Route dynamic adjustment method, device and system

Abstract

The invention discloses a method, a device and a system for dynamically adjusting a route. The method comprises the following steps: monitoring the resource use condition of each service node; calculating the routing weight of each service node according to the resource use condition of each service node; and performing routing forwarding on the application service request according to the calculated routing weight. According to the scheme provided by the invention, better service node load balancing can be realized.

Description

Technical Field

[0001] The present invention relates to the field of network technologies, and in particular, to a method, device, and system for dynamically adjusting routing. Background Art

[0002] In a microservices architecture system, it is a widely adopted and efficient method to improve the concurrent service ability of the system through load balancing and multi-node deployment of applications. Currently, most load balancing routing strategies for applications are based on static configurations, and common routing strategies include round-robin mode, random mode, weight mode, etc.

[0003] As the application service continues to run, the performance differences and resource usage of service nodes change significantly. For example, the central processing unit (CPU) or memory occupancy ratio of some service nodes may be higher than that of other nodes. If the request forwarding is still carried out according to the static configuration strategy, the requests received by the nodes with higher CPU or memory usage do not decrease, which may lead to serious problems such as triggering alarms, system jams, or even crashes on these nodes, while the resources of the nodes with lower CPU or memory usage are wasted, and the maximum utilization of resources is not achieved. Summary of the Invention

[0004] The present invention provides a method, device, and system for dynamically adjusting routing, which can achieve better load balancing of service nodes.

[0005] According to one aspect of the present invention, there is provided a method for dynamically adjusting routing, including:

[0006] Monitoring the resource usage of each service node;

[0007] Calculating the routing weight of each service node according to the resource usage of each service node;

[0008] Performing routing forwarding on application service requests according to the calculated routing weights.

[0009] In one embodiment, monitoring the resource usage of each service node includes:

[0010] Periodically obtaining the resource usage of each service node.

[0011] In one embodiment, the resource usage includes processor resource usage and memory resource usage.

[0012] In one embodiment, calculating the routing weight of each service node according to the resource usage of each service node includes:

[0013] Calculating the remaining processing capacity of each service node according to the resource usage of each service node;

[0014] Calculate the routing weights of each service node according to the remaining processing capabilities of each service node. The routing weight of each service node represents the proportion of the remaining processing capabilities of each node in the total processing capabilities of all service nodes.

[0015] In one embodiment, routing and forwarding the application service request according to the calculated routing weights includes:

[0016] Prioritize forwarding the application service request to the service node with a higher routing weight.

[0017] According to another aspect of the present invention, there is provided a routing dynamic adjustment device, including:

[0018] A resource monitoring module for monitoring the resource usage of each service node;

[0019] A weight calculation module for calculating the routing weights of each service node according to the resource usage of each service node;

[0020] An intelligent routing module for routing and forwarding the application service request according to the calculated routing weights.

[0021] In one embodiment, the resource monitoring module is specifically configured to periodically obtain the resource usage of each service node.

[0022] In one embodiment, the resource usage includes the processor resource usage and the memory resource usage.

[0023] In one embodiment, the weight calculation module is specifically configured to calculate the remaining processing capabilities of each service node according to the resource usage of each service node; calculate the routing weights of each service node according to the remaining processing capabilities of each service node. The routing weight of each service node represents the proportion of the remaining processing capabilities of each node in the total processing capabilities of all service nodes.

[0024] In one embodiment, the intelligent routing module is specifically configured to prioritize forwarding the application service request to the service node with a higher routing weight.

[0025] According to another aspect of the present invention, there is provided a routing dynamic adjustment system, including:

[0026] A node resource monitoring service module deployed on each service node for monitoring the resource usage of the service node where it is located;

[0027] A monitoring center service module deployed on the routing control node for collecting the resource usage of each service node;

[0028] A dynamic weight calculation service module deployed on the routing control node for calculating the routing weights of each service node according to the resource usage of each service node;

[0029] The intelligent routing service module is deployed on the routing control node and is used to perform routing forwarding on the application service request according to the calculated routing weight.

[0030] In one embodiment, the node resource monitoring service module is specifically used to monitor the resource usage of the service node where it is located; and periodically report the resource usage of the service node where it is located to the monitoring center service module.

[0031] In one embodiment, the resource usage includes the processor resource usage and the memory resource usage.

