More and more applications are organized in the form of meshed micro-services which can be deployed on the popularcontainer orchestration platform Kubernetes. Designing appropriate container auto-scaling methods for s...More and more applications are organized in the form of meshed micro-services which can be deployed on the popularcontainer orchestration platform Kubernetes. Designing appropriate container auto-scaling methods for such applicationsin Kubernetes is beneficial to reducing costs and guaranteeing Quality of Services (QoS). However, most existing resourceprovisioning methods focus on a service without considering interactions among meshed services. Meanwhile, synchronouscalls among services have different impacts on the processing ability of containers as the proportion of different businesstype’s requests changes which is not considered in existing methods too. Therefore, in this article, an adaptive queuing modeland queuing-length aware Jackson queuing network based method is proposed. It adjusts the processing rate of containersaccording to the ratio of synchronous calls and considers queuing tasks when calculating the impact of bottleneck tiers toothers. Experiments are performed on a real Kubernetes cluster, which illustrate that the proposal obtains the lowest percentageof Service Level Agreement (SLA)-violations (decreasing about 6.33%-12.29%) with about 0.9% additional costscompared with existing methods of Kubernetes and other latest methods.展开更多
基金supported by the National Natural Science Foundation of China(Grant No.61972202,61872186,61973161,61991404)the Fundamental Research Funds for the Central Universities(No.30919011235).
文摘More and more applications are organized in the form of meshed micro-services which can be deployed on the popularcontainer orchestration platform Kubernetes. Designing appropriate container auto-scaling methods for such applicationsin Kubernetes is beneficial to reducing costs and guaranteeing Quality of Services (QoS). However, most existing resourceprovisioning methods focus on a service without considering interactions among meshed services. Meanwhile, synchronouscalls among services have different impacts on the processing ability of containers as the proportion of different businesstype’s requests changes which is not considered in existing methods too. Therefore, in this article, an adaptive queuing modeland queuing-length aware Jackson queuing network based method is proposed. It adjusts the processing rate of containersaccording to the ratio of synchronous calls and considers queuing tasks when calculating the impact of bottleneck tiers toothers. Experiments are performed on a real Kubernetes cluster, which illustrate that the proposal obtains the lowest percentageof Service Level Agreement (SLA)-violations (decreasing about 6.33%-12.29%) with about 0.9% additional costscompared with existing methods of Kubernetes and other latest methods.
