O.S. Abdul Qadir et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(3), May – June 2020, 2527 – 2534 2527 ABSTRACT Cloud computing takes into account to permit the sharing of resources like networks, servers, storage, applications, and services to achieve access of these resources from any computer in the world through the Internet. The task scheduling algorithms are of NP-hard in nature. The importance of task scheduling is to map the tasks with the appropriate resources for execution. Makespan is the difference in time from start to end of the scheduling. Load balancing is the sharing of workload among the resources. The objective of any scheduling algorithm is to reduce the makespan with proper utilization of resources. In this paper a new algorithm (DOTS) is developed with the realistic dual-objective criteria that minimize the makespan and balance the load across the resources. The distribution of tasks among the resources is also evaluated using the coefficient of variation. The dual objectives are transformed into a single score using weighted sum method. The results clearly indicate that the proposed DOTS technique performs well when compared with seven other scheduling algorithms. Key words: Cloud Computing, Task Scheduling, Makespan, Load balancing, ETC Matrix and Coefficient of Variation. 1. INTRODUCTION 1.1 Cloud Computing Cloud computing is a model that share the resources like servers, storage, applications and services in the Internet to provide various on demand services. Cloud computing is one of the fastest growing technology and is applied in everywhere business operations. The services of cloud computing are usually classified as Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS). 1.2 Task Scheduling Tasks are a form of incoming request. Task Scheduling is the concept of allocating the tasks to the resources based on some principles. The length of scheduling is the time taken to finish all the tasks. Similarly, the schedule length of a particular resource is the time taken to execute all the tasks assigned to it. Different resources will have different schedule length depending on the scheduling techniques used. 1.3 Makespan Any scheduling algorithm will try to execute all the tasks in the shortest possible time by appropriately mapping with the available resources. Hence every resource will be allocated with a specific set of tasks based on the scheduling policy. The length of execution of a particular resource is the total execution of all the tasks allocated to it. The maximum value in the set of the total execution time of every resource is termed as the makespan. In other words, makespan is the length of the schedule of a resource having the maximum total execution time. 1.4 Load balancing Load balancing is the process of reassigning the total load of the entire system to individual resources in order to have an equal or near to equal utilization of all the resources. This will lead to the reduction of the response time of the tasks. Load balancing removes the situation where some resources are overloaded in comparison with some other resources which are underloaded. The main goal of load balancing is the effective utilization of resources there by reducing the overall execution time (makespan). 1.5 ETC Matrix The expected time to compute (ETC) matrix model defines the specification of the execution time of all the tasks across the available resources. For example, a value of [3,4] in the ETC matrix represents the execution time for the 3rd task on the 4th resource. In general, a value of [i, j] in the matrix Dual Objective Task Scheduling Algorithm in Cloud Environment O.S. Abdul Qadir 1 , Dr. G. Ravi 2 1 Research Scholar in Computer Science, Jamal Mohamed College (Autonomous) (Affiliated to Bharathidasan University) Tiruchirappalli, Tamil Nadu, India abdulqadir@jmc.edu 2 Associate Professor & Head, PG and Research Department of Computer Science Jamal Mohamed College (Autonomous), (Affiliated to Bharathidasan University), Tiruchirappalli, Tamil Nadu, India ravi_govindaraman@yahoo.com ISSN 2278-3091 Volume 9, No.3, May - June 2020 International Journal of Advanced Trends in Computer Science and Engineering Available Online at http://www.warse.org/IJATCSE/static/pdf/file/ijatcse07932020.pdf https://doi.org/10.30534/ijatcse/2020/07932020