[0032] In one embodiment, the dynamic weight calculation service module is specifically used to calculate the remaining processing capacity of each service node according to the resource usage of each service node; and calculate the routing weight of each service node according to the remaining processing capacity of each service node. The routing weight of each service node represents the proportion of the remaining processing capacity of each node in the total processing capacity of all service nodes.

[0033] In one embodiment, the intelligent routing service module is specifically used to preferentially forward the application service request to the service node with a higher routing weight.

[0034] The technical solution of the embodiment of the present invention monitors the resource usage of each service node, and then calculates the routing weight of each service node according to the resource usage of each service node, so that the application service request can be routed and forwarded according to the calculated routing weight. For an application service that provides multiple differentiated interfaces and adopts a multi-center and multi-active deployment architecture, better load balancing can be achieved, the reasonable utilization of the resources of each service node can be guaranteed, and the concurrency performance can be improved.

[0035] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0037] Figure 1 It is a schematic flowchart of a routing dynamic adjustment method provided by an embodiment of the present invention;

[0038] Figure 2Schematic structural diagram of a routing dynamic adjustment device provided by an embodiment of the present invention;

[0039] Figure 3 Schematic diagram of the routing adjustment interaction process of a routing dynamic adjustment system provided by an embodiment of the present invention. Detailed implementation manners

[0040] In order to enable those skilled in the art of the present technology to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention. The acquisition, storage, use, processing, etc. of data in the technical solutions of this application all comply with the relevant regulations of national laws and regulations.

[0041] Microservices (or microservice architecture) is a cloud-native architecture approach that includes numerous small, loosely coupled, and independently deployable components or services in a single application. The microservice architecture is suitable for scenarios such as banks that need to deploy service nodes at multiple locations. Taking a bank as an example, when handling personal card applications, card binding, and other services, the bank needs to verify personal identity information. Due to the wide business application scenarios of the identity information verification service, the average daily business volume of medium and large-sized banks is about tens of millions of transactions. The processing efficiency and business continuity of the identity information verification service directly affect the bank's business capabilities and user experience. To ensure the processing efficiency, business continuity, and growing business needs of the identity information verification service, the bank uses the microservice architecture to deploy service nodes in multiple locations to ensure the multi-center multi-active operation of this service.

[0042] Due to the large average daily business volume and high business peaks of the identity information verification service, if static configuration-based load balancing routing is used to forward requests, the requests may be forwarded to nodes with poor processing performance, resulting in the response time of the requests not meeting expectations, or the possibility of request failure due to exceeding the maximum waiting time, affecting the user experience.

[0043] Since the identity information verification service provides multiple verification interfaces, some verification interfaces have simple processing logics and fast responses, while some have complex processing logics and slow responses. If the load balancing routing based on the minimum number of connections is used to forward requests, assuming that a certain node accepts only complex verification requests, resulting in the connections not being released all the time, its connection number will be in a high state. However, the CPU and memory usage of this node is not high. But due to its high connection number, requests will no longer be forwarded to this node, causing waste of node resources. And some nodes may process requests more simply and release requests faster, so their connection numbers are lower, resulting in more requests being forwarded to this node, causing a resource performance bottleneck for this node.

[0044] Since the service nodes of the identity information verification service are distributed in different city centers, there are significant differences in their network environments. If the load balancing routing based on the shortest response time is used to forward requests, it may cause requests to be distributed to nodes with better network environments, resulting in a high load on this node, while the resources of nodes with relatively poor network environments cannot be reasonably utilized.

[0045] In summary, the current static routing adjustment strategy is difficult to adapt to the microservice architecture with multiple service nodes providing services together.

[0046] Figure 1 The flowchart of a routing dynamic adjustment method provided by an embodiment of the present invention is as Figure 1 shown. The routing dynamic adjustment method provided in this embodiment includes:

[0047] Step S110, monitor the resource usage of each service node.

[0048] The routing dynamic adjustment method provided by an embodiment of the present application is applied to a scenario where multiple service nodes provide business services for users and is executed by a routing dynamic adjustment system deployed on each service node and the central routing control node. Each step in the routing dynamic adjustment method provided by an embodiment of the present application can be executed by a device node, a hardware module or a software module deployed in the device node.

