Task grouping and optimized deep learning based VM sizing for hosting containers as a service

Abstract Containers as a service (CaaS) are kind of services that permits the organization to handle containers more effectively. lightweight, require less computing resources, portable, and facilitate better support for microservices. In CaaS model, deployed in virtual machines, machine runs on physical machine. The objective this paper is estimate resource required by VM execute number containers. sizing directorial process where system administrators have optimize allocated resources within permitted virtualized space. work, carried out using Deep Convolutional Long Short Term Memory Network (Deep-ConvLSTM), weights updated Fractional Pelican Optimization (FPO) Algorithm. Here, FPO configured hybridizing concept Calculus (FC) location Algorithm (POA). Moreover, task grouping done with Embedded Clustering (DEC), established respect various parameters, such length, submission rate, scheduling class, priority, usage, latency, Task Rejection Rate (TRR). addition, performance analysis taking 100, 200, 300, 400 tasks. We got best utilization 0.104 300 tasks, response time 262ms 100 TRR 0.156 tasks makespan 0.5788