Anil Kumar’s research while affiliated with Guru Nanak Dev University and other places

In cloud computing technology, task scheduling is one of the research challenges. For these various algorithms, works such as particle swarm optimization (PSO), firefly algorithm, ant colony optimization (ACO) and genetic algorithm (GA). PSO is inspired by the bird’s movement, and ACO is based on the behaviour of ants. GA works based on the natural evolution process. This paper presents the hybrid of PSO-ACO-GA for task scheduling on virtual machines of cloud computing known as ant particle swarm genetic algorithm (APSGA). Here, GA and PSO will perform iteration to get the task basis on fitness value and further ACO will distribute the task on specific virtual machines. This paper has achieved improved results for parameters such as CPU utilization, makespan and execution time. Our proposed algorithm has achieved makespan that is reduced by 27.1%, 19.45% and 21.24% with compare to PSO, ACO and GA, respectively. It has achieved maximum of CPU utilization and execution time.

Cloud computing (CC) is a pool of services which includes services in means of infrastructure, memory, applications, platform, storage and many more. Cloud provides service as per requests, at a given time numerous users might be working to get the resources provided by the service provider. The main issue here for the service provider is how to handle these requests, so they have to do it in such a way that the quality promised in not compromised. For this purpose, we need to choose such a scheduling mechanism that could efficiently perform the task of mapping resources from that populous pool of resource. Keeping in account various metrics such as execution time and makespan that effect the parameter of task scheduling like response time, load balancing, completion time, etc. In this paper, we will be having a systematic review of various task scheduling algorithm choosing different metrics to work on.