Topology-Aware Scheduling Framework for Microservice Applications in Cloud

Loosely coupled and highly cohesived microservices running in containers are becoming the new paradigm for application development. Compared with monolithic applications, applications built on microservices architecture can be deployed and scaled independently, which promises to simplify software development and operation. However, the dramatic increase in the scale of microservices and east-west network traffic in the data center have made the cluster management more complex. Not only does the scale of microservices cause a great deal of pressure on cluster management, but also cascading QoS violations present a substantial risk for SLOs (Service Level Objectives). In this paper, we propose a Microservice-Oriented Topology-Aware Scheduling Framework (MOTAS), which effectively utilizes the topologies of microservices and clusters to optimize the network overhead of microservice applications through a heuristic graph mapping algorithm. The proposed framework can also guarantee the cluster resource utilization. To deal with the dynamic environment of microservice, we propose a mechanism based on distributed trace analysis to detect and handle QoS violations in microservice applications. Through real-world experiments, the framework has been proved to be effective in ensuring cluster resource utilization, reducing application end-to-end latency, improving throughput, and handling QoS violations.

Automated summary

Loosely coupled and highly cohesive microservices running in containers have become the new paradigm for application development. Compared to monolithic applications, microservices architecture allows for independent deployment and scaling, promising to simplify software development and operation. However, the increase in microservices scale and east-west network traffic in data centers has made cluster management more complex. This paper proposes a Microservice-Oriented Topology-Aware Scheduling Framework (MOTAS) that optimizes the network overhead of microservice applications by effectively utilizing microservices and cluster topologies through a heuristic graph mapping algorithm. The framework also guarantees cluster resource utilization and incorporates a mechanism for detecting and handling QoS violations in dynamic microservice environments.