power access methods applied to a computational power requesting device, or any computational power access method applied to a target node.

The embodiment of the application has the beneficial effects that:

in the technical scheme provided by the embodiment of the application, the message header of the application layer, such as the first preset message header, is defined. And filling target path information through the first preset message header, so that the computing force request equipment can access the target equipment along a path corresponding to the target path information, and the computing force service is acquired. Because the target path information is carried in the first preset message header of the application layer, the first preset message header can be filled with the target path information, the application layer can identify the address of any type of network, therefore, the expression form of the target path information can not be limited by the address of the IPv6 network, and the address of the IPv4 network or other types of networks can also be limited, thereby greatly promoting the large-scale evolution of the computing network. In addition, in the embodiment of the application, the target path information is filled in the message header of the application layer, and belongs to the operation of the application layer, the operation of the application layer is user mode operation, the influence of the user mode operation on the operating system is small, the updating of the computing power network characteristic is facilitated, and the updating speed of the computing power network characteristic is further improved.

Of course, it is not necessary for any one product or method of practicing the application to achieve all of the advantages set forth above at the same time.

Drawings

In order to more clearly illustrate the embodiments of the application or the technical solutions in the prior art, the drawings used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the application, and other embodiments may be obtained according to these drawings to those skilled in the art.

FIG. 1 is a schematic diagram of a centralized computing network in the related art;

FIG. 2 is a schematic diagram of a distributed computing power network in the related art;

FIG. 3 is a diagram of a message format SRv in the related art;

FIG. 4 is a schematic diagram of an SRH format according to the related art;

FIG. 5 is a schematic diagram of message forwarding in the related art;

FIG. 6 is another diagram of a message format SRv in the related art;

FIG. 7 is a schematic diagram of a computing power access system according to an embodiment of the present application;

FIG. 8 is a schematic structural diagram of a networking architecture of a computing power network according to an embodiment of the present application;

FIG. 9 is a first flowchart of a method for accessing computing power according to an embodiment of the present application;

Fig. 10 is a schematic diagram of a first header format according to an embodiment of the present application;

FIG. 11 is a schematic diagram of an IPv4 address packet according to an embodiment of the present application;

FIG. 12 is a schematic diagram of an IPv6 address packet according to an embodiment of the present application;

FIG. 13 is a detailed view of step S91 in FIG. 9;

fig. 14 is a schematic diagram of a second header format according to an embodiment of the present application;

fig. 15 is a schematic diagram of a third header format according to an embodiment of the present application;

FIG. 16 is a second flow chart of a method for accessing computing power according to an embodiment of the present application;

FIG. 17 is a third flow chart of a method for accessing computing power according to an embodiment of the present application;

FIG. 18 is a schematic diagram of a computing power access device according to an embodiment of the present application;

FIG. 19 is a schematic view of another architecture of a computing power access device according to an embodiment of the present application;

fig. 20 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by the person skilled in the art based on the present application are included in the scope of protection of the present application.

Currently, new generation information technologies such as 5G (5 th Generation Mobile Communication Technology, fifth generation mobile communication technology), AI (Artificial Intelligence ), big data, etc. are remodelling the world, and the intelligent world of "everything interconnected, everything perceived, everything intelligent" is accelerating. GSMA (Global System for Mobile communications Association, global mobile communications system association) data shows that, by 2025, global 5G permeability would be expected to reach 24%; while IDC (Internet Data Center ) predicts that the global internet of things market size will reach $ 1.1 trillion by 2025. In the ages of intelligent internetworking, data are production factors, and calculation power is productivity. The huge market facing industry digital transformation, each industry has put forward urgent demands on computing power and networks, and next generation network transformation represented by computing power networks gradually becomes a hot spot of common attention of all industry parties.

In a broad sense, an optical network is a new type of information infrastructure that allocates and flexibly schedules computing resources, storage resources, and network resources as needed among clouds, networks, edges, according to business needs. In popular terms, the power calculation ratio is electricity, the power calculation network ratio is a power grid, and the power grid supports the use of various electric appliances in daily life of people. In the universal intelligent networking era, the power network can meet the real-time computing requirements of emerging applications such as automatic driving, cloud gaming, face recognition, virtual Reality (VR), augmented Reality (AR Augmented Reality) and the like.

The industry divides the construction and development of the power network into 3 stages, and the 3 stages are respectively: a ubiquitous collaboration stage (starting stage), a fusion unification stage (development stage), and an integral endophytic stage (crossing stage). In the starting stage, the core concept is 'collaboration', and a network with 'collaboration' characteristics such as network follow-up, collaborative arrangement, collaborative operation, one-station service and the like is created, so that the computing power is more stereoscopic and ubiquitous. In the development stage, a network with 'fusion' characteristics such as calculation network fusion, intelligent arrangement, unified operation, fusion service and the like is created, so that the network is connected with cloud, edge and end ubiquitous calculation power resources, and various novel business requirements are met. In the crossing stage, the network strong calculation is promoted, the advantages of gathering calculation power and exerting calculation power clusters are realized by using a network, the core concept is 'integration', the integrated calculation network, the symbiotic calculation network, the intelligent endophytic, the innovative operation and the integrated service are realized, and the system of 'network in calculation and network in calculation' is realized.

In general, a power network is a new type of information infrastructure. The algorithm here exists in the form of cloud in general, including a center cloud and an edge cloud. The computing power network provides integrated services for clients by sensing network states and cloud computing power states at the same time, and the computing power network has two forms of a centralized computing power network and a distributed computing power network on a technical route.

As shown in fig. 1, fig. 1 is a schematic structural diagram of a centralized computing power network in the related art, where the centralized computing power network includes a central control computing network brain, a network controller, a cloud management platform, an edge cloud, a cloud access device and a network access device. In the embodiment of the present application, the access device may be a PE (Provider Edge) device, and based on this, the cloud access device may be simply referred to as a cloud PE, and the network access device may be simply referred to as a network PE. The cloud PE and the network PE are located in a cloud private network. The computing network brain is responsible for collecting computing network information of the whole network, including computing power information and network information, maintaining computing power topology and network topology, selecting a proper cloud for a user based on the maintained computing power information and network information, and opening a network path.

As shown in fig. 2, fig. 2 is a schematic structural diagram of a distributed computing power network in the related art, where the distributed computing power network includes a computing network brain, a network controller, a cloud management platform, an edge cloud, a cloud PE and a network PE. The distributed computing power network transmits computing power information in the network, so that all router nodes in the network maintain computing power information and routing information, and policy routing is performed based on comprehensive consideration of the network information and the computing power information by each router node, so that proper computing power service is provided for users.

