Delay-guaranteed cloud edge collaborative optimization balanced deployment method

Through joint optimization of computing-network-assembly, comprehensive consideration of resource and delay requirements, and optimized cloud-edge collaborative deployment, the problems of uneven resource utilization and insufficient time delay guarantee in traditional deployment strategies are solved, and efficient resource utilization and user satisfaction are achieved Improved degree.

CN118939412AActive Publication Date: 2024-11-12中国司法大数据研究院有限公司
View PDF 6 Cites 0 Cited by

Patent Information

Application Number
CN202410928317.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-11
Publication Date
2024-11-12
Estimated Expiration
2044-07-11

AI Technical Summary

Technical Problem

The traditional cloud-edge collaborative deployment strategy fails to effectively comprehensively consider the user distribution and application delay requirements, resulting in uneven resource utilization and the inability to deploy important applications, affecting service quality.

Method used

Adopting a method based on computing-network-assembly joint optimization, comprehensively considering basic resource capacity and cluster resource load balancing, combining the application's delay requirements and importance, optimized and deployed through the delay matrix and weighted computing power value to ensure rapid application Efficient use of response and resources.

Benefits of technology

It realizes the guarantee of time delay while ensuring balanced business deployment, improves user experience and resource utilization, avoids the problem of uneven resource share of computing nodes, and improves user satisfaction.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118939412A_ABST
    Figure CN118939412A_ABST
Patent Text Reader

Abstract

The invention discloses a delay-guaranteed cloud edge collaborative optimization balanced deployment method, which is based on calculation-network-storage joint optimization, comprehensively considers basic resource capacity and cluster resource load balancing, combines time delay requirements and importance of applications, optimizes and deploys the applications, and realizes quick response of the applications and efficient utilization of resources. Wherein the business applications are deployed from three aspects of construction attributes, business attributes and status attributes of the applications, after comprehensive consideration, sorting is performed according to business volumes after importance weights are introduced, and optimization deployment is performed according to differential guarantee requirements of the business applications; when cloud edge coordination deployment is carried out, in the thought of balancing the resource utilization rate, the time delay of the application used by the user in the cloud edge computing environment is considered, and efficient utilization of resources is achieved while the normal use experience of the user is guaranteed. According to the method, resource limitation and real-time requirements in edge calculation can be effectively met.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of cloud computing and edge computing, and relates to a cloud-edge collaborative optimization and balanced deployment method with guaranteed delay. Background Art

[0002] Cloud-edge collaboration is a combination of cloud computing and edge computing, which gives full play to the advantages of both and makes up for their respective shortcomings. In the cloud-edge collaboration architecture, cloud center nodes are generally considered to have strong computing capabilities, deploy applications with high computing resource requirements and insensitive to latency, and are responsible for processing large-scale data analysis and complex computing tasks; while edge nodes have limited computing resources, generally deploy applications with low computing resource requirements and sensitive to latency, and are responsible for processing computing tasks with high real-time requirements and small data volumes. The two realize data sharing and complementary computing capabilities through efficient network connections and collaboration mechanisms. Through cloud-edge collaboration, resource allocation is optimized, IT operating costs are reduced, and the overall benefits of information construction are improved.

[0003] How to choose the right computing nodes to deploy business applications will directly affect the performance, response speed and service quality of the application. Traditional application deployment strategies consider factors such as node resource capacity and cluster load balancing, but do not comprehensively consider user distribution and application latency requirements, which reduces the service quality of the application; do not carefully consider the adaptability of basic resources, resulting in uneven utilization of cluster resources; and do not consider the differentiated security requirements of applications when deploying multiple applications, which may result in the inability to deploy important applications. Summary of the invention

[0004] In response to the problems existing in the prior art, the purpose of the present invention is to provide a cloud-edge collaborative optimization and balanced deployment method with delay guarantee, which takes into account delay guarantee while ensuring balanced deployment of business applications, so as to meet the needs of business operation and achieve the best user experience.

[0005] The present invention is based on the joint optimization of computing, network and storage, comprehensively considers the basic resource capacity (such as CPU, memory and disk storage, etc.) and cluster resource load balancing, combines the latency requirements and importance of the application, optimizes the deployment of the application, and realizes rapid response of the application and efficient use of resources.

