An Evaluation of Job Scheduling Strategies for Divisible Loads on Grid Platforms
نویسندگان
چکیده
In this paper we study distributed job scheduling in grid environments when each job is a DL application. The scheduling goal is to minimize the average steady-state job turnaround time. In this context, we identify in which regimes classes of scheduling strategies are efficient, namely for which platforms and which communication to computation ratios. We also quantify what level of global information about the platform is required for efficient scheduling. All our findings are obtained via simulation of wide ranges of application and platform scenarios. Our most significant findings are that the use of grid information is only necessary at high workload, and that at high workload using dynamic information improves performance by around 10% when compared to using static information.
منابع مشابه
Grid Scheduling Divisible Loads from Multiple Sources via Linear Programming
To date solutions for optimal finish time and job allocation in divisible load theory are largely obtained only for network topologies with a single load originating (root) processor. However in large-scale data intensive problems with geographically distributed resources, load is generated from multiple sources. This paper introduces a new divisible load scheduling strategy for tree networks w...
متن کاملDeadline-Based Scheduling Algorithm for Divisible-Load in Grid Computing
The divisible load model is motivated by divisible load theory, where both communication and computation can be arbitrarily divisible into as many independent partitions as required and facilitates a good approximation for many real-world application systems such as those arising in large physics experiments. Scheduling an application with divisible load in data grid is significantly important ...
متن کاملAn Efficient Hybrid Job Scheduling Algorithm for Computational Grids
Scheduling of jobs is a challenging problem in grid. Efficient job scheduling is essential for the effective utilization of the resources. We propose a grid model as a collection of clusters. In this paper, we apply Divisible Load Theory (DLT) and Least Cost Method (LCM) to model the grid scheduling problem involving multiple worker nodes in each cluster. We propose a hybrid job scheduling algo...
متن کاملA New Job Scheduling in Data Grid Environment Based on Data and Computational Resource Availability
Data Grid is an infrastructure that controls huge amount of data files, and provides intensive computational resources across geographically distributed collaboration. The heterogeneity and geographic dispersion of grid resources and applications place some complex problems such as job scheduling. Most existing scheduling algorithms in Grids only focus on one kind of Grid jobs which can be data...
متن کاملAn Effective Approach to Job Scheduling in Decentralized Grid Environment
Scheduling of jobs and resource management are the important challenging work in a grid environment. Processing time minimization of the jobs arriving at any computer site in a grid system is one of the major objectives in the research area of computing. In this paper, we propose a decentralized grid system model as a collection of clusters. We then introduce a decentralized job scheduling algo...
متن کامل