[0049] For scenarios that need to face large-scale service demands, multiple service nodes need to be deployed to meet the requirements of large-scale concurrent application services. The multiple service nodes can be deployed at the same location or at multiple different locations. And to avoid the situation where service nodes become unavailable due to unexpected circumstances, which may further lead to the termination of application services, the multiple service nodes are preferably deployed at different locations. The resource configurations of different service nodes may be the same or different, so the service capabilities that each service node can provide may also be the same or different. The service capabilities that a service node can provide are determined according to its resource usage. Among them, the resource usage of a service node is the usage of resources that have an impact on the service capabilities that can be provided. In this embodiment, the resource usage of each service node is monitored, that is, the resource changes of each service node are monitored.

[0050] In the embodiments of this application, the resource usage of the processor (CPU) and the memory resources of each service node can be used as the resource usage of each service node. The CPU and the memory are the resources of the service node that have the greatest impact on routing forwarding. Among them, the CPU affects the computing power of the service node, and the memory affects the caching ability of the computing resources of the service node. That is, the usage of the processor and the content usage of each service node are monitored. The processor usage and the content usage can be expressed as percentages.

[0051] Step S120, calculate the routing weights of each service node according to the resource usage of each service node.

[0052] After obtaining the resource usage of each service node, the routing weights of each service node can be calculated. The more resources a service node has already used, the smaller the ability of the service node to provide application services. The more resources a service node has already used means a higher load. At this time, even if the service node still has available resources, it will affect the performance of the service node. Therefore, using a service node with a lower load to provide services for the application can make the operation efficiency of the service node higher, and the service experience of the application service is also better. Therefore, the routing weights of each service node can be calculated based on the resource usage of each service node. The routing weights of each service node are calculated according to the resource usage that can affect the application service of the service node. The calculation of the routing weights comprehensively considers the resource usage of each service node, and finally calculates the respective routing weights of all service nodes. The higher the routing weight, the more remaining available resources of the service node.

[0053] In one embodiment, to calculate the routing weights of each service node, the remaining processing capacity of each service node may first be calculated according to the resource usage of each service node; then, the routing weights of each service node are calculated based on the remaining processing capacity of each service node, where the routing weight of each service node represents the proportion of the remaining processing capacity of each node in the total processing capacity of all service nodes.

[0054] Since the resource configuration of each service node may be different, when calculating the routing weights of each service node, the initial routing weight of each service node also needs to be considered. The higher the resource configuration ability of the service node, the higher the initial routing weight. Then, the remaining processing capacity of each service node is calculated according to the resource usage of each service node. Next, the proportion of the remaining processing capacity of each service node in the total processing capacity is calculated based on the remaining processing capacity of each service node. Finally, considering the proportion of the remaining processing capacity of each service node in the total processing capacity of all service nodes, the routing weights of each service node can be determined, that is, the proportion of the remaining processing capacity of each service node in the remaining processing capacity of all service nodes.

[0055] When the resource usage of the service node includes multiple resources, the usage of different resources may have the same weight coefficient or different weight coefficients. For different business and application usage requirements, weight coefficients matching different resources can be configured. For example, when the resource usage of the service node includes two resources, CPU and memory, for a business with a higher demand for computing resources, the weight coefficient of the CPU can be set higher, while for a business with a larger amount of data, the weight coefficient of the content can be set higher.

[0056] Step S130: Route and forward the application service request according to the calculated routing weight.

[0057] After calculating the routing weight, the application service request can be routed and forwarded according to the routing weight. After receiving the application service request, the application service request can be preferentially forwarded to the service node with a higher routing weight, so that the resources of each service node can be evenly used, thereby achieving load balancing between service nodes.

[0058] The routing dynamic adjustment method provided in this embodiment monitors the resource usage of each service node, then calculates the routing weights of each service node according to the resource usage of each service node, and thus can route and forward the application service request according to the calculated routing weights. For an application service that provides multiple differentiated interfaces and adopts a multi-center and multi-active deployment architecture, it can achieve better load balancing, ensure the reasonable utilization of the resources of each service node, and improve the concurrency performance.

[0059] In one embodiment, monitoring the resource usage of each service node can be to periodically obtain the resource usage of each service node. Frequent adjustment of the routing forwarding policy for each service node may also cause additional resource consumption. Therefore, the routing forwarding policy can be adjusted periodically. The resource usage of each service node can be obtained at a certain preset period, and the preset period can be determined according to the concurrency of the actual application service. The higher the concurrency of the application service, the shorter the preset period can be set, and vice versa. After each preset period expires, collect the resource usage of each service node, calculate the routing weights based on the periodically collected resource usage of each service node, and perform routing forwarding for this preset period based on the calculated routing weights of each service node.

