The cloud-native architecture and microservice technologies are revolutionizing the design, development, and management of cloud applications and services by offering greater elasticity, scalability, and flexibility. However, managing service-to-service traffic and handling faults turn out to be more difficult for modern, sophisticated cloud-native applications. The research community responded to the technical challenges by exploring efficient scheduling schemes that deploy constituent services to a node. Despite those efforts, current solutions are unable to handle real-time traffic dynamics, which could lead to resource waste and unnecessary communication delays. In this work, service partitions are used to improve resource distribution and traffic control in microservice-based applications. This strategy uses graph-based techniques to effectively cluster services, optimize resource usage, and boost communication efficiency, while continually monitoring application behaviors. We found that it can reduce response times by up to 15% during times of high network latency. The performance and dependability of microservices in cloud-native environments can be significantly improved using the proposed approach.
Ahmed et al. (Thu,) studied this question.