CN110636007A - Routing arrangement method, routing arrangementr and system for edge computing network - Google Patents

Abstract

The invention discloses a routing arrangement method, a routing arrangement device and a system for an edge computing network. The method includes: receiving a service connection opening request reported by an edge computing node; querying whether there is an existing connection with the source node, The route corresponding to the destination node; if there is a corresponding route in the working routing table, the service connection is performed according to the route and the stored bandwidth information; if the corresponding route does not exist in the working routing table, check whether it exists in the pre-routing table and judge Whether the routing information meets the bandwidth requirement of the service request; if there is a corresponding route in the pre-routing table, and the initial bandwidth information meets the bandwidth requirement of the service request, and the information in the pre-routing table is updated, the corresponding route and service request bandwidth to send. The routing arrangement method, routing arrangement device and system of the edge computing network provided by the present invention can realize rapid establishment of edge computing services and efficient utilization of optical network resources.

Description

Routing arrangement method, routing arrangementr and system for edge computing network

technical field

The present invention relates to an edge computing network, in particular to a routing arrangement method, a routing arrangement device and a system for an edge computing network.

Background technique

Edge computing refers to an open platform that integrates network, computing, storage, and application core capabilities on the side close to the source of objects or data, and provides the nearest end services. The core is to provide computing power at the edge of the network, generate faster network service response, and meet the basic needs of the industry in real-time business, application intelligence, security and privacy protection. Edge computing is between physical entities and industrial connections, or on top of physical entities. With cloud computing, historical data from edge computing can still be accessed.

Edge computing nodes are usually located at the edge of the access network or metropolitan area network, and are interconnected with the cloud data center through the optical transport network. Due to the limited processing capacity, a large number of distributed edge nodes cannot provide all services completely independently. They need to cooperate with other edge or cloud data centers through the optical transport network. The optical network connecting edge nodes and cloud data centers needs to be fine-grained in bandwidth , The edge computing service with small transmission delay provides a large number of routes to support the collaborative interaction requirements between distributed edge nodes and cloud centers.

Edge computing is a multi-application and multi-service system, and each service has different performance requirements for the allocated optical network routing according to its service characteristics.

Based on this, the inventors of the present application found that the low-latency and high-efficiency support of optical networks for edge computing services has become an urgent problem to be solved, especially the interaction delay and resource utilization optimization issues between edge computing services and optical network routing.

The information disclosed in this Background section is only intended to increase the understanding of the general background of the invention and should not be taken as an acknowledgment or any form of suggestion that the information constitutes the prior art that is already known to those of ordinary skill in the art.

Contents of the invention

The object of the present invention is to provide a routing arrangement method, a routing arrangement device and a system for an edge computing network, which can improve the utilization rate of edge computing services and optical network routing resources.

To achieve the above object, the present invention provides a routing arrangement method for an edge computing network. The edge computing network includes a plurality of edge computing nodes. The routing arrangement method includes: receiving a service connection activation request reported by the edge computing node, wherein the service The connection opening request includes the source node of the service connection, the destination node and the bandwidth of the service request; whether there is an existing route corresponding to the source node and the destination node in the working routing table, wherein the working routing table stores the established A set of routes for service connections and bandwidth information corresponding to the routes; if there are existing routes corresponding to the source node and the destination node in the working routing table, the service connection is performed according to the routes and the stored bandwidth information; if the working There is no existing route corresponding to the source node and the destination node in the routing table, then query whether there is a route corresponding to the source node and the destination node in the pre-routing table, and judge whether the route meets the bandwidth of the service request Requirements, wherein the pre-routing table stores the pre-established route and the initial bandwidth information corresponding to the route; if there is a route corresponding to the source node and the destination node in the pre-routing table, and the initial bandwidth information meets the bandwidth requirement of the service request , then replace the initial bandwidth information corresponding to the route with the bandwidth of the service request, update the information in the pre-routing table, and migrate the routes corresponding to the source node and destination node from the pre-routing table to the working routing table In: sending the route corresponding to the source node and the destination node and the bandwidth of the service request to the OpenFlow controller.