[0060] The routing weight calculation strategy provided by the embodiments of the present application will be described in detail below. Taking the CPU and memory usage of the service node as an example for illustration.

[0061] 1. It is necessary to define the initial weights of each service node, set an initial weight for each service node, representing its default service capacity, and the sum of the initial weights of all service nodes should be 100%. Suppose there are two service nodes, node1 and node2, and the resource configurations of the two service nodes are the same, then the initial weights of the two service nodes are both 50%, which means that the CPU and content configurations of each service node are the same.

[0062] 2. Define the weight coefficients. According to business requirements and factors such as whether the application is mainly compute-intensive or input / output (IO)-intensive, define the CPU configuration weight coefficient (w_cpu) and the memory configuration weight coefficient (w_mem). The sum of the weight coefficients should be 100%. Assuming that the CPU and memory configurations have the same impact on the service, the weight coefficients are both 50%.

[0063] 3. Define the initial processing capacity of the service node: pc_nodei = the CPU configuration of each node * w_cpu + the memory configuration * w_mem, where i is the number of the service node.

[0064] 4. Define the average percentage of CPU and memory usage of the service node: The average percentages of CPU and memory usage of each service node within a routing cycle are avg_cpu_nodei and avg_mem_nodei respectively.

[0065] 5. Dynamic weight calculation

[0066] First, calculate the remaining processing capacity (rpc_nodei) of each service node

[0067] rpc_nodei = cpu configuration * (1 - avg_cpu_nodei) * w_cpu + memory configuration * (1 - avg_mem_nodei) * w_mem 2)

[0068] Then calculate the percentage (rpc_ratio) of the sum of the remaining processing capabilities of all service nodes to the sum of the initial total processing capabilities

[0069] rpc_ratio = ∑rpc_nodei / ∑pc_nodei

[0070] Finally, calculate the routing weight value (weight_nodei) of each service node

[0071] weight_nodei = weight_nodei * rpc_ratio + rpc_nodei / sum of the remaining processing capabilities of all nodes * (1 - rpc_ratio)

[0072] The original weight values are first reduced by the percentage of the used processing capabilities to the initial processing capabilities, and then the weights of the percentage of the unused processing capabilities to the initial processing capabilities are re - distributed using the newly calculated remaining processing capabilities

[0073] Figure 2 This is a schematic structural diagram of a routing dynamic adjustment device provided by an embodiment of the present invention. As Figure 2 shown, the routing dynamic adjustment device provided in this embodiment includes:

[0074] A resource monitoring module 21, used to monitor the resource usage of each service node; a weight calculation module 22, used to calculate the routing weights of each service node according to the resource usage of each service node; an intelligent routing module 23, used to perform routing forwarding on application service requests according to the calculated routing weights

[0075] The routing dynamic adjustment device provided in this embodiment is used to implement Figure 1 the routing dynamic adjustment method provided by the embodiment shown. Its implementation principle and technical effects are similar, and will not be elaborated here

[0076] In one embodiment, the resource monitoring module 21 is specifically used to periodically obtain the resource usage of each service node

[0077] In one embodiment, the resource usage includes the processor resource usage and the memory resource usage

[0078] In one embodiment, the weight calculation module 22 is specifically configured to calculate the remaining processing capacity of each service node according to the resource usage of each service node; calculate the routing weight of each service node according to the remaining processing capacity of each service node, and the routing weight of each service node represents the proportion of the remaining processing capacity of each node in the total processing capacity of all service nodes.

[0079] In one embodiment, the intelligent routing module 23 is specifically configured to preferentially forward the application service request to the service node with a higher routing weight.

[0080] The embodiment of the present application further provides a routing dynamic adjustment system, including:

[0081] The node resource monitoring service module is deployed on each service node and is used to monitor the resource usage of the service node where it is located.

[0082] The monitoring center service module is deployed on the routing control node and is used to collect the resource usage of each service node.

[0083] The dynamic weight calculation service module is deployed on the routing control node and is used to calculate the routing weight of each service node according to the resource usage of each service node.

[0084] The intelligent routing service module is deployed on the routing control node and is used to perform routing forwarding on the application service request according to the calculated routing weight.

