Multi-objective microservice deployment optimization via a knowledge-driven evolutionary algorithm

Abstract

For the deployment and startup of microservice instances in different resource centres, we propose an optimization problem model based on the evolutionary multi-objective theory. The objective functions of the model consider the computation and storage resource utilization rate, load balancing rate, and actual microservice usage rate in resource service centres. The constraints of the model are the completeness of service, total amount of storage resources, and total number of microservices. In this study, a knowledge-driven evolutionary algorithm (named MGR-NSGA-III) is proposed to solve the problem model and seek the optimal deployment and startup strategy of microservice instances in different resource centres. The proposed model and solution have been evaluated via real data experiments. The results show that our approach is better than the traditional microservice instance deployment and startup strategy. The average computation rate, storage idle rate, and actual microservice idle rate were 13.21%, 5.2%, and 16.67% lower than those in NSGA-III, respectively. After 50, 100, and 150 evolutionary generations in serval operations, the population members in NGR-NSGA-III dominated the population members in NSGA-III 6,270, 3,581, and 7,978 times in average, respectively, which means that NGR-NSGA-III can converge to the optimal solution much quicker than NSGA-III.

A container deployment strategy for server clusters with different resource types

“…Here, we evaluate the proposed AF‐CSDS deployment strategy based on the above experimental data and the simulation running environment. In the experiment, it was compared with existing deployment strategies such as APSO‐TSDS, 5 MSG‐NSGA‐III, 20 ACO‐MCMS, 19 GA‐NSGA‐II, 17 and the Spread algorithm implemented in Docker Swarm 8 The APSO‐TSDS strategy is a service deployment strategy based on an accelerated particle swarm optimization algorithm.…”

“…Lin et al 19 designed an improved ant colony optimization (ACO) algorithm that realized the container deployment of microservices by optimizing network transmission costs, load balancing of clusters, and average failure rate of microservices. Ma et al 20 built a knowledge‐driven evolutionary algorithm using the NSGA‐II algorithm for reference points (i.e., NSGA‐III) to solve the deployment and start‐up of microservices. The algorithm considers the actual idle rate of microservices, idle rate of computing and storage resources, and load balancing of computing and storage resources.…”

MHDNNL

“…The proposed algorithm achieves the best performance on the DTLZ test suite, comparing with NSGAIII, GrEA, KnEA, VaEA and Two-Arch2. (Ma, Wang, Gu, Meng & Wu, 2021) proposed a multi-objective knowledge-driven evolutionary algorithm (MGR-NSGA-III) for microservice deployment and startup strategy problem in different data centers. The MGR-NSGA-III takes completeness of service, total amount of storage resources, and total number of microservices as the constraints, considers the computation and storage resource utilization rate, load balancing rate, and actual microservice usage rate in resource service centers, and gains a better performance than the traditional microservice instance deployment and startup strategy does.…”

From Monolithic to Microservice Architecture: The Case of Extensible and Domain-Specific IDEs

“…The question of microservices deployment and more generally of cloud deployment is complex due to the multiple factors taken into account in addition to the response time [23]. Elements such as availability, free storage, CPU usage or memory usage are constraints that are imposed to the deployment centers, which have specific needs [24].…”

“…A Sirius Web application relies on a set of modules called Sirius Components 23 . In this work, we focus in particular on the backend components managing the back-end part of Sirius Web.…”