Online application dynamic capacity expansion and shrinkage method based on micro-service call dependence perception

Abstract

According to the online application dynamic capacity expansion and shrinkage method based on the micro-service call dependence perception, an MAPE mode is adopted, and the whole system is divided intofour parts including a monitoring module, an analysis module, a planning module and an execution module; the monitoring module first reads operation data; the operation data comprises cluster resource use data and service call related data, the analysis module finds and selects an object to be subjected to capacity expansion or capacity reduction, the planning module calculates the number of containers needing capacity expansion or capacity reduction, and finally the execution module adjusts the number of copies in a specified container set to achieve micro-service level capacity expansion orcapacity reduction. The system can analyze the micro-service capability based on the calling dependency relationship and the application delay between the micro-services by analyzing the calling dependency relationship and the delay between the micro-services, comparing the calling dependency relationship and the delay with historical data, calculating and judging the current micro-service running state and the service capability when the request flow is changed, and performing capacity expansion and shrinkage on the specified micro-service; the resource utilization rate is improved while theservice quality is guaranteed.

Description

technical field [0001] The invention relates to the field of microservice architecture, in particular to a method for dynamically expanding and contracting online applications based on microservice call dependency perception. Background technique [0002] Among the large number of applications and services active on the Internet, online applications account for a large proportion. Many of these user-oriented applications are very sensitive to latency, such as web search, online ticketing systems, and e-commerce systems. User traffic for these applications often exhibits periodic or sudden fluctuations over time. Traffic fluctuations are workload fluctuations, which affect the resources required by the application and thus affect the quality of service. It is an important direction for online service optimization to meet users' requirements for delay while reducing resource utilization. The proposal and wide application of microservice architecture, thanks to its flexible ...

AI Technical Summary

Problems solved by technology

[0007] However, existing technologies often perform expansion and contraction judgments and operations on a single microservice, and seldom regard the microservice application as a whole, considering the impact of call dependencies on the expansion and contraction method

Images