A novel hybrid of Shortest job first and round Robin with dynamic variable quantum time task scheduling technique

Abstract

Cloud computing is a ubiquitous network access model to a shared pool of configurable computing resources where available resources must be checked and scheduled using an efficient task scheduler to be assigned to clients. Most of the existing task schedulers, did not achieve the required standards and requirements as some of them only concentrated on waiting time or response time reduction or even both neglecting the starved processes at all. In this paper, we propose a novel hybrid task scheduling algorithm named (SRDQ) combining Shortest-JobFirst (SJF) and Round Robin (RR) schedulers considering a dynamic variable task quantum. The proposed algorithms mainly relies on two basic keys the first having a dynamic task quantum to balance waiting time between short and long tasks while the second involves splitting the ready queue into two sub-queues, Q1 for the short tasks and the other for the long ones. Assigning tasks to resources from Q 1 or Q 2 are done mutually two tasks from Q 1 and one task from Q 2 . For evaluation purpose, three different datasets were utilized during the algorithm simulation conducted using CloudSim environment toolkit 3.0.3 against three different scheduling algorithms SJF, RR and Time Slice Priority Based RR (TSPBRR) Experimentations results and tests indicated the superiority of the proposed algorithm over the state of art in reducing waiting time, response time and partially the starvation of long tasks.

864 Published By: Blue Eyes Intelligence Engineering & Sciences Publication Retrieval Number: E6080018520 /2020©BEIESP DOI:10.35940/ijrte.E6080.018520 Journal Website: www.ijrte.org Multi Level Queue Scheduling With Particle Swarm Optimization (Mlqs-Pso) Of Vms in Queueing Heterogeneous Cloud Computing Systems

“…Elmougy, et al [16] developed a new hybrid scheduling depends on task is represented as (SRDQ) integrating Round Robin (RR) and Shortest-Job-First (SJF) schedulers with vibrant unpredictable quantity job. The algorithms suggested relying generally on two fundamental keys the first, to have a complex task of balancing the waiting period between tasks with small sized tasks as well as lengthy tasks Whereas lengthy tasks includes forming two sub-queues from the ready queue, Q1 representing short and Q1 representing lengthy task.…”

“…Next in step 6, (14) and (15) are computed and achieve Hβ=T' (16) layer of hidden and T' for output transposition. Using the least square method of measuring the weight matrix values of the minimum error to achieve a unique solution [15,16].…”

Novel Approach to Task Scheduling and Load Balancing Using the Dominant Sequence Clustering and Mean Shift Clustering Algorithms

“…The combination in both versions of SJF and RR is evaluated with dynamic and static quanta, and optimality in the task scheduling and load balancing methods is achieved via evaluation. Second, the ready queue is divided into two sub-queues, in which Q1 denotes short tasks, whereas Q2 denotes long tasks [4].…”

“…Task rescheduling increases computational time and complexity. In [4], a hybrid of SJF and RR schedulers that considers only dynamic variable quantum tasks is adopted to complete task scheduling. The ready queue is divided into Q1 and Q2, where Q1 is for storing short tasks, whereas Q2 queue is for storing long tasks.…”

“…Slow convergence time and poor local search capability caused by the PSO algorithm are determined. In the current research, we use efficient task scheduling and load balancing algorithms to solve the problems highlighted in [1,2,4,9].…”