Currently, the power network uses a bearer network based on SRv6 technology as a bottom network, that is, SRv network is used as a data plane, each node in the power network selects SRv a path based on network information and power information, and power services are acquired through the selected SRv path.

SRv6 technology is based on IPv6 networks and uses a source routing mechanism for forwarding. SRv6 technology mainly complies with the following RFC (Request For Comment, request for comments) protocol standards:

RFC8402, SRA (Segment Routing Architecture, segmented routing architecture): describing a source routing technical architecture;

RFC8754, IPv6 SRH: describe SRv source routing header message format;

RFC8986, SRv6 network programming (Network Programming): SRv6 network programming approach is described.

SRv6 an IPv6 extension header, namely an SRH, is newly defined by adopting a routing extension header defined in the IPv6 standard specification (RFC 2460), the extension header specifies an explicit path of IPv6, and stores Segment List (Segment List) information of IPv6, an SRH extension header is added to an IPv6 message by a header node, and an intermediate SRv node can process and forward according to path information contained in the SRH extension header, instead of the SRv node only needing to perform conventional forwarding according to a standard destination IPv6 address.

As shown in fig. 3, fig. 3 is a schematic diagram of a format of a SRv message in the related art, and SRv message includes a MAC (Media Access Control, medium access control) header, an IPv6 standard header, an SRH and a message payload. The format of the SRH is shown in fig. 4, where the SRH includes:

the Next Header (Next Header), 8bits long, is used to identify the type of the Next Header.

The SRH length (Hdr Ext Len), which is 8bits in length, represents the length of the SRH in 8 bytes, excluding the first 8 bytes.

The Routing Type (Routing Type) is 8bits long, and the value is 4, which indicates that the SRH is carried.

Segment index (SL), 8bits long, indicates the number of segment identifiers (Segment Identifier, SIDs) to be found, and the initial value is n, where n indicates the number of SIDs encapsulated in the SRH, and the value of SL is decremented by 1 every node.

The Last hop (Last Entry), 8bits in length, is the number of the first SID of the actual forwarding path of the message in the SRH.

Flag bits (Flags), 8bits in length, are flag bit information.

A Tag (Tag) of length 16bits for marking a group of messages having the same characteristics.

Segment List (Segment List), which is a SID List, shows an IPv6 address of 128 bits (bits), including Segment List [0] -Segment List [ n-1], and the nodes on the packet forwarding path are arranged in order from far to near, i.e., segment List [0] shows the last SID of the forwarding path, segment List [1] shows the last-to-last SID of the forwarding path, and so on.

The Optional TLV (Type Length Value ) objects (variable) is an Optional TLV object (variable).

In the SRv network, each time a SRv node passes, the SRv node reads information in the SRH, executes related SID instructions, copies the active Segment List to the destination address in the IPv6 standard header, decrements the SL (Segments Left) field by 1, updates the related pointer, and finally performs IPv 6-only "very simple" forwarding.

By way of example, as shown in FIG. 5, there are five nodes in SRv domain, namely node 1, node 2, node 3, node 4 and node 5, wherein node 1, node 3, node 4 and node 5 are the nodes supporting SRv6, simply referred to as SRv6 node, node 2 is a normal IPv6 node, the SID of node 1 is 2001:1, the SID of node 3 is 2001:3, the SID of node 4 is 2001:4, and the SID of node 5 is 2001:5; the data packet is forwarded from node 1 to node 5, node 1 being the source node and node 5 being the tail node. Wherein the data packet includes an IPv6 standard header, an SRH, and a payload. The IPv6 standard header (IPv 6 Hdr) includes a Version number (Version), a Flow Class (Traffic Class), a Flow Label (Flow Label), a Payload length (Payload length), a Next header (Next), a Hop limit (Hop limit), a SA (source address), and a DA (Destination Address ). The example shown in fig. 5, at node 1, the SA is the SID of node 1, 2001::1, and the da is the SID of node 3, 2001::3, next header (Next) =43. The structure of the SRH is described in the section of fig. 4 above. The example shown in fig. 5, at node 1, the SRH includes segment lists 2001:5, 2001:4, 2001:3, sl=2, route type=4, SRH length (Len) =6, and the last hop value is 2.

The forwarding process of the data packet from node 1 to node 5 is as follows:

a. the data packet enters the node 1, the node 1 adds SRH to the data packet, and specifies the relevant operation of the SRv node in the whole path, and the outer layer encapsulates the IPv6 standard header. Since 3 nodes SRv need to pass from node 1 to node 5, the SRH includes three Segment lists, and SL corresponding to the first Segment List of the forwarding path is 2; at this point, the SA of the IPv6 standard header of the outer layer is 2001:1, i.e., the SID of node 1, and DA is 2001:3, i.e., the SID of node 3, copied from Segment List [2] in the SRH. Node 1 forwards the packet according to the DA in the IPv6 standard header.

b. The data packet is forwarded from the node 1 to the node 2, because the node 2 only supports the conventional IPv6 and does not support SRv, when the node 2 receives the data packet, the node 3 does not process the extension header when the DA of the data packet is not the network segment address of the node itself according to the rule of IPv6 RFC, and forwards the data packet directly according to the DA in the external layer IPv6 standard header, namely, forwards the data packet to the node 3.

c. When the node 3 receives the data packet, according to DA of the outer layer IPv6 standard header, namely 2001:3, searching a local SID table, if the local SID table is hit, subtracting 1 from an SL field after executing a related instruction, pointing a pointer to Segment List [1] (this is an active Segment List), and copying an address of the Segment List [1] to DA of the outer layer IPv6 standard header, wherein DA of the outer layer IPv6 standard header is 2001:4, namely the address of the node 4. The node 3 forwards the data packet according to the destination address of the outer layer IPv6 standard header, i.e. forwards the data packet to the node 4.

d. The processing procedure of the node 4 is consistent with that of the node 3, the node 4 searches a local SID table according to DA of an external layer IPv6 standard header, namely 2001:4, if the local SID table is hit, after a related instruction is executed, the SL field is subtracted by 1, the pointer points to Segment List [0], and the address of the Segment List [0] is copied to DA in the external layer IPv6 standard header, wherein DA of the external layer IPv6 standard header is 2001:5, namely SID of the node 5. The data packet is forwarded according to the DA of the outer layer IPv6 standard header, i.e. the data packet is forwarded to the node 5.

e. When the node 5 receives the data message, it recognizes that the DA of the outer layer IPv6 standard header is the SID of the node, and the SL field value is 0. At this point, node 5 strips the SRH and IPv6 standard header, reads the real Payload (Payload), and completes forwarding according to the relevant instruction in the SID.

