Study of Scheduling Strategies in a Dynamic Data Grid Environment
Data grids seek to harness geographically distributed resources for large-scale data-intensive problems. Such problems involve loosely coupled jobs and large data sets mostly distributed geographically. Data grids have found applications in scientific research, in the field of high-energy Physics, Life Sciences etc. The issues that need to be considered in the data grid research area include: resource management including computation management and data management. Computation management include scheduling of jobs, scalability, response time involved in such scheduling, while data management include data replication in selected sited, data movement when required. Therefore, scheduling and replication assumes great importance in a data grid environment. In this paper, we have developed several scheduling strategies based on a developed replication strategy. The scheduling strategies are called Matching based Scheduling (MJS), Cost base Scheduling (CJS) and Latency based Scheduling (LJS). Among these