The Information Technology sector has undergone tremendous changes arising due to the emergence and prevalence of Cloud Computing. Microservice Architectures have also been attracting attention from several industries and researchers. Due to the suitability of microservices for the Cloud environments, an increasing number of Cloud applications are now provided as microservices. However, this transition to microservices brings a wide range of infrastructural orchestration challenges. Though several research works have discussed the engineering of microservice-based applications, there is an inevitable need for research on handling the operational phases of the microservice components. Microservice application deployment in containerized datacenters must be optimized to enhance the overall system performance. In this research work, the deployment of microservice application modules on the Cloud infrastructure is first modelled as a Binary Quadratic Programming Problem. In order to reduce the adverse impact of communication latencies on the response time, the interaction pattern between the microservice components is modelled as an undirected doubly weighted complete Interaction Graph. A novel, robust heuristic approach IntMA is also proposed for deploying the microservices in an interaction-aware manner with the aid of the interaction information obtained from the Interaction Graph. The proposed allocation policies are implemented in Kubernetes. The effectiveness of the proposed approach is evaluated on the Google Cloud Platform, using different microservice reference applications. Experimental results indicate that the proposed approach improves the response time and throughput of the microservice-based systems.

由于云计算的出现和普及，信息技术领域发生了巨大的变化。微服务架构也已经引起了多个行业和研究人员的关注。由于微服务适用于云环境，因此现在提供越来越多的云应用程序作为微服务。但是，这种向微服务的过渡带来了各种各样的基础架构流程挑战。尽管一些研究工作已经讨论了基于微服务的应用程序的工程设计，但是不可避免地需要进行有关处理微服务组件的操作阶段的研究。必须优化容器化数据中心中的微服务应用程序部署，以增强整体系统性能。在这项研究工作中，微服务应用程序模块在云基础架构上的部署首先被建模为二进制二次编程问题。为了减少通信延迟对响应时间的不利影响，微服务组件之间的交互模式被建模为无向双权完整交互图。新颖，强大的启发式方法还提出了IntMA，用于借助从交互图获得的交互信息，以交互感知的方式部署微服务。建议的分配策略在Kubernetes中实施。使用不同的微服务参考应用程序，在Google Cloud Platform上评估了所建议方法的有效性。实验结果表明，所提出的方法提高了基于微服务的系统的响应时间和吞吐量。