In a preferred embodiment, the target node is: a cloud computing center node, and after receiving the service connection opening request reported by the edge computing node, the query whether there is an existing connection with the source node in the working routing table 1. Before the route corresponding to the destination node, it also includes: calculating a route to the cloud computing center for each edge computing node, and generating a pre-routing table.

In a preferred embodiment, calculating a route to the cloud computing center for each edge computing node, generating a pre-routing table includes: setting the initial bandwidth information and delay requirements required by all edge computing nodes; The requested bandwidth, the total bandwidth of link i, j and the total number of routes from the source node to the destination node in the network, calculate the sharing coefficient of link i, j, wherein, the link i, j is the edge computing node i and the edge Calculate the links between nodes j; circularly count the sharing coefficients of all links, and establish a network resource parameter table, wherein the network resource parameter table stores edge computing nodes, initial bandwidth information corresponding to the edge computing nodes, and Time delay requirements, and the sharing coefficient of the link corresponding to the edge computing node; in all links, according to the sharing coefficient, use the D algorithm to calculate the remaining bandwidth and the route that meets the delay requirements, and store it in the pre-routing table, wherein, A route consists of multiple links from a source node to a destination node.

In a preferred embodiment, updating the information in the pre-routing table includes: recalculating the number of routes passing through link i, j; if the number of routes passing through link i, j is more increase, update the sharing coefficients of links i and j, and store the updated sharing coefficients and remaining bandwidth of links i and j in the network resource parameter table; repeat the above steps until all links are updated; In the route, use the D algorithm to calculate the route with the remaining bandwidth and delay meeting the requirements according to the sharing coefficient, and update the pre-routing table.

In a preferred embodiment, the update of the sharing coefficient of link i, j includes: recalculating the link i, j according to the recalculated number of routes passing through link i, j and the total number of routes from the source node to the destination node in the network. The sharing coefficient of i, j; the recalculated sharing coefficient of link i, j is multiplied by the coefficient (1+1/N), as the updated sharing coefficient of link i, j.

In a preferred embodiment, the calculation of the sharing coefficient of link i, j according to the number of routes passing through link i, j and the total number of routes from the source node to the destination node in the network includes: calculating link i according to formula one , the sharing coefficient θ(ij) of j, the formula one is:

Among them, Link(i,j) is the link between edge computing node i and edge computing node j, B _i,j is the total bandwidth of link Link(i,j), and Rl is the link passed by route l A collection of , Bl is the bandwidth occupied by route l, and L is the total number of routes in the network.

In order to achieve the above object, the present invention also provides a routing arranger for the edge computing network, which is used to execute the above-mentioned routing arrangement method for the edge computing network.

In order to achieve the above object, the present invention also provides a routing arrangement system for an edge computing network, including the above-mentioned routing arrangementr, and also includes: an OpenFlow controller, which is used to transmit the route from the source node to the destination node according to the routing arrangementr. The information and the bandwidth requested by the service generate an OpenFlow flow table.

In a preferred embodiment, it also includes: an optical switch node; the OpenFlow controller is also used to send the OpenFlow flow table to a corresponding optical switch node; the optical switch node is used to configure the wavelength resource according to the flow table, and send the OpenFlow The controller replies with a confirmation message; the edge computing node performs business interaction and data transmission with the cloud computing center node according to the configured wavelength resources.

In a preferred embodiment, after the edge computing node completes the service interaction and data transmission, it sends a service connection end request to the OpenFlow controller; the OpenFlow controller forwards the request to the routing orchestrator; Migrate from the working routing table to the pre-routing table, and update the routing bandwidth parameter to the initial bandwidth information; the OpenFlow controller sends a flow table command to the optical switch node to release the optical link resources.

Compared with the prior art, according to the method of combining the pre-routing table and the working routing table of the present invention, routing calculation and resource dynamic allocation are performed between all edge computing nodes and cloud computing nodes from the perspective of the global network, which can realize The routing control of the entire network and the dynamic planning and policy optimization of resource allocation further realize the rapid establishment of edge computing services and the efficient utilization of optical network resources.

