An Approach of Leading Sequence Clustering (LSC) Algorithm based Scheduling and Agglomerative Mean Shift Clustering for Load Balancing in Cloud

Ashish Sharma and Anjani Rai

In cloud environment, the Quality of Service can be increased or decreased dramatically by load balancing and task scheduling techniques. The disadvantages of existing methods are resolved by using a proposed efficient scheduling and load balancing based mechanism. In this work, load balancing is done by Leading Sequence Clustering (LSC) algorithm. Trapping into local minima is avoided by using this algorithm and for scheduling problems also it can be utilized. The LSC algorithm clusters the user’s task and represents it in terms of a graph with one or more clusters. Predict Earliest Finish Time (PEFT) scheduling is combined with Genetic Algorithm (GA) to rank the tasks. Based on the processing cost and execution time, the task with high priority is first scheduled. Agglomerative Mean Shift clustering (AMSC) is used with fuzzy functions to cluster the virtual machines (VM). Weighted least connection (WLC) algorithm is sued at last to perform load balancing. The server’s weight, capacity and connectivity to server, loads are distributed by this algorithm. The response time is increased by selecting the server with high weight or least connectivity to allocate a task. Different cloud parameters are used to show the efficiency of proposed method in experimentation.

Volume 11 | 10-Special Issue

Pages: 618-624

DOI: 10.5373/JARDCS/V11SP10/20192850