[0006] When deploying business applications, the present invention does not simply deploy them in a default order, or simply sort them according to the required resource volume, but starts from the three aspects of application construction attributes, business attributes, and status attributes. After comprehensive consideration, it sorts them according to the business volume after introducing importance weights, and optimizes the deployment according to the differentiated security needs of business applications.

[0007] When performing cloud-edge coordinated deployment, the present invention takes into account the latency of users using applications in a cloud-edge computing environment on the basis of balancing resource utilization, thereby achieving efficient resource utilization while ensuring the user's normal usage experience (i.e., latency).

[0008] The specific scheme flow chart of the present invention is as follows Figure 1 As shown, the steps include:

[0009] (1) Obtain a list of business applications P and a list of computing nodes K to be deployed;

[0010] (2) Calculate the delay matrix T of the terminal farthest from each edge computing node k in the node list K for each service application in the service application list P. The terminal refers to the user end that uses the service application, and the service application is deployed on the computing node. If the service application p is used by multiple terminals, if the service application p is deployed on the edge computing node k, the delay of the terminal farthest from the node k can be calculated as t pk , then the value T of the position of the element in the pth row and kth column of the delay matrix T pk is 1, if this delay t pk The value of is smaller than the maximum delay of the application p, that is, the delay matrix T selects the computing nodes that meet the delay requirements of all users of the application.

[0011] (3) Based on the list of all applications to be deployed and their feature information, the weighted computing power value is calculated and sorted in descending order to obtain a new business application list P_decay;

[0012] (4) Determine whether there are any undeployed applications in the business application list P_decay. If so, proceed to step (5); otherwise, terminate the deployment process.

[0013] (5) Select the top-ranked undeployed business application from the business application list P_decay to obtain the various types of resources required by it; that is, the values ​​of the various types of resources involved in the application feature information of the undeployed business application;

[0014] (6) Calculate the remaining computing power resources of all current computing nodes, count their standard computing power, and obtain the computing node matrix Q whose remaining computing power meets the business deployment requirements;

[0015] (7) If the current application to be deployed has affinity with the deployed application and the remaining computing power of the computing node where the affinity application is located meets the requirements of the current business application to be deployed, the current application to be deployed is deployed on the computing node with the most affinity services. The current application to be deployed is deployed and the process goes to step (11). Otherwise, the process goes to step (8).

[0016] (8) Based on the matrix Q and the matrix T, find the computing nodes among all computing nodes that meet the computing power resources and latency of the current application to be deployed, and then go to step (9). If all computing nodes only meet the computing power but not the latency, then go to step (10);

[0017] (9) Sort the computing nodes that meet the computing power and latency requirements according to the computing power standard values ​​to obtain a list of computing nodes in descending order, and deploy the current application to be deployed on the computing node ranked first. The current application to be deployed is deployed, and the process proceeds to step (11);

[0018] (10) Calculate the average latency from the computing nodes that meet the computing power value to each terminal that uses the current business application to be deployed, sort them according to the average latency, obtain a list of computing nodes in ascending order, and deploy the current application to be deployed on the computing node with the first ranking. The current application to be deployed is deployed, and enter step (11);

[0019] (11) Update the remaining resource information of the computing node, remove the currently deployed application from the business application list P_decay and return to step (4).

[0020] Furthermore, in the step (1), the service application list P to be deployed includes all applications to be deployed and feature information of the applications, as shown in Table 1 below:

[0021] Table 1 is the characteristic information table of the application

[0022]

[0023] At the same time, the computing node list K includes all the deployable computing nodes and their characteristics, as shown in Table 2:

[0024] Table 2 is the node feature table

[0025] feature meaning <![CDATA[C k ]]> The CPU resource quantification value of the computing node <![CDATA[R k ]]> Quantified value of RAM resources of this computing node <![CDATA[S k ]]> Quantified value of the hard disk storage resources of the computing node <![CDATA[G k ]]> The GPU resource quantification value of this computing node TR The throughput of the computing node

[0026] Furthermore, the latency described in step (2) refers to the time required for a terminal device to make a complete request to a business application deployed on an edge computing node and receive the calculation result. The specific process is: the device initiates a request, reaches the edge management gateway through a network link, is further routed to the edge computing node or cloud center for calculation, and finally returns the data result to the terminal device. Figure 2 Complete request process for the terminal device.

