Micro-service dynamic flow limiting system and method

Abstract

The invention relates to a micro-service dynamic flow limiting system. The system comprises a service distributor, a service instance configurator, a bill pool and an important service instance queue, the service distributor is respectively connected with the service instance configurator, the bill pool and the important service instance queue; the service distributor receives a service request and determines which service instance of the service the request is sent to; the service instance configurator is used for recording the information instanceinfo of each service instance; the ticket pool (ticketpool) is used for storing a ticket (ticket); the important service instance queue stores service instance information specially used for important service requests. According to the invention, the flow distribution can be dynamically and smoothly adjusted according to the real-time change of the service request number, and the full utilization of the service capability is ensured.

Description

technical field [0001] The present invention relates to the field of microservice architecture, in particular to a system and method for dynamic current limiting of microservices. Background technique [0002] In the microservice architecture, it is often necessary to limit the flow of some of the microservices to prevent the influx of a large number of service requests from being able to handle the congestion of the microservices in a high concurrency situation, which in turn affects other microservices and causes the system to be paralyzed . [0003] Existing common current limiting methods, including common current limiting methods such as waiting queue, leaky bucket, token bucket, and counter method, can well limit the number of requests per unit time and ensure the normal operation of the service. However, these methods often impose a unified flow limit on all service requests. Regardless of whether a service request is important or not, it cannot be executed under the...

AI Technical Summary

Problems solved by technology

However, these methods often impose a unified flow limit on all service requests. Regardless of whether a service request is important or not, it cannot be executed under current limit conditions.

On this basis, some current limiting methods add priority configuration, give priority to high-priority service requests, and block and limit low-priority service requests, but often occur when concurrent requests are not high, and too many resources are reserved The traffic is sent to high priority requests, and the service capacity is idle, but the low priority requests are still not executed

Images