Description of drawings

Fig. 1 is a flow chart of a routing arrangement method for an edge computing network according to an embodiment of the present invention.

Fig. 2 is a schematic structural diagram of a router orchestrator of an edge computing network according to an embodiment of the present invention.

Detailed ways

Below in conjunction with accompanying drawing, specific embodiment of the present invention is described in detail, but it should be understood that protection scope of the present invention is not limited by specific embodiment.

Unless expressly stated otherwise, throughout the specification and claims, the term "comprise" or variations thereof such as "includes" or "includes" and the like will be understood to include the stated elements or constituents, and not Other elements or other components are not excluded.

Example 1

As shown in Figure 1, it is a flow chart of a routing arrangement method for an edge computing network according to a preferred embodiment of the present invention. The edge computing network includes a plurality of edge computing nodes, and the routing arrangement method includes: steps S1-S6.

In step S1, the service connection opening request reported by the edge computing node is received, wherein the service connection opening request includes the source node of the service connection, the destination node and the bandwidth of the service request; wherein, the target node is: a cloud computing center node.

In step S2, query whether there is an existing route corresponding to the source node and the destination node in the working routing table, wherein the working routing table stores a set of routes for which business connections have been established and bandwidth information corresponding to the routes , the route is path information from the source node to the destination node.

Specifically, a link refers to the optical fiber directly connected between two adjacent nodes, and a route is an end-to-end connection formed by a series of links; the number of routes is how many routes pass through the link.

In step S3, if there is an existing route corresponding to the source node and the destination node in the working routing table, the service connection is performed according to the route and the stored bandwidth information.

In step S4, if there is no existing route in the working routing table, then query whether there is a route corresponding to the source node and the destination node in the pre-routing table, and judge whether the route meets the bandwidth requirement of the service request, wherein The pre-routing table stores pre-established routes and initial bandwidth information corresponding to the routes.

In step S5, if there is a route corresponding to the source node and the destination node in the pre-routing table, and the initial bandwidth information meets the bandwidth requirement of the service request, then replace the initial bandwidth information corresponding to the route with the bandwidth of the service request , and update the information in the pre-routing table, and migrate the routes corresponding to the source node and the destination node from the pre-routing table to the working routing table.

In step S6, the route corresponding to the source node and the destination node and the bandwidth of the service request are sent to the OpenFlow controller.

Therefore, this embodiment uses the method of combining "pre-routing table" and "working routing table" to perform routing calculation and resource dynamic allocation among all "edge computing nodes-cloud computing nodes" from the perspective of the global network, which can Realize network-wide routing control and dynamic planning and policy optimization of resource allocation, and further realize the rapid establishment of edge computing services and efficient utilization of optical network resources.

In an implementation manner, after step S1 and before S2, the method may further include: step S0.

In step S0, a route to the cloud computing center is calculated for each edge computing node, and a pre-routing table is generated.

Specifically, steps S01-S04 may be included.

In step S01, the initial bandwidth information and delay requirements required by all edge computing nodes are set.

In step S02, according to the bandwidth of the service request, the total bandwidth of the link i, j and the total number of routes from the source node to the destination node in the network, the sharing coefficient of the link i, j is calculated, wherein the link i, j is the link between edge computing node i and edge computing node j.

Specifically, the sharing coefficient θ(ij) of link i, j is calculated according to Formula 1, which is:

Among them, Link(i,j) is the link between edge computing node i and edge computing node j, B _i,j is the total bandwidth of link Link(i,j), and Rl is the link passed by route l A collection of , Bl is the bandwidth occupied by route l, and L is the total number of routes in the network.

In step S03, the sharing coefficients of all links are cyclically counted, and a network resource parameter table is established, wherein the network resource parameter table stores edge computing nodes, initial bandwidth information and delay requirements corresponding to the edge computing nodes, A sharing coefficient of a link corresponding to the edge computing node.

