serverless edge computing environments use lightweight containers to run different services on a need basis. Container caching at edge nodes is an effective strategy to further reduce the startup latency related to th...
详细信息
ISBN:
(纸本)9798350310900
serverless edge computing environments use lightweight containers to run different services on a need basis. Container caching at edge nodes is an effective strategy to further reduce the startup latency related to the preparation of container images. However, the capacity limitation of the edge nodes requires an efficient caching strategy that can capture underlying service request patterns. Hence, this paper proposes an EXplainable Reinforcement Learning (XRL)-based container caching strategy to increase the hit rate of cached containers. While a few studies already proposed RL-based caching algorithms, our proposal focuses more on the explainability part of the caching decisions based on a causal graph. The generated explanations from our approach can indicate which caching actions specifically contribute to the increase in the hit rate, which implies the underlying request patterns. Our experiments in a simple network topology demonstrate the validity of the generated explanations.
As the edge cloud technology evolves, research is underway to provide customers with a serverless architecture-based FaaS (Function as a Service) on the edge nodes. However, serverless architecture-based FaaS on the e...
详细信息
ISBN:
(纸本)9781728147840
As the edge cloud technology evolves, research is underway to provide customers with a serverless architecture-based FaaS (Function as a Service) on the edge nodes. However, serverless architecture-based FaaS on the edge nodes has a problem with resource management due to edge nodes with relatively lower computing power than the cloud datacenters. And also it has a late service provision problem due to time to prepare cloud instances for customer API call requests. In this paper, we propose the method of dynamic container layer replacement considering a resource-limited environment on the edge nodes while mitigating cold start and then shows the performance.
暂无评论