In the related art, an edge router such as a PE device selects a cloud and a corresponding path (for example, SRv paths) that meet the requirements of a user SLA (Service Level Agreement ) for a user according to a computing power routing table maintained by the edge router, such as the node 1 described above. After selecting the path for the user, the edge router forwards the message.

SRv6 Segment is typically abbreviated SRv6 SID or SID, which presents a 128bit (bits) IPv6 address, and the Segment list presents an ordered set of IPv6 address lists inserted in the SRH, not all IPv6 addresses being SIDs. When SRv node explicitly declares a certain IPv6 address as a SID, that IPv6 address is a SID. The SID may also be the address of a certain node or the address of a certain interface, where the SID contains related instructions and carriable parameters. The SRv6 message shown in fig. 6, which includes an IPv6 standard header, an SRH, and a payload, the SRH includes a plurality of SIDs, such as the 128bits SID shown in fig. 6. SID is composed of two parts, namely a route location (Locator) and a Function (Function), wherein the Locator represents an identifier allocated to one network node in a network and can be used for routing and forwarding data packets; the Function represents an instruction of the device, can support customization, and is mainly used for indicating the node to perform corresponding operation. The Function part can also divide an optional parameter segment (parameters). Thus, the format of the SID may also be: the Locator is located in the field, wherein the Locator occupies the high-order bits of the IPv6 address, the Function occupies the rest of the IPv6 address, and the optional parameter is located in the field, occupies the low-order bits of the IPv6 address.

While IPv6 networks are already under construction, the internet is currently largely an IPv4 network, while SRv6 is necessarily IPv 6-based, and thus suffers from large-scale deployment problems if the computing network uses SRv networks as the data plane. Moreover, the current computational power network based on SRv can only be deployed under certain specific private networks, and is not suitable for large-scale evolution. In addition, if the SRv network is used as the data plane, the upper layer application is required to carry the calculation information and the calculation requirement in the Optional TLV object variable field of the SRH. The variable field of the Optional TLV object of SRv is generally processed in the Linux kernel, and the influence of changing the kernel to the operating system is large and inconvenient, so that the computing power network characteristics such as the computing power information, the network information and the path information cannot be updated quickly.

To solve the above-mentioned problems, an embodiment of the present application provides a computing power access system, as shown in fig. 7, including a computing power request device and a plurality of nodes, where the plurality of nodes may include a computing power gateway and an intermediate node, where the computing power gateway is connected to the computing power request device and/or connected to a device that provides computing power resources, where the device that provides computing power resources is located in a computing power resource pool, and one computing power resource pool may include one or more devices, such as device 1-device N in fig. 7, and each device is capable of providing a corresponding computing power resource, that is, a corresponding computing power service.

Based on the foregoing power access system, an embodiment of the present application provides a networking architecture of a power network, as shown in fig. 8, where the power network includes power resource pools site1 and site2, a power request device (Host), power gateways CGW1-CGW3, and intermediate nodes a-E. Fig. 8 illustrates only 2 computing resource pools, 1 computing request device, 3 computing gateways, and 5 intermediate nodes, and is not limiting. Wherein:

calculation force requesting device (Host): a client (source) for issuing a calculation request.

Computing power resource pool (Site): may be in the form of a center cloud, an edge cloud, etc. The service end providing the computing power service, i.e. the device providing the computing power resource, is located within site (destination end). The pool of computational resources may also be referred to as a pool of computational resources.

Computing Gateway (CGW): and providing a calculation force request service and a calculation force registration service respectively by the Host and the site.

Intermediate Node (Node): i.e. a force-calculating intermediate router, a router for transmitting force-calculating information and forwarding force-calculating service messages.

In the power computing network, each power computing resource pool is accessed to the whole power computing network through a power computing gateway. Each computing power resource pool needs to specify a corresponding computing power gateway and register computing power information such as CPU, gpu and the like of the computing power resource pool to the connected computing power gateway. The computing power information here is the computing power information of each device included in the computing power resource pool, that is, each device registers its computing power information to the computing power gateway of the computing power resource pool to which it belongs. The computing force gateway announces the collected computing force information in the whole network and receives the computing force information announced by other routers. The whole power network can be an IPv4 network or an IPv6 network, and the intermediate node does not need to fully support the power access method provided by the embodiment of the present application, that is, the intermediate node does not need to fully support the data plane of the power network described in the embodiment of the present application, for example, in fig. 8, the node C-node E may be a node that only supports normal IPv4 or IPv6 forwarding, and does not support the data plane of the power network described in the embodiment of the present application, and the power request device, the power gateway, and part of the intermediate nodes (for example, the node a and the node B in fig. 8) support the data plane described in the embodiment of the present application. For ease of illustration, the intermediate nodes described below all refer to nodes that support the data plane described in embodiments of the present application.

Based on the foregoing algorithm power access system, an embodiment of the present application provides an algorithm power access method, where a header of an application layer, such as a first preset header, is defined in the method. And filling target path information through the first preset message header, so that the computing force request equipment can access the target equipment along a path corresponding to the target path information, and the computing force service is acquired. Because the target path information is carried in the first preset message header of the application layer, the first preset message header can be filled with the target path information, the application layer can identify the address of any type of network, therefore, the expression form of the target path information can not be limited by the address of the IPv6 network, and the address of the IPv4 network or other types of networks can also be limited, thereby greatly promoting the large-scale evolution of the computing network. In addition, in the embodiment of the application, the target path information is filled in the message header of the application layer, and belongs to the operation of the application layer, the operation of the application layer is user mode operation, the influence of the user mode operation on the operating system is small, the updating of the computing power network characteristic is facilitated, and the updating speed of the computing power network characteristic is further improved.

The following describes in detail the method for accessing computing power provided by the embodiment of the present application through a specific embodiment.

Referring to fig. 9, fig. 9 is a first flowchart of a computing power access method according to an embodiment of the present application, where the method is applied to a computing power request device, and includes the following steps:

step S91, obtaining target path information from a first computing gateway to a second computing gateway, wherein the first computing gateway is a computing gateway connected with computing request equipment, and the second computing gateway is a computing gateway connected with target equipment for providing target computing resources.