[0027] In this transmission process, the total delay is mainly composed of two parts: request uplink delay and calculation delay. The calculation formula is:

[0028] D tot =D up +Dcal

[0029] Among them, D tot Denotes the total delay, D up Indicates the requested uplink time, D cal Indicates the calculation delay.

[0030] Furthermore, the request uplink time in step (2) mainly includes the total time required for the device terminal to transmit to the edge management gateway and then to the edge computing node through each routing node. Ignoring the transmission delay between the terminal device and the edge management gateway, and considering the edge management gateway as a routing node, the transmission rate v between the nodes with a total number of nodes X can be recorded as:

[0031]

[0032] in,

[0033] Therefore, the routing time required for the application to send W traffic from the terminal device is:

[0034]

[0035] Furthermore, the calculation delay in step (2) is mainly for a certain computing node. When the computing resources and storage resources of the node meet the requirements of business application operation, the time required for calculation is related to the amount of business data and the device throughput, which is:

[0036]

[0037] In summary, when the lth service in step (2) is deployed to the kth computing node, the latency of the terminal farthest from the kth computing node is D lk The delay matrix T is:

[0038]

[0039] where t lk The value of is: If the application is deployed on this node and causes the farthest terminal delay D lk is greater than the maximum delay D_max allowed by the lth service application, then t lk =0; otherwise, t lk =1. Therefore, for the lth service, the number of nodes that meet the delay requirement is N delayAllowed,l ={k,if t lk ≤D_max},

[0040] Furthermore, the weighted computing power value calculation formula of the computing power node in step (3) is:

[0041] ω=ln(Js+B+St+2)*Ck *R k *S k *G k (1)

[0042] Furthermore, the standard computing power calculation formula of the computing node in step (6) is:

[0043] CP=C k *R k *S k *G k

[0044] The remaining computing power that meets the computing node matrix Q of the business deployment is:

[0045]

[0046] where v lk The value of is: If the computing power of the kth computing node does not meet the application requirements when the lth business is deployed on the kth computing node, then v lk =0; if the node computing power meets the requirement, then v lk =1.

[0047] The computing power of the computing node meets the requirements, which specifically means that the CPU, RAM, hard disk storage, and GPU occupied by all applications deployed on the edge node should not exceed the initial available resources of the node. kl , R kl , S kl , G kl The available resources of CPU, RAM, hard disk storage, and GPU allocated to the lth business for the kth computing node are:

[0048]

[0049]

[0050]

[0051]

[0052] Furthermore, the affinity mentioned in step (7) means that there are frequent interactions between certain applications and these interactions affect the application latency. In this case, the two applications should be deployed on the same server or in the same computer room. For certain applications that are heavily dependent on a data center and are sensitive to latency, the application should be deployed as close to the data center as possible.

[0053] Furthermore, in step (9), when ranking the computing power of each computing node, the central node needs to be considered after a delay, that is, when ranking, the central node is ranked last by default.

[0054] Furthermore, the specific operation steps of updating the remaining resource information of the computing node in step (11) are to subtract the quantified value of resources required for the business application of the currently deployed application during calculation from the computing node on which the business application is newly deployed.

[0055] The application and the corresponding deployed nodes obtained through the above steps are the final deployment solution. The present invention introduces the delay situation that has not been considered in the traditional Kubernetes scheduling method, and makes a certain guarantee for the delay while balanced deployment, so as to ensure the user experience during use as much as possible.

[0056] The advantages of the present invention are as follows:

[0057] The cloud-edge optimized balanced deployment strategy proposed in the present invention comprehensively considers differentiated security requirements such as application importance and affinity, and reasonably selects nodes for deploying business applications. It ensures that business latency is met, allocates computing power as fairly as possible, and optimizes cloud-edge resource utilization, which can effectively cope with resource constraints and real-time requirements in edge computing.

