Performance analysis of optimal scheduling for healthcare workflows using hybrid approaches in cloud environment

Abstract

Cloud Computing is an effective platform for storing and maintaining a large number of data in remote manner. More number of tasks are operated in the user system environments with low hardware specifications and they cannot be updated frequently due to its high cost.The enormous amount of information retained in the hard disk of the system consumes more storage space which affects the processing capability and performance of the system. To overcome such limitations in an existing system, the tasks for different applications can be executed in a remote cloud server with higher-end hardware specifications. Cloud computing aims to store information in a remote environment where users can access it anytime, anywhere via the internet. The main limitation of the conventional system is that it consumes high Makespan (MS) and Execution Time (ET) with different loads, hence it requires computing and allocating priority for each executing task and based on the higher priority, the particular task is executed in the cloud system. The novelty of the research work mainly combines the optimization algorithm with Machine Learning (ML) approach for workflow scheduling that achieves minimum MS and ET compared to various state-of-art methods. The main objective of this research work is to propose an effective workflow scheduling methodology using ML, Deep Learning (DL) and evolutionary approaches in healthcare applications. newlineThe research work develops an efficient Optimal Workflow Scheduling (OWS) for cloud systems that integrates the ML method Adaptive Neuro-Fuzzy Inference System (ANFIS) and Genetic Algorithm (GA). The Amazon Elastic Compute Cloud (Amazon EC2) is used as a cloud platform for evaluating the performance of the proposed workflow scheduling. newline