In step S04, in all links, according to the sharing coefficient, use the D algorithm to calculate the route with the remaining bandwidth and delay meeting the requirements, and store it in the pre-routing table, where the route includes multiple links from the source node to the destination node , the remaining bandwidth of the link is the difference between the total bandwidth of the link and the bandwidth of the service request corresponding to the bandwidth of the link.

Specifically, the D algorithm is the Dijkstra algorithm.

Therefore, by pre-calculating the overall route, and then dynamically adjusting the network bandwidth allocation method according to the actual needs of the business, the fast connection establishment of the edge computing business can be realized.

Updating the information in the pre-routing table in step S5 includes: step S51-step S55.

In step S51, the number of routes passing through link i, j is calculated again.

In step S52, if the number of routes passing through link i, j has increased compared with the previous calculation, update the sharing coefficient of link i, j, and update the sharing coefficient of link i, j and the remaining bandwidth Stored in the network resource parameter table.

Specifically, updating the sharing coefficient of link i, j includes: recalculating the sharing coefficient of link i, j according to the recalculated number of routes passing through link i, j and the total number of routes from the source node to the destination node in the network; The recalculated sharing coefficient of link i, j is multiplied by the coefficient (1+1/N), as the updated sharing coefficient of link i, j.

In step S53, if the number of routes passing through link i, j has not increased compared with the previous calculation, the sharing coefficient of link i, j will be calculated again as the updated sharing coefficient of link i, j.

Step S54, repeating the above steps until the updating of all links is completed;

Step S55, in all links after updating, use the D algorithm to calculate the route whose remaining bandwidth and time delay meet the requirements according to the sharing coefficient, and update the pre-routing table.

Therefore, if the recalculated number of routes passing through link i, j is greater than the previous number of routes, the recalculated sharing coefficient of link i, j is multiplied by the coefficient (1+1/N) as the updated link The sharing coefficient of route i, the link weight with a high sharing coefficient gets a reward value, and the link weight with a low link sharing degree gets a penalty value, and the link with a high sharing coefficient is preferentially selected to share network resources as much as possible and improve resource utilization. utilization rate.

Example 2

As shown in FIG. 2 , it is a schematic structural diagram of a routing orchestrator of an edge computing network according to a preferred embodiment of the present invention. This embodiment also provides a routing orchestrator of an edge computing network, which is used to implement the edge The routing arrangement method of the computing network will not be described in detail in this embodiment.

The routing orchestrator mainly includes: routing calculation module 1, pre-routing module 2, working routing module 3 and northbound interface module 4. A pre-routing table is stored in the pre-routing module 2, and a working routing table is stored in the working routing module 3.

The routing calculation module 1 is used to receive a request for opening a service connection reported by an edge computing node, wherein the request for opening a service connection includes a source node, a destination node of the service connection, and the bandwidth of the service request; wherein the target node is: a cloud computing center node .

Routing calculation module 1 inquires whether there is an existing route corresponding to the source node and the destination node in the working routing table in the working routing module 3, wherein the working routing table stores the set of routes that have established business connections and the routing information associated with the routing table. The bandwidth information corresponding to the route is the path information from the source node to the destination node. If there is an existing route corresponding to the source node and the destination node in the working routing table, then the service connection is carried out according to the route and the stored bandwidth information. If there is no existing route corresponding to the source node and the destination node in the work routing table, whether there is a route corresponding to the source node and the destination node in the pre-routing table of the query pre-routing module 2, and judge the route Whether the route meets the bandwidth requirement of the service request, wherein the pre-routing table stores the pre-established route and the initial bandwidth information corresponding to the route. If there is a route corresponding to the source node and the destination node in the pre-routing table, and the initial bandwidth information meets the bandwidth requirements of the service request, then the initial bandwidth information corresponding to the route is replaced with the bandwidth of the service request, and the pre-routing The information in the table is updated, and the routes corresponding to the source node and the destination node are migrated from the pre-routing table to the working routing table.

The route calculation module sends the route corresponding to the source node and the destination node and the bandwidth of the service request to the OpenFlow controller through the northbound interface module.

The route arranger can also include a network topology auxiliary diagram module 5, and the network topology auxiliary diagram module 5 is used for storing a network resource parameter table.