[0058] This solution not only takes into account the balance of application deployment, but also guarantees latency and improves user satisfaction. Compared with the deployment solution with minimum latency, this solution solves the problem of concentrated application deployment, which greatly improves the resource utilization overhead of computing nodes and avoids the situation where some computing nodes have too high resource occupancy and other resource nodes have too low occupancy when users make requests. Compared with the Kubernetes native scheduling model, this solution introduces the feature of latency guarantee. When deploying, it gives priority to deploying computing nodes that meet the maximum latency acceptable to users, reducing the average service processing latency and improving user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 The figure is a flow chart of the method of the present invention.

[0060] Figure 2 Complete request process diagram for terminal equipment.

[0061] Figure 3 This is a schematic diagram of the cloud-edge-end network experimental environment. DETAILED DESCRIPTION

[0062] The present invention is further described in detail below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention but not to limit the scope of the present invention.

[0063] Example 1:

[0064] (1) Obtain a list of business applications P and a list of computing nodes K to be deployed;

[0065] We designed Figure 3 The cloud-edge-end network environment shown has 7 routing nodes, 4 edge gateways, and 10 computing nodes, of which N0 is the central node. As an example, three applications and their respective user distributions are plotted here.

[0066] In the experimental environment, we assume that the network speed between computing nodes, routing nodes, and edge gateways, computing node computing resource information, and business application attribute information are as shown in Tables 3 and 4.

[0067] Table 3 is the list of computing nodes K

[0068]

[0069] Table 4 is a list of business applications P

[0070]

[0071]

[0072] (2) Calculate the delay matrix T of the terminal farthest from each edge computing node k for each service application; the network transmission rate between nodes is shown in Table 5:

[0073] Table 5 is the network transmission rate table

[0074]

[0075]

[0076] The delay matrix T is obtained as follows:

[0077] n0 n1 n2 n3 n4 n5 n6 n7 n8 n9 n10 app1 1 0 0 0 0 0 0 0 0 0 0 app2 1 1 1 1 1 0 1 0 0 1 1 app3 1 0 0 0 1 0 1 0 0 1 1 app4 1 0 0 0 1 0 1 0 0 1 1 app5 1 0 0 0 0 0 0 0 0 0 0 app6 1 1 1 1 1 1 1 1 1 1 1 app7 1 0 0 0 0 0 0 0 0 0 0 app8 1 0 0 0 1 0 1 0 0 1 1 app9 1 0 0 0 0 0 0 0 0 0 0 app10 0 0 0 0 0 0 0 0 0 0 0 app11 1 0 0 0 1 0 1 0 0 1 1 app12 1 1 1 1 1 1 1 1 1 1 1 app13 1 0 0 0 0 0 1 0 0 1 0 app14 1 0 0 0 0 0 1 0 0 1 0 app15 1 1 1 1 1 1 1 1 0 1 1 app16 1 1 1 1 1 1 1 1 1 1 1 app17 1 1 1 1 1 0 1 1 0 1 1 app18 1 0 0 0 1 0 1 0 0 1 1 app19 0 0 0 0 0 0 0 0 0 0 0 app20 1 1 1 1 1 1 1 1 1 1 1 app21 1 1 1 1 1 1 1 1 1 1 1 app22 1 1 1 1 1 1 1 1 1 1 1 app23 1 0 0 0 1 0 1 0 0 1 0 app24 1 0 0 0 1 0 1 0 0 1 1 app25 1 0 0 0 0 0 0 0 0 0 0

[0078] (3) All business applications to be deployed are sorted in descending order according to the weighted computing power value, and a new business application list P_decay is obtained as shown in Table 6:

[0079] Table 6 is a list of new business applications P_decay

[0080]

[0081]

[0082] (4) Determine whether there are any undeployed applications in the service application list P_decay. Based on the new service application list P_decay, it can be seen that there are still undeployed service applications, so it is possible to proceed to step (5).

