A high performance machine learning algorithm TspINA; scheduling multifariousness destined tasks by better efficiency
Shantanu Anandrao Lohi, Nandita Tiwari · 2020 Fourth World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4) · 2020
The ability to successfully execute task scheduling for any computing system is a performance measurement in any multi-processing environment, such as working of cloud or continuous scheduling and much more. Over the years, for many cloud deployments, cost-effective resource scheduling (CERS) algorithms and dynamic and integrated resource scheduling (DAIRS) algorithms were commonly used. But these algorithms have certain drawbacks in terms of retention and efficient task response time and setting threshold values along with causing lot of overhead. Through this paper we try to put forward an algorithm based on machine learning (ML) paradigm, Task scheduling process Improved Novel Algorithm (TspINA). Our proposed algorithm explores the methods used by CERS and DAIRS algorithms and provides a reasonable enhancement by integrating machine learning features techniques for prior-learning and persistent adaptation to minimize the completion time for a defined series of objectives and tasks. Our suggested methodology would be further compared to the existing CERS and DAIRS algorithms by using generic datasets and the outcomes will be correlated with the amount of machine effort, delay and responsiveness.