Therefore, this embodiment uses the method of combining "pre-routing table" and "working routing table" to perform routing calculation and resource dynamic allocation among all "edge computing nodes-cloud computing nodes" from the perspective of the global network, which can Realize network-wide routing control and dynamic planning and policy optimization of resource allocation, and further realize the rapid establishment of edge computing services and efficient utilization of optical network resources.

Example 3

This embodiment also provides a routing arrangement system for an edge computing network, including the routing editor in Embodiment 2, and also includes: an OpenFlow controller and an optical switch node.

The OpenFlow controller is used to generate the OpenFlow flow table according to the routing information from the source node to the destination node sent by the routing orchestrator and the bandwidth of the service request. The OpenFlow controller is also used to send the OpenFlow flow table to the corresponding optical switch node.

The optical switch node is used to configure the wavelength resource according to the flow table, and reply the confirmation message to the OpenFlow controller.

The edge computing node performs business interaction and data transmission with the cloud computing center node according to the configured wavelength resources. After the edge computing node completes business interaction and data transmission, it sends a business connection end request to the OpenFlow controller;

The OpenFlow controller forwards the request to the routing orchestrator; the routing orchestrator migrates the route corresponding to the service connection from the working routing table to the pre-routing table, and updates the route bandwidth parameter to the initial bandwidth information; the OpenFlow controller sends the optical switch node Send flow table commands to release optical link resources.

Therefore, this embodiment implements network-wide routing control and resource allocation dynamic planning and strategy optimization through the network-wide "routing orchestration" mechanism, so as to realize the rapid establishment of edge computing services and efficient utilization of optical network resources.

Those skilled in the art should understand that the embodiments of the present application may be provided as methods, systems, or computer program products. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

The present application is described with reference to flowcharts and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and/or block in the flowchart and/or block diagram, and a combination of the flow and/or block in the flowchart and/or block diagram can be realized by computer program instructions. These computer program instructions may be provided to a general purpose computer, special purpose computer, embedded processor, or processor of other programmable data processing equipment to produce a machine such that the instructions executed by the processor of the computer or other programmable data processing equipment produce a An apparatus for realizing the functions specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing apparatus to operate in a specific manner, such that the instructions stored in the computer-readable memory produce an article of manufacture comprising instruction means, the instructions The device realizes the function specified in one or more procedures of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions can also be loaded onto a computer or other programmable data processing device, causing a series of operational steps to be performed on the computer or other programmable device to produce a computer-implemented process, whereby the The instructions provide steps for implementing the functions specified in the flow chart or blocks of the flowchart and/or the block or blocks of the block diagrams.

The foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. These descriptions are not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling others skilled in the art to make and use various exemplary embodiments of the invention, as well as various Choose and change. It is intended that the scope of the invention be defined by the claims and their equivalents.

Claims (10)

1. A routing arrangement method of an edge computing network, the edge computing network comprising a plurality of edge computing nodes, the routing arrangement method comprising:

receiving a service connection opening request reported by an edge computing node, wherein the service connection opening request comprises a source node and a destination node of service connection and the bandwidth of the service request;

inquiring whether existing routes corresponding to the source node and the destination node exist in a working routing table, wherein the working routing table stores a set of routes with established service connection and bandwidth information corresponding to the routes;

if existing routes corresponding to the source node and the destination node exist in the working routing table, performing service connection according to the routes and the stored bandwidth information;

if the working routing table does not have existing routes corresponding to the source node and the destination node, inquiring whether the routes corresponding to the source node and the destination node exist in a pre-routing table or not, and judging whether the routes meet the bandwidth requirement of a service request or not, wherein the pre-routing table stores the pre-established routes and initial bandwidth information corresponding to the routes;

if the pre-routing table has routes corresponding to the source node and the destination node and the initial bandwidth information meets the bandwidth requirement of the service request, replacing the initial bandwidth information corresponding to the route with the bandwidth of the service request, updating the information in the pre-routing table, and migrating the routes corresponding to the source node and the destination node from the pre-routing table to the working routing table;

and sending the routes corresponding to the source node and the destination node and the bandwidth of the service request to an OpenFlow controller.