[0085] The service node is a node that provides business services. The node resource monitoring service module is deployed on each service node. The node resource monitoring service module can be a physical module or a software module, and the node resource monitoring service module can also be an independent device. The routing control node is an independent network node, or the routing control node can also be a hardware or software module deployed in any network node. The monitoring center service module, the dynamic weight calculation service module, and the intelligent routing service module are software or hardware modules deployed on the reason control node.

[0086] In one embodiment, the node resource monitoring service module is specifically configured to monitor the resource usage of the service node where it is located; periodically report the resource usage of the service node where it is located to the monitoring center service module.

[0087] In one embodiment, the resource usage includes the processor resource usage and the memory resource usage.

[0088] In one embodiment, the dynamic weight calculation service module is specifically configured to calculate the remaining processing capacity of each service node according to the resource usage of each service node; calculate the routing weight of each service node according to the remaining processing capacity of each service node, and the routing weight of each service node represents the proportion of the remaining processing capacity of each node in the total processing capacity of all service nodes.

[0089] In one embodiment, the intelligent routing service module is specifically configured to preferentially forward an application service request to a service node with a higher routing weight.

[0090] Figure 3 FIG. is a schematic diagram of a routing adjustment interaction process of a routing dynamic adjustment system provided based on an embodiment of the present invention. Figure 3 Taking two service nodes as an example for illustration, Figure 3 The monitoring center service module is not shown in FIG.. Actually, the resource usage conditions of the service nodes sent by the node resource monitoring service modules of each service node are first sent to the monitoring center service module, and then sent by the monitoring center service module to the public security dynamic weight calculation service module. The node resource monitoring service modules of each service node collect the average CPU usage rate and the average memory usage rate of the node where they are located every routing cycle, and send the data to the dynamic weight calculation service module through the monitoring center service module. The dynamic weight calculation service module collects the CPU and memory data of each service node within a routing cycle, recalculates the weights of each node according to the calculation method, and compares with the previous weight data. If they are inconsistent, the new weight data is pushed to the intelligent routing service module, so that it can forward the requests of the application using the new routing policy.

[0091] Next, taking the case where a bank develops an identity information verification service for services such as card application and card binding and adopts a multi-center and multi-active deployment framework, the routing dynamic adjustment method provided by the embodiment of the present application will be described.

[0092] The bank develops an identity information verification service for use in services such as card application and card binding. The service provides multiple verification modes, such as a simple item mode, a portrait mode, etc., and the response times of different modes vary greatly. The identity information verification service adopts a microservice architecture and is deployed on multiple nodes. Two nodes, AppAA and AppBB, are deployed in the AA center and the BB center respectively, and the network environments are quite different. The load balancing router (i.e., the routing control node) is deployed in the AA center. To ensure the balanced load of business traffic, the load balancing routing policy adopts the dynamic routing adjustment based on the running conditions of application service resources provided by this solution.

[0093] The node resource monitoring services are respectively deployed on the node servers where the application services are located, that is, on the two nodes, AppAA and AppBB.

[0094] The monitoring center service, the dynamic weight calculation service, and the intelligent routing service are deployed on the node where the load balancing router is located.

[0095] Since the CPU and memory configurations of the two nodes, AppAA and AppBB, are the same, assumed to be 16C32G (16 cores and 32GB), the initial routing ratios of the two nodes are 50% and 50% respectively. When a request is sent to the router, it will be evenly distributed to the two nodes for processing.

[0096] Since this application service has no special requirements for CPU and memory, the weight systems of CPU and memory are set to be equal, 50% and 50% respectively. It can be obtained that the total processing capacity of each node is 16 * 50% + 32 * 50% = 24, and the sum of the initial processing capacities of the two nodes is 48.

[0097] As the business progresses, the CPU and memory occupancy rates of the two nodes change. The node resource monitoring service AA reports that the avg_cpu_nodeAA of the AppAA node in the most recent 1 hour is 45% and the avg_mem_nodeAA is 55%. The node resource monitoring service BB reports that the avg_cpu_nodeBB of the AppBB node in the most recent 1 hour is 20% and the avg_mem_nodeBB is 25%.

[0098] The dynamic weight calculation service collects the CPU and memory occupancy rates of the two nodes, and calculates the routing weights of the two nodes according to the weight policy calculation formula: The remaining processing capacity rpc_nodeAA of the AppAA node is 16 * (1 - 45%) * 50% + 32 * (1 - 55%) * 50% = 11.6. The remaining processing capacity rpc_nodeBB of the AppBB node is 16 * (1 - 20%) * 50% + 32 * (1 - 25%) * 50% = 18.4.