In the embodiment of the application, the target device is a device for satisfying the calculation force request of the calculation force request device, the calculation force request device is connected with one or more calculation force gateways, the first calculation force gateway can be any calculation force gateway connected with the calculation force request device, the target device is connected with one or more calculation force gateways, the second calculation force gateway can be any calculation force gateway connected with the target device, and the calculation force resource pool comprising the target device is the target calculation force resource pool. The target path information is the path information from the first computing gateway to the second computing gateway. The path information may include addresses of nodes on a path from the first computing gateway to the second computing gateway, where the nodes are nodes supporting the data plane described in the embodiments of the present application, and the addresses may be IPv4 addresses or IPv6 addresses, which is not limited. When the address of the node is an IPv4 address, the path information may further include an upper application port number, so that the upper application calls a port (port) corresponding to the upper application port number, and performs related operations such as security or format conversion on the packet.

When the computational power services such as image rendering, AI training and reasoning, high-performance computing and the like are required to be acquired, the computational power request equipment acquires target path information so as to facilitate the subsequent access to target equipment in a target computational power resource pool along a path indicated by the target path information, and the required computational power services are acquired.

Step S92, filling the target path information in a first preset message header of the application layer to obtain a first message header.

In the embodiment of the application, a message header template processed by an application layer, such as a first preset message header, is preset in the calculation force request device. The computing force request equipment fills the acquired target path information into a first preset message header, and the first preset message header filled with the target path information is the first message header. In the embodiment of the present application, the first packet header may be referred to as a CNC (Computing and Network Convergence, computing network) header, where the application layer is located above a transport layer to which UDP (User Datagram Protocol, user data protocol) belongs, and the first packet header may also be referred to as a CNC over UDP (CNCoU) header.

In some embodiments, the first header may include a first message Type (Msg Type) field, a flow identification length (FlowID Len) field, a flow identification (FlowID) field, a Segment Type (Segment Type) field, a Segment routing header length (SR Hdr Len) field, a Segment List (Segment List) field, a current Segment index (Segment Left) field, and a first Segment index (Last Entry) field, as shown in fig. 10. In fig. 10, numerals 0 to 9 of the first row denote bit positions.

The first message type field is used for filling type identification of the power access message. Here, the value of the first message type field indicates that the message is a Computing transit (arithmetic access) message.

The flow identification length field is used to populate the length of the identification of the flow to which the computational access message belongs. The stream identifier is an ID (Identity) of a stream, and can be used for tracking (trace).

The flow identification field is used to populate an identification of the flow to which the computational power access message belongs. In the embodiment of the present application, the flow identification field is an optional field, that is, the first header may not include the flow identification field.

The segment routing header length field is used to populate the length of the segment list.

The segment list field includes one or more elements for populating the target path information. The Segment List field refers to the field where the Segment List [0] -Segment List [ n ] is located, namely the field where the Segment List [0] -Segment List [ n ] is located, wherein the Segment List [0] -Segment List [ n ] respectively represents one element of the Segment List, and the Segment List [0] -Segment List [ n ] forms a complete path from the first computing gateway to the second computing gateway.

The current segment index field is used for filling the sequence number of the current element to be searched in the segment list. The initial value of the current segment index field is n, n represents the number of elements encapsulated in the segment list, and the value of the current segment index field is subtracted by 1 every time a node is passed.

The first segment index field is used to populate the sequence number of the first element of the segment list. The value of the first segment index field is the number of the first element of the actual forwarding path of the message in the segment list.

The segment type field is used to populate the types of elements in the segment list. For example, the value of the segment type field is 1, which indicates that the type of the element in the segment list is in the form of IPv4 address+port (port); the value of the segment type field is 2, which indicates that the type of the element in the segment list is in IPv6 form. In the embodiment of the application, the segment type field can adopt other values to distinguish different types of elements, which is not particularly limited.

In other embodiments, the first header may further include: a Source Port (Source Port) field, a Source address (Source address) field, and a Source address length (SRC Len) field, as shown in fig. 10.

The source port field is used to populate the source port of the flow indicated by the flow identification field. The flow indicated by the flow identification field is the flow to which the calculation power access message belongs.

The source address field is used to populate the source address of the flow indicated by the flow identification field. The source address is the address of the head node, such as node 1 in fig. 5.

The source address length field is used to populate the length of the source address indicated by the source address field. Wherein the length of the IPv4 address is 4 bytes, and the length of the IPv6 address is 16 bytes.

In the embodiment of the application, the source address and the source port are stored in the first message header. In this way, the problem that the destination cannot trace the source due to the modification of the source address by NAT (Network Address Translation ) and the like in the forwarding process can be avoided.

In the embodiment of the present application, the first header may further include an optional TLV, that is, an optional TLV object (variable), so that additional information that needs to be used is carried by the first header. In addition, the length of each field in the first header may be set according to actual requirements, and fig. 10 is only an example and is not meant to be limiting.

Step S93, the first message head is packaged for the original message, and the calculation power access message is obtained.

The computing force request device obtains an original message for accessing the computing force service. And after the first message header is obtained, the computing power request equipment encapsulates the first message header for the original message to obtain a computing power access message. The power access message may include a first message header and UDP header and IP header required for transmission. The structure of the power access messages for different types of IP addresses can be seen in fig. 11 and 12. Fig. 11 is a message structure of an IPv4 address, including an IPv4 header, a UDP header, a CNCoU header, and an initial Packet (original Packet). Fig. 12 is a message structure of an IPv6 address, including an IPv6 header, a UDP header, a CNCoU header, and an initial packet.

Step S94, sending the computing power access message to the first computing power gateway so that the first computing power gateway forwards the computing power access message to the target equipment according to the target path information.

The power request equipment sends a power access message to a first power gateway, the first power gateway can forward the power access message to target equipment according to target path information carried in the power access message, namely, the power access message is forwarded to a second power gateway from the first power gateway along a path indicated by the target path information, the second power gateway strips a first message header of the power access message to obtain an original message, and the power access message is restored to a standard UDP message (original message); and forwarding the original message to target equipment in the target computing power resource pool. And then the target device in the target computing power resource pool provides computing power service for the computing power request device.

In some embodiments, an implementation manner of step S91 is further provided in the embodiments of the present application, as shown in fig. 13, which may include the following steps.

Step S131: and acquiring target computing power information and target network information.