2. The routing orchestration method according to claim 1, wherein the target node is: the cloud computing center node, after receiving the service connection fulfillment request reported by the edge computing node, and before querying whether existing routes corresponding to the source node and the destination node exist in the working routing table, further includes:

and calculating a route to the cloud computing center for each edge computing node, and generating a pre-routing table.

3. The routing orchestration method according to claim 2, wherein the computing a route to the cloud computing center for each edge computing node, and wherein generating the pre-routing table comprises:

setting initial bandwidth information and time delay requirements needed by all edge computing nodes;

calculating a sharing coefficient of a link i, j according to the bandwidth of the service request, the total bandwidth of the link i, j and the total number of routes from a source node to a destination node in the network, wherein the link i, j is a link between an edge calculation node i and the edge calculation node j;

circularly counting the sharing coefficients of all links, and establishing a network resource parameter table, wherein the network resource parameter table stores edge computing nodes, initial bandwidth information and time delay requirements corresponding to the edge computing nodes, and the sharing coefficients of the links corresponding to the edge computing nodes;

and in all links, calculating routes with the residual bandwidth and the time delay meeting the requirements by adopting a D algorithm according to the sharing coefficient, and storing the routes into a pre-routing table, wherein the routes comprise a plurality of links from the source node to the destination node.

4. The routing orchestration method according to claim 3, wherein the updating information in the pre-routing table comprises:

calculating the number of routes passing through the link i, j again;

if the number of routes passing through the link i, j is increased compared with the last calculation, updating the sharing coefficient of the link i, j, and storing the updated sharing coefficient of the link i, j and the residual bandwidth in a network resource parameter table;

repeating the steps until the updating of all links is completed;

and in all links, calculating the routes with the residual bandwidth and the time delay meeting the requirements by adopting a D algorithm according to the sharing coefficients, and updating the pre-routing table.

5. The routing orchestration method according to claim 4, wherein the updating the sharing coefficients of links i, j comprises:

according to the recalculated number of the routes passing through the links i and j and the total number of the routes from the source node to the destination node in the network, recalculating the sharing coefficient of the links i and j;

and multiplying the recalculated sharing coefficient of the link i, j by a coefficient (1+1/N) to obtain the updated sharing coefficient of the link i, j.

6. The routing arrangement method of claim 3, wherein calculating the sharing coefficient of the link i, j according to the number of routes passing through the link i, j and the total number of routes from the source node to the destination node in the network comprises:

calculating the sharing coefficient theta (ij) of the link i, j according to a formula I:

wherein Link (i, j) is a Link between an edge computing node i and an edge computing node j, B_i,jThe total bandwidth of the Link (i, j), Rl is the set of links passed by the route L, Bl is the bandwidth occupied by the route L, and L is the total number of routes in the network.

7. A route orchestrator of an edge computing network, for performing the method of route orchestration of the edge computing network according to any one of claims 1-6.

8. A routing orchestration system for an edge computing network, comprising the routing orchestrator of claim 7, further comprising:

and the controller of the OpenFlow is used for generating an OpenFlow flow table according to the routing information sent by the routing composer from the source node to the destination node and the bandwidth of the service request.

9. The routing orchestration system of claim 8, further comprising: an optical switch node;

the OpenFlow controller is further configured to send an OpenFlow flow table to a corresponding optical switch node;

the optical switch node is used for configuring wavelength resources according to the flow table and replying a confirmation message to the OpenFlow controller;

and the edge computing node performs service interaction and data transmission with the cloud computing center node according to the configured wavelength resources.

10. The routing orchestration system of claim 9,

after finishing service interaction and data transmission, the edge computing node sends a service connection ending request to the OpenFlow controller;

the OpenFlow controller forwards the request to the routing orchestrator;

the route orchestrator transfers the route corresponding to the service connection from the working route table to the pre-route table, and updates the route bandwidth parameter to the initial bandwidth information;

and the OpenFlow controller sends a flow table instruction to the optical switch node and releases optical link resources.

Images