Balanced Leader Distribution Algorithm in Kubernetes Clusters

Abstract

Container-based virtualization is becoming a de facto way to build and deploy applications because of its simplicity and convenience. Kubernetes is a well-known open-source project that provides an orchestration platform for containerized applications. An application in Kubernetes can contain multiple replicas to achieve high scalability and availability. Stateless applications have no requirement for persistent storage; however, stateful applications require persistent storage for each replica. Therefore, stateful applications usually require a strong consistency of data among replicas. To achieve this, the application often relies on a leader, which is responsible for maintaining consistency and coordinating tasks among replicas. This leads to a problem that the leader often has heavy loads due to its inherent design. In a Kubernetes cluster, having the leaders of multiple applications concentrated in a specific node may become a bottleneck within the system. In this paper, we propose a leader election algorithm that overcomes the bottleneck problem by evenly distributing the leaders throughout nodes in the cluster. We also conduct experiments to prove the correctness and effectiveness of our leader election algorithm compared with a default algorithm in Kubernetes.

Node-Based Horizontal Pod Autoscaler in KubeEdge-Based Edge Computing Infrastructure

“…Having the leaders of multiple applications concentrated in a particular node in a Kubernetes cluster may cause the system to bottleneck. So some studies like [22], [23] provide leader selection algorithms that solve the bottleneck problem by uniformly distributing the leaders among nodes in the cluster. Although the experimental results of the study mentioned above show that the developed mechanisms are highly effective, they cannot be thoroughly applied in such an edge computing environment because they do not consider unstable or unestablished communication links resulting from the geographical distribution of the edge nodes.…”

Infrastructure Optimization in Kubernetes Cluster

“…In research paper [1] the creators Nguyen Dinh Nguyen and Taehong Kim portray the Kubernetes design and the pioneer based system for keeping up predictable information stockpiling among reproductions of a stateful application in the Kubernetes group. The creators propose a pioneer political race calculation that not just works with the utilization of pioneer political decision in Kubernetes yet in addition equitably disperses the pioneers all through every one of the hubs in the group.…”

Estimating the Deployment Time for Cloud Applications using Novel Google Kubernetes Cloud Service over Microsoft Kubernetes Cloud Service

“…Docker containers are supported by GKE. To help develop and serve application containers, Google Cloud provides continuous integration and continuous delivery technologies (Nguyen and Kim 2021;Burns, Beda, and Hightower 2019). Cloud Build may be used to create container images (such as Docker) from a variety of source code repositories, and Artifact Registry or Container Registry can be used to store and serve them.…”