In the embodiment of the application, the calculation force information can comprise information such as a service name of the calculation force service, the requirement of the calculation force request equipment on the calculation force and the like; the network information may be information such as a demand of the computing power requesting device for the network. The information of the demand of the computing force request device on computing force can be called computing force constraint information, and the computing force constraint information can comprise the information of the demands on performances such as CPU (Central Processing Unit ), GPU (Graphics Processing Unit, graphic processor), memory and storage; the service requirement information on the network may also be referred to as network constraint information, where the network constraint information may include requirement information on network performance such as delay, jitter, bandwidth, packet loss rate, and the like. In the power calculation network, the power calculation constraint information, the network constraint information and the path information all belong to the characteristics of the power calculation network.

The target calculation force information is calculation force constraint information required by the calculation force service requested by the calculation force request equipment at the time, and the target network information is network constraint information required by the calculation force service requested by the calculation force request equipment at the time. The user may input target computing power information and target network information to the computing power requesting device. The computing power request device acquires target computing power information and target network information input by a user.

Step S132: and filling the target calculation force information and the target network information in a second preset message header of the application layer to obtain a second message header.

In the embodiment of the application, a message header template processed by an application layer, such as a second preset message header, is preset in the calculation force request device. The power calculation request equipment fills the acquired target power calculation information and target network information in a second preset message header, and the second preset message header filled with the target power calculation information and the target network information is the second message header. In the embodiment of the present application, the second packet header may be referred to as a CNC header, where the application layer is located above the transport layer to which UDP belongs, and the second packet header may also be referred to as a CNCoU header.

In some embodiments, the second header may include a second message Type (Msg Type) field, a calculation region identification (Service Region ID) field, a Bitmap (Bitmap) field, a length of calculation service Name (Name Len) field, a calculation service Name (Computing Service Name) field, a Transaction identification (Transaction ID) field, a calculation and network constraint field. As shown in fig. 14. In fig. 14, numerals 0 to 9 of the first row denote bit positions. Wherein, the liquid crystal display device comprises a liquid crystal display device,

the second message type field is used to populate a type identifier of the power request message, where the value of the second message type field indicates that the message is a Computing Request (power request) message.

The power calculation region identification field is used to populate an identification of the requested power calculation region. One computing force area includes one or more pools of computing force resources. When the power computing resource pool is accessed to the power computing network, the power computing area accessed by the power computing resource pool can be designated, namely, the power computing area identifier corresponding to the power computing resource pool is designated. The value of the power calculation region identification field indicates the power calculation region where the request target power calculation resource pool is located, and in the embodiment of the present application, the power calculation region identification field may not limit the identification of the power calculation region, for example, when the value of the power calculation region identification field is 0, the identification of the power calculation region is indicated to be not limited, that is, the power calculation region where the request target power calculation resource pool is located is not limited.

The bitmap field is used for indicating whether the computing power and network constraint field carries computing power constraint information and network constraint information; the bitmap field may indicate by bit whether computational power and network constraint information is contained. As shown in fig. 14, the bitmap field has a length of 8 bits, and the 8 bits respectively indicate from top to bottom whether the eight constraint information of CPU, memory (Mem), storage (storage), GPU, delay (Delay), jitter (Jitter), bandwidth (Bandwidth), and packet Loss rate (Loss) are included. The constraint information is the requirement of the target network information and the target calculation force information on calculation force;

The length field of the name of the service of the computing power is used for filling the length of the name of the service of the computing power, namely the length of the name of the computing power used for filling the request;

the service name field of the computing power is used for filling the service name of the computing power, namely the service name of the computing power service used for filling the request;

the calculation power and network constraint fields are used to populate the target calculation power information and the target network information, as shown in fig. 14, CPU, memory, storage, GPU, latency, jitter, bandwidth, and packet loss rate.

In the embodiment of the present application, the number of fields included in the second header and the length of each field may be set according to actual requirements, and fig. 14 is only an example and is not meant to be limiting.

Step S133: and packaging a second message header for the original message to obtain the calculation force request message.

The computing force requesting device acquires an original message (original Packet) for requesting a computing force service. And after the second message header is obtained, the force calculation request equipment encapsulates the second message header for the original message to obtain a force calculation request message. The calculation request message may include a second message header and UDP header and IP header required for transmission. The structure of the power access messages for different types of IP addresses can be seen in fig. 11 and 12.

Step S134: and sending an algorithm force request message to the first algorithm force gateway so that the first algorithm force gateway extracts target algorithm force information and target network information from a second message header carried by the algorithm force request message, determines target equipment conforming to the target algorithm force information and the target network information, and determines target path information of a second algorithm force gateway connected from the first algorithm force gateway to the target equipment.

In the power computing network, one power computing gateway can record power computing information such as CPU, GPU and the like of a power computing resource pool connected with each power computing gateway in the power computing network in a power computing database of the power computing gateway. After the power calculation request equipment sends the power calculation request message to the first power calculation gateway, the first power calculation gateway extracts target power calculation information and target network information from a second message header carried by the power calculation request message, and determines a target power calculation resource pool conforming to the target power calculation information and the target network information by inquiring a power calculation database of the first power calculation gateway, so as to determine target equipment, and calculates target path information of the second power calculation gateway connected from the first power calculation gateway to the target equipment.

For example, after the first computing gateway receives the computing request message sent by the computing request device, the computing database of the first computing gateway queries the computing database according to the service name of the computing service of the current request, the requirement of the computing service of the current request for computing information, and the requirement of the computing service of the current request for network information carried in the computing request message, so as to calculate and obtain the IP address of the computing service meeting the constraint information condition, namely the IP address of the target device meeting the constraint condition, and the target path information of the second computing gateway connected from the first computing gateway to the target device, wherein the computing resource pool where the computing service meeting the constraint information condition is located is the target computing resource pool, namely the remote computing resource pool capable of providing the image rendering service.

Step S135: when a computing force response message sent by a first computing force gateway is received, acquiring target path information carried by the computing force response message from the computing force response message.

And after the first computing force gateway determines the target equipment and the target path information, returning the target path information to the computing force request equipment. Subsequently, the computing force requesting device may execute step S91 to step S94 according to the target path information.

In the embodiment of the present application, the first computing gateway may notify the computing request device of the target path information in any form, which is not limited.

In some embodiments, a header template, such as a third preset header, processed by the application layer is preset in the first computing gateway. The first computing gateway fills the acquired target path information into a third preset message header, and the third preset message header filled with the target path information is the third message header. In the embodiment of the present application, the third header may be referred to as a CNC header, where the application layer is located above the transport layer to which UDP belongs, and the third header may also be referred to as a CNCoU header.