[0083] (5) Select the top-ranked undeployed application from the business application list P_decay, which should be app22. Its properties are shown in Table 7:

[0084] Table 7 shows app22 and its application attributes

[0085] Application properties app22 CPU resource quantification value (C / time slice) 10 RAM resource quantification value (R / kb) 359 Hard disk storage resource quantification value (S / kb) 65 GPU resource quantization value (G) 5 Maximum delay allowed by the user (D_max / s) 0.8 Business volume (W / kb) 1729 Construction attributes (Js) 2 Business attributes (B) 1 Status attribute (St) 0 Affinity Application

[0086] (6) Calculate the remaining computing power resources of all current computing nodes, count their standard computing power, and obtain a list Q of computing nodes whose remaining computing power meets the business deployment requirements, as shown in Table 8 below:

[0087] Table 8 is a list of computing nodes that meet app22

[0088] n0 n1 n2 n3 n4 n5 n6 n7 n8 n9 n10 app22 1 1 1 1 1 1 1 1 1 1 1

[0089] (7) If the current application to be deployed has affinity with the already deployed application and the remaining computing power of the computing node where the affinity application is located meets the requirements, the current application to be deployed is deployed on the computing node with the most affinity services. Because app22 has no affinity business application, this step is skipped and the deployment node is found from the perspective of latency and computing power.

[0090] (8) Find the computing nodes among all computing nodes that meet the computing resources and latency of the current application to be deployed based on matrix Q and matrix T. Based on matrix Q and matrix T, it can be seen that all current computing nodes meet the requirements, so only one computing node with the largest computing resources needs to be found to deploy the application.

[0091] (9) Sort the computing nodes that meet the computing power and latency requirements according to the computing power standard values ​​to obtain a list of computing nodes in descending order, as shown in Table 9:

[0092] Table 9 is a list of computing node computing power rankings

[0093] Compute Node Hashrate n10 1084473000 n8 960000000 n9 881600000 n7 422240000 n3 393750000 n6 276276000 n5 240994000 n4 236544000 n2 231000000 n1 105000000 n0 12000000000

[0094] Since we try to avoid deploying applications on the central node n0 during deployment, even if n0 has the largest computing power, it will be placed at the end of the queue. Therefore, app22 is deployed on computing node n10.

[0095] (11) Update the remaining resource information of the computing node, remove the currently deployed application app22 from the business application list P_decay and return to step (4) to continue deploying the remaining business applications.

[0096] After the above cycle steps, the final deployment plan is shown in Table 10:

[0097] Table 10 shows the application deployment results

[0098] Application Name Deploy Node app1 n4 app2 n3 app3 n6 app4 n6 app5 n0 app6 n8 app7 n6 app8 n4 app9 n0 app10 n0 app11 n10 app12 n8 app13 n9 app14 n9 app15 n5 app16 n7 app17 n7 app18 n9 app19 n0 app20 n2 app21 n8 app22 n10 app23 n4 app24 n10 app25 n0

[0099] The comparison table with the results of other model deployments is shown in Table 11:

[0100] Table 11 is a comparison table of the evaluation results of each model deployment

[0101]

[0102]

[0103] It can be found that our deployment method has the highest score in user satisfaction. The satisfaction of minimum delay is not as good as this model. This is because in areas with a large number of users and tight resources, the minimum delay model cannot balance the resource utilization in a small range, resulting in some applications being unable to be deployed nearby and choosing nodes with the lowest delay that are still beyond the user's tolerance range. This drawback is also reflected in the fact that the minimum delay model has the highest resource utilization overhead. In terms of average business processing delay and resource utilization overhead, our model has similar resource utilization overhead to the balanced deployment model, but the average delay is reduced by 10%. In summary, our deployment method has the best overall effect.

[0104] In this solution, the formula for setting the weight in the deployment order of business applications is: ln(Js+B+St+2), but other weight formulas are introduced such as: It can also replace this solution.

[0105] Although the specific embodiments of the present invention are disclosed for the purpose of illustration, the purpose is to help understand the content of the present invention and implement it accordingly, those skilled in the art will understand that various substitutions, changes and modifications are possible without departing from the spirit and scope of the present invention and the appended claims. Therefore, the present invention should not be limited to the content disclosed in the best embodiment, and the scope of the present invention is subject to the scope defined in the claims.

Claims

1. A delay-guaranteed cloud-edge collaborative optimization balanced deployment method, the steps of which include: (1) Obtain a list P of business applications to be deployed and a list K of computing nodes used to deploy the business applications in the list P of business applications; Acquiring location information of a terminal using the service application; (2) For each edge computing node in the node list K, calculate the latency of the terminal farthest from the edge computing node accessing the service application on the edge computing node to obtain a latency matrix T; (3) Calculate the weighted computing power required by the corresponding business application according to the characteristic information of each business application in the application list P; sort the business applications in descending order according to the weighted computing power required by the business applications to obtain a business application list P_decay; (4) Determine whether there are any undeployed business applications in the business application list P_decay. If yes, proceed to step (5); otherwise, terminate the deployment process. (5) Select the top-ranked undeployed business application p from the business application list P_decay and obtain the various resources it needs to occupy; (6) Calculate the remaining computing power resources of each edge computing node in the current node list K, and calculate its standard computing power value, to obtain the edge computing node matrix Q whose remaining computing power resources meet the requirements for deploying the business application p; (7) If the business application p to be deployed and the deployed business application q have affinity, and the remaining computing power of the edge computing node k where the business application q is located satisfies the business application p, then the business application p is deployed on the edge computing node k and the process goes to step (11); otherwise, the process goes to step (8); (8) Based on the edge computing node matrix Q and the delay matrix T, find an edge computing node among the edge computing nodes that meets the computing power resources and delay requirements of the business application p. If found, proceed to step (9); if an edge computing node that only meets the computing power resources required by the business application p but does not meet the delay requirements of the business application p is found, proceed to step (10); (9) Sort the edge computing nodes that meet the computing power resources and latency requirements of the business application p according to the standard computing power value to obtain a list of edge computing nodes in descending order; deploy the business application p on the edge computing node ranked first in the list of edge computing nodes, and proceed to step (11); (10) Calculate the average delay from each edge computing node that meets the computing power required by the service application p but does not meet the delay requirement of the service application p to each terminal using the service application p, deploy the service application p on the edge computing node with the smallest average delay, and proceed to step (11); (11) Update the remaining resource information of each edge computing node in the node list K, remove the currently deployed application from the business application list P_decay and return to step (4).

2. The method according to claim 1, characterized in that: The characteristic information of the service application includes the CPU resource quantization value C, RAM resource quantization value R, hard disk storage resource quantization value S, GPU resource quantization value G required for the service application calculation, and the maximum delay D allowed by the service application in the terminal. max , the business application calculates the business processing data size W of a request, the user attribute U, construction attribute Js, business attribute B, status attribute St of the business application, and the application name Affinity with affinity; The computing power resources of the edge computing node include: CPU resource quantization value C k 、RAM resource quantization value R k , hard disk storage resource quantification value S k , GPU resource quantization value G k and throughput TR.

3. The method according to claim 2, characterized in that The weighted computing power value ω required by the business application = ln(Js+B+St+2)*C k *R k *S k *G k .

4. The method according to claim 2, characterized in that: The standard computing power value of the edge computing node CP = C k *R k *S k *G k .

5. The method according to claim 1, characterized in that The delay matrix The value of the element in the lth row and the kth column is based on D lk OK, D lk is the latency of the terminal farthest from edge computing node k when business application l is deployed to edge computing node k; if the latency D lk is greater than the maximum delay D_max allowed by the service application l, then t lk =0; On the contrary, t lk =1.

6. The method according to claim 1, characterized in that If the business application p to be deployed currently has affinity with multiple deployed business applications, the business application p is deployed to the edge computing node that has the most affinity with the business application p and whose remaining computing power satisfies the business application p.

7. The method according to claim 1, characterized in that Whether the two service applications have affinity is determined according to the feature information of the two service applications.

Citation Information

Patent Citations

  • Service chain deployment method and device for cloud-side collaborative Internet of Things

    CN114071582A

  • Service scheduling deployment method for edge computing node resources of Internet of Things

    CN114077485A

  • Computing power resource scheduling method adaptive to dynamic environment in edge-end cooperation scene

    CN116647604A

  • Heterogeneous multi-edge cloud collaborative micro-service deployment and routing joint optimization method and system

    CN116915686A

  • Joint optimization method and system for delay and spectrum occupation in cloud-edge collaborative network

    US20230421501A1