Online Deployment Algorithms for Microservice Systems with Complex Dependencies

TL;DR

This paper addresses the challenge of deploying microservices with complex dependencies and multiple instances in cloud and edge environments by formulating the problem as a fractional polynomial problem, transforming it into a quadratic sum-of-ratios fractional problem, and proposing greedy algorithms that outperform existing methods.

Contribution

It introduces a novel formulation of the service placement problem considering complex dependencies and multiple instances, and presents efficient greedy algorithms for its solution.

Findings

The proposed algorithms outperform existing approaches in quality.

The algorithms demonstrate faster computation times.

The models effectively handle complex microservice dependencies.

Abstract

Cloud and edge computing have been widely adopted in many application scenarios. With the increasing demand of fast iteration and complexity of business logic, it is challenging to achieve rapid development and continuous delivery in such highly distributed cloud and edge computing environment. At present, microservice-based architecture has been the dominant deployment style, and a microservice system has to evolve agilely to offer stable Quality of Service (QoS) in the situation where user requirement changes frequently. Many research have been conducted to optimally re-deploy microservices to adapt to changing requirements. Nevertheless, complex dependencies between microservices and the existence of multiple instances of one single microservice in a microservice system have not been fully considered in existing works. This paper defines SPPMS, the Service Placement Problem in Microservice Systems that feature complex dependencies and multiple instances, as a Fractional Polynomial Problem (FPP) . Considering the high computation complexity of FPP, it is then transformed into a Quadratic Sum-of-Ratios Fractional Problem (QSRFP) which is further solved by the proposed greedy-based algorithms. Experiments demonstrate that our models and algorithms outperform existing approaches in both quality and computation speed.