[0099] rpc_ratio = (11.6 + 18.4) / 48 = 62.5%

[0100] The routing weight weight_nodeAA of the AppAA node is 50% * 62.5% + 11.6 / 30 * 37.5% = 45.75%

[0101] The routing weight weight_nodeBB of the AppBB node is 50% * 62.5% + 18.4 / 30 * 37.5% = 54.25%

[0102] The routing weights of the two nodes change. The weight of the AppBB node increases, and the weight of the AppAA node decreases.

[0103] The dynamic weight calculation service pushes the new routing policy to the intelligent routing service. After the next request arrives at the router, the request will be preferentially given to the AppBB node for processing.

[0104] For an application service that provides multiple differentiated interfaces and adopts a multi - center and multi - active deployment architecture, better load balancing can be achieved based on this method and device, ensuring the reasonable utilization of resources of each node and improving the concurrent performance.

[0105] It should be understood that various forms of processes shown above can be used, re - ordering, adding or deleting steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is made herein.

[0106] The above - mentioned specific implementation manners do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub - combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for dynamically adjusting a route, characterized in that including: monitoring the resource usage of each service node; calculating the routing weights of each service node according to the resource usage of each service node; routing and forwarding the application service requests according to the calculated routing weights.

2. The method according to claim 1, wherein The monitoring of the resource usage of each service node includes: periodically obtaining the resource usage of each service node.

3. The method according to claim 1 or 2, characterized in that The resource usage includes the processor resource usage and the memory resource usage.

4. The method according to claim 3, wherein The calculating of the routing weights of each service node according to the resource usage of each service node includes: calculating the remaining processing capacity of each service node according to the resource usage of each service node; calculating the routing weights of each service node according to the remaining processing capacity of each service node, where the routing weights of each service node represent the proportion of the remaining processing capacity of each node in the total processing capacity of all service nodes.

5. The method according to claim 4, wherein The routing and forwarding of the application service requests according to the calculated routing weights includes: preferably forwarding the application service requests to the service node with a higher routing weight.

6. A routing dynamic adjustment device, characterized in that, including: a resource monitoring module for monitoring the resource usage of each service node; a weight calculation module for calculating the routing weights of each service node according to the resource usage of each service node; an intelligent routing module for routing and forwarding the application service requests according to the calculated routing weights.

7. The device according to claim 6, characterized in that, The resource monitoring module is specifically used for periodically obtaining the resource usage of each service node.

8. The device according to claim 6 or 7, characterized in that, The resource usage includes the processor resource usage and the memory resource usage.

9. The device according to claim 8, characterized in that, The weight calculation module is specifically used for calculating the remaining processing capacity of each service node according to the resource usage of each service node; calculating the routing weights of each service node according to the remaining processing capacity of each service node, where the routing weights of each service node represent the proportion of the remaining processing capacity of each node in the total processing capacity of all service nodes.

10. The device according to claim 9, characterized in that, The intelligent routing module is specifically used for preferably forwarding the application service requests to the service node with a higher routing weight.

11. A routing dynamic adjustment system, characterized in that, including: a node resource monitoring service module deployed on each service node for monitoring the resource usage of the service node where it is located; a monitoring center service module deployed on the routing control node for collecting the resource usage of each service node; a dynamic weight calculation service module deployed on the routing control node for calculating the routing weights of each service node according to the resource usage of each service node; an intelligent routing service module deployed on the routing control node for routing and forwarding the application service requests according to the calculated routing weights.

12. The system according to claim 11, wherein The node resource monitoring service module is specifically used for monitoring the resource usage of the service node where it is located; periodically reporting the resource usage of the service node where it is located to the monitoring center service module.

13. The system according to claim 11 or 12, characterized in that, The resource usage includes the processor resource usage and the memory resource usage.

14. The system according to claim 13, wherein, The dynamic weight calculation service module is specifically used for calculating the remaining processing capacity of each service node according to the resource usage of each service node; calculating the routing weights of each service node according to the remaining processing capacity of each service node, where the routing weights of each service node represent the proportion of the remaining processing capacity of each node in the total processing capacity of all service nodes.

15. The system according to claim 14, wherein, The intelligent routing service module is specifically used to preferentially forward application service requests to service nodes with higher routing weights.

Images