In some embodiments, the third header includes a third message Type (Msg Type) field, a power area identification (Service Region ID) field, a Segment length (Segment Len) field, a power service address length (IP Len) field, a Transaction identification (Transaction ID) field, a power service address (Service IP adress) field, and a Duan Liebiao (Segment List) field. As shown in fig. 15. In fig. 15, numerals 0 to 9 of the first row denote bit positions. Wherein, the liquid crystal display device comprises a liquid crystal display device,

The third message type field is used for filling the type identifier of the calculation force response message. Here, the value of the third message type field indicates that the message is a Computing Request (arithmetic response) message.

The computing power area identification field is used for filling the identification of the computing power area of the computing power resource pool where the target device is located. When the power calculation region identification field of the power calculation request message does not specify the identification of the power calculation region, calculating the value of the power calculation region identification field in the power calculation response message by the first power calculation gateway to obtain the power calculation region of the target power calculation resource pool where the target equipment is located; when the power calculation region identification field of the power calculation request message designates the power calculation region identification, the value of the power calculation region identification field in the power calculation response message is the same as the power calculation region identification field of the power calculation request message designates the power calculation region identification, namely the power calculation region identification field of the power calculation request message designates the power calculation region identification which is the power calculation region identification of the target power calculation resource pool where the target equipment is located.

The segment length field is used to populate the length of the target path information. As shown in fig. 15, the segment length field occupies 8 bits in total. The Segment (Segment) may be an IPv4 address or an IPv6 address, and the highest bit indicates whether the address in the Segment list is an IPv4 address or an IPv6 address, and the remaining 7 bits indicate the number of elements in the Segment list.

The power service address length field is used to populate the length of the address of the target device. Wherein, the length word of the IPv4 address is 4 bytes, and the length of the IPv6 address is 16 bytes.

The transaction identification field is used to populate the transaction identification of the current computational force response. The transaction identification of the computational force response corresponds to the transaction identification of the computational force request.

The power service address field is used for filling the address of the target device, namely, the IP address corresponding to the power service meeting the constraint information condition, which is calculated by the first power gateway, and the IP address can be an IPv4 address or an IPv6 address.

The segment list field includes one or more elements for populating the target path information. The Segment List field refers to the field where the Segment List [0] -Segment List [ n ] is located, namely the field where the Segment List [0] -Segment List [ n ] is located, and the Segment List [0] -Segment List [ n ] respectively represents one element of the Segment List. The element types in the segment list may be in the form of IPv4 address + port (port) or IPv 6.

In the embodiment of the present application, the number of fields included in the third header and the length of each field may be set according to actual requirements, and fig. 15 is only an example and is not meant to be limiting.

In some embodiments, in order to facilitate distinguishing the power response message from other messages, so as to facilitate accurate access to the power service, the power request device and the first power network negotiate an application layer port, i.e. a preset port, corresponding to the power response message in advance. The power calculation request equipment receives a message, such as a first message, sent by a first power calculation gateway; identifying whether a target port of an application layer carried by the first message is a preset port; and when the destination port of the application layer carried by the first message is a preset port, determining the first message as a calculation response message. The subsequent calculation force request equipment can accurately access the calculation force service according to the target path information carried by the calculation force response message.

The destination port of the application layer may be obtained from a UDP header of the power calculation response message, or may be obtained from a designated location of the power calculation response message that is negotiated in advance between the power calculation request device and the first power calculation network, which is not limited.

In the technical scheme provided by the embodiment of the application, the power calculation request equipment sends the power calculation request message carrying the acquired target power calculation information and the target network information to the first power calculation gateway. After the first power gateway receives the power request message, determining target equipment and target path information according to target power information and target network information carried by the power request message, and sending a power response message carrying the determined target path information to the power request equipment. Therefore, the computing force request equipment can obtain the target path information from the first computing force gateway to the second computing force gateway by extracting the target path information in the computing force response message, so that the first computing force gateway can forward the computing force access message to the target equipment according to the target path information conveniently for subsequent realization. Meanwhile, the second preset message header is positioned in the application layer, so that the IPv6 address, the IPv4 address or other types of addresses can be identified at the same time, and the large-scale evolution of the power network can be greatly promoted. Because the second preset message header is positioned in the application layer, the operation of the application layer is user mode operation, the influence on an operating system is small, the updating of the computing power network characteristics is facilitated, and the updating speed of the computing power network characteristics is further improved.

Corresponding to the calculation power access method, the embodiment of the application also provides a calculation power access method. Referring to fig. 16, fig. 16 is a schematic flow chart of a second method for accessing a computing force according to an embodiment of the present application, where the method is applied to a target node in a computing force network, the target node is any node on a path from a first computing force gateway to a second computing force gateway in the computing force network, for example, the target node may be a first node on a path from the first computing force gateway to the second computing force gateway, i.e. a first computing force gateway, the target node may be a tail node on a path from the first computing force gateway to the second computing force gateway, i.e. a second computing force gateway, and the target node may be any intermediate node on a path from the first computing force gateway to the second computing force gateway. The first computing gateway is a computing gateway connected with computing request equipment, and the second computing gateway is a computing gateway connected with target equipment for providing target computing resources; the method comprises the following steps:

step S161: and receiving a calculation power access message.

And the target node receives the calculation power access message sent by the upstream equipment.

In the embodiment of the application, when the target node is the first node on the path from the first computing power gateway to the second computing power gateway, namely the first computing power gateway, the upstream equipment is computing power request equipment, and the first computing power gateway receives the computing power access message sent by the computing power request equipment. After the first computing force gateway receives the computing force access message sent by the computing force request device, the computing force access message can be processed and forwarded to the next hop node on the path from the first computing force gateway to the second computing force gateway.

When the target node is an intermediate node on the path from the first computing gateway to the second computing gateway, the upstream device is a last-hop node of the intermediate node in the path from the first computing gateway to the second computing gateway, and the intermediate node receives the computing power access message sent by the last-hop node. After the intermediate node receives the calculation power access message sent by the previous-hop node, the calculation power access message can be processed and forwarded to the next-hop node of the intermediate node on the path from the first calculation power gateway to the second calculation power gateway.

When the target node is a tail node on a path from the first computing power gateway to the second computing power gateway, namely the second computing power gateway, the upstream equipment is a last-hop node of the second computing power gateway in the path from the first computing power gateway to the second computing power gateway, and the second computing power gateway receives a computing power access message sent by the last-hop node. After the second computing force gateway receives the computing force access message sent by the previous hop node, the first message header of the computing force access message can be stripped to obtain an original message, the original message is forwarded to target equipment in a target computing force resource pool, and the target equipment in the target computing force resource pool provides computing force service for computing force request equipment.

In some embodiments, in order to facilitate distinguishing the power access message from other messages, so as to facilitate accurate access to the power service, each node and the power request device are preconfigured with an application layer port corresponding to the power access message, that is, a preset port. The first computing gateway receives a message, such as a second message, sent by computing force request equipment; identifying whether a target port of an application layer carried by the second message is a preset port; and when the destination port of the application layer carried by the second message is a preset port, determining the second message as a calculation access message. Each subsequent node can accurately access the computing power service according to the target path information carried by the computing power access message.

Step S162: extracting a first message header of an application layer carried by the computing power access message, wherein the first message header comprises target path information from a first computing power gateway to a second computing power gateway.

In the embodiment of the application, after the target node determines that the received message is the power access message, extracting a first message header of an application layer carried by the power access message, and analyzing target path information from a first power gateway to a second power gateway carried in the first message header. The structure of the first header is shown in fig. 10.

Step S163: and forwarding the computing power access message to the target equipment according to the target path information.

In the embodiment of the application, after the target node obtains the target path information carried in the computing power access message, the computing power access message can be forwarded to the target device according to the target path information, that is, the computing power access message is forwarded from the first computing power gateway to the second computing power gateway along the path indicated by the target path information. When the message is forwarded to the second power gateway, the second power gateway strips the first message header of the power access message to obtain an original message, and the power access message is restored to a standard UDP message; and forwarding the original message to target equipment in the target computing power resource pool, so that the target equipment in the target computing power resource pool provides computing power service for computing power request equipment.

In the technical scheme provided by the embodiment of the application, after receiving the power-calculation access message sent by the upstream equipment, the target node extracts the target path information in the first message header of the application layer carried by the power-calculation access message, and forwards the power-calculation access message to the target equipment according to the target path information so as to realize message forwarding. Because the first message header carrying the target path information in the power access message received by the target node is positioned in the application layer, the IPv4 address, the IPv6 address or other types of addresses can be identified at the same time, the problem that the power network cannot evolve on a large scale is avoided, and meanwhile, the process of forwarding the power access message to the target device according to the target path information is also performed in the application layer, and the problem that the characteristics of the power network cannot be updated rapidly is avoided.

In some embodiments, the present application further provides a method for accessing computing power, as shown in fig. 17, which may include the following steps:

step S171: and receiving a calculation power access message. The same as in step S161 described above.

Step S172: extracting a first message header of an application layer carried by the computing power access message, wherein the first message header comprises target path information from a first computing power gateway to a second computing power gateway.

When the target node is determined to be the first computing gateway or the intermediate node according to the target path information, step S173 is executed; when it is determined that the target node is the second computing gateway according to the target path information, step S174 is performed.

Step S173, the address of the next hop node is obtained from the target path information; updating the destination address of the calculation power access message into the acquired address to obtain an updated calculation power access message; and forwarding the updated calculation power access message.

Step S174, stripping the first message header to obtain the original message and forwarding the original message.

Taking the first message header format shown in fig. 10 as an example, after receiving the power access message, the target node analyzes the first message header, determines that the current target node is a first power gateway if the current segment index sl=first segment index is obtained, and determines that the current target node is a second power gateway if the current segment index SL is obtained to be equal to 0; otherwise, determining the current target node as an intermediate node.

When the target node is determined to be the first power gateway or the intermediate node, the target node may subtract 1 from the SL to obtain an updated SL, take a Segment pointed by the updated SL, that is, take Segment List [ SL ], update the destination address in the IP header of the power access message to Segment List [ SL ], and further forward the updated power access message.

When the target node is determined to be the second computing gateway, the target node pops up the first message header, restores the computing access message to a standard UDP message, and sends the standard UDP message to the UDP application layer for processing.

In the technical scheme provided by the embodiment of the application, a target node is determined to be a first computational power gateway or an intermediate node or a tail node according to target path information, and when the target node is the first computational power gateway or the intermediate node, the address of a next-hop node is obtained from the target path information; updating the destination address of the calculation power access message into the acquired address to obtain an updated calculation power access message; and forwarding the updated calculation power access message; and when the target node is the second computing gateway, the first message header is stripped to obtain an original message, and the original message is forwarded. And the forwarding mode of the calculation access message and the change condition of each address information in the forwarding process are defined when the current target node is the first calculation gateway, the intermediate node or the tail node.

In some embodiments, the first header may include a source port field, a source address field, and a source address length field; the source port field is used for filling a source port of a stream to which the computation access message belongs; the source address field is used for filling the source address of the stream to which the computational power access message belongs; the source address length field is used to populate the length of the source address indicated by the source address field.

When the target node is determined to be a first computing gateway according to the target path information, the first computing gateway can acquire a source address and a source port of a computing access message; and filling the acquired source address and source port into a source port field and a source address field of the first message header.

In this case, the step of peeling the first header to obtain the original message includes: copying the source address of the source address field of the first message header to the source address field of the IP header of the power access message, copying the source port of the source port field of the first message header to the source port field of the UDP header of the power access message, and stripping the first message header to obtain the original message.

In some embodiments, when the target node is determined to be the first computing gateway according to the target path information, the computing method may further include: receiving an calculation force request message sent by calculation force request equipment; extracting target calculation power information and target network information from a second message header carried by the calculation power request message; determining target equipment which accords with the target computing power information and the target network information; determining target path information of a second computing gateway connected from the first computing gateway to target equipment; filling the target path information in a third preset message header of the application layer to obtain a third message header; encapsulating a third message header for the original message to obtain a calculation force response message; and sending an calculation response message to the calculation request terminal so that the calculation request terminal can acquire the target path information. The structure of the second header and the third header can be seen from the related description of fig. 14 and fig. 15.

Corresponding to the above calculation force access method, the embodiment of the application also provides a calculation force access device, as shown in fig. 18, applied to a calculation force request device, including:

the obtaining module 181 is configured to obtain target path information from a first computing gateway to a second computing gateway, where the first computing gateway is a computing gateway connected to the computing request device, and the second computing gateway is a computing gateway connected to a target device that provides a target computing resource;

a filling module 182, configured to fill the target path information in a first preset header of an application layer, to obtain a first header;

the encapsulation module 183 is configured to encapsulate the first header for the original packet, to obtain a calculation access packet;

and a sending module 184, configured to send the power access packet to the first power gateway, so that the first power gateway forwards the power access packet to the target device according to the target path information.

In some embodiments, the obtaining module 181 is specifically configured to:

acquiring target computing power information and target network information;

In some embodiments, the computing power access device may further include:

Corresponding to the foregoing method for accessing the computing power, the embodiment of the present application further provides a computing power access device, as shown in fig. 19, which is applied to a target node in a computing power network, where the target node is any node on a path from a first computing power gateway to a second computing power gateway in the computing power network, where the first computing power gateway is a computing power gateway connected to a computing power request device, and the second computing power gateway is a computing power gateway connected to a target device that provides a target computing power resource, and the computing power access device includes:

a first receiving module 191, configured to receive a computing power access message;

an extracting module 192, configured to extract a first header of an application layer carried by the computing power access packet, where the first header includes target path information from the first computing power gateway to the second computing power gateway;

and a forwarding module 193, configured to forward the computing power access packet to the target device according to the target path information.

In some embodiments, the computing power access device may further include:

In some embodiments, forwarding module 193 is specifically configured to:

A forwarding module 193, specifically for: copying the source address of the source address field of the first message header to the source address field of the IP header of the power-calculation access message, copying the source port of the source port field of the first message header to the source port field of the UDP header of the power-calculation access message, and stripping the first message header to obtain an original message.

In some embodiments, the computing power access device may further include:

In correspondence with the above-mentioned computing power access method, the embodiment of the present application further provides an electronic device, such as the computing power request device and the target node, as shown in fig. 20, including a processor 201 and a machine-readable storage medium 202, where the machine-readable storage medium 202 stores machine-executable instructions that can be executed by the processor 201, and the processor 201 is caused by the machine-executable instructions to: any computational power access method applied to the computational power request equipment or any computational power access method applied to the target node is realized.

The machine-readable storage medium may include random access Memory (Random Access Memory, RAM) or may include Non-Volatile Memory (NVM), such as at least one disk Memory. In the alternative, the machine-readable storage medium may also be at least one memory device located remotely from the foregoing processor.

The processor may be a general-purpose processor including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but also digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

In yet another embodiment of the present application, a computer readable storage medium is also provided, in which a computer program is stored, which when executed by a processor, implements any one of the computing power access methods applied to the computing power requesting device, or implements any one of the computing power access methods applied to the target node.

In yet another embodiment of the present application, a computer program product comprising instructions that, when run on a computer, cause the computer to perform, or implement, any of the computational power access methods applied to a computational power requesting device is also provided.

In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present application, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in or transmitted from one computer-readable storage medium to another, for example, by wired (e.g., coaxial cable, optical fiber, digital Subscriber Line (DSL)), or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid State Disk (SSD)), etc.

It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.

In this specification, each embodiment is described in a related manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus, electronic devices, storage media, and program product embodiments, the description is relatively simple as it is substantially similar to method embodiments, as relevant points are found in the partial description of method embodiments.

The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application are included in the protection scope of the present application.

Claims

1. A computing power access method, characterized by being applied to a computing power requesting device, the method comprising:

2. The method of claim 1, wherein the step of obtaining target path information for the first computing force gateway to the second computing force gateway comprises:

Acquiring target computing power information and target network information;

3. The method of claim 2, wherein the second header includes a second message type field, a power calculation region identification field, a bitmap field, a length of power calculation service name field, a transaction identification field, a power calculation and network constraint field;

4. The method of claim 2, wherein the power response message comprises a third header comprising a third message type field, a power region identification field, a segment length field, a power service address length field, a transaction identification field, a power service address field, and a Duan Liebiao field;

5. The method according to claim 2, wherein the method further comprises:

receiving a first message sent by the first computing gateway;

6. The method of any of claims 1-5, wherein the first header includes a first message type field, a flow identification length field, a flow identification field, a segment type field, a segment routing header length field, a segment list field, a current segment index field, and a first segment index field;

7. The method of claim 6, wherein the first header further comprises: a source port field, a source address field, and a source address length field;

8. The power computing access method is characterized by being applied to a target node in a power computing network, wherein the target node is any node on a path from a first power computing gateway to a second power computing gateway in the power computing network, the first power computing gateway is a power computing gateway connected with power computing request equipment, and the second power computing gateway is a power computing gateway connected with target equipment for providing target power computing resources; the method comprises the following steps:

Receiving a calculation power access message;

9. The method of claim 8, wherein the method further comprises:

receiving a second message sent by upstream equipment;

10. The method of claim 8, wherein forwarding the power access message to the target device according to the target path information comprises:

11. The method of claim 10, wherein the first header further comprises a source port field, a source address field, and a source address length field; the source port field is used for filling a source port of a flow to which the computing power access message belongs; the source address field is used for filling the source address of the flow to which the computing power access message belongs; the source address length field is used for filling the length of the source address indicated by the source address field;

the method further comprises the steps of:

12. The method of claim 8, wherein when the target node is a first computing gateway, the method further comprises:

extracting target calculation force information and target network information from a second message header carried by the calculation force request message;

13. The method of claim 12, wherein the second header includes a second message type field, a power calculation region identification field, a bitmap field, a length of power calculation service name field, a transaction identification field, a power calculation and network constraint field;

14. The method of claim 12, wherein the third header includes a third message type field, a power region identification field, a segment length field, a power service address length field, a transaction identification field, a power service address field, and a Duan Liebiao field;

15. The method according to any of claims 8-14, wherein the first message header comprises a first message type field, a source address length field, a flow identification length field, a source port field, a source address field, a flow identification field, a segment type field, a segment routing header length field, a segment list field, a current segment index field, and a first segment index field;

16. A computing power access apparatus for use with a computing power requesting device, the apparatus comprising:

17. The power computing access device is characterized by being applied to a target node in a power computing network, wherein the target node is any node on a path from a first power computing gateway to a second power computing gateway in the power computing network, the first power computing gateway is a power computing gateway connected with power computing request equipment, and the second power computing gateway is a power computing gateway connected with target equipment for providing target power computing resources; the device comprises:

18. An electronic device comprising a processor and a machine-readable storage medium storing machine-executable instructions executable by the processor, the processor being caused by the machine-executable instructions to: a method according to any one of claims 1 to 7 or a method according to any one of claims 8 to 15.

19. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when executed by a processor, implements the method of any of claims 1-7 or implements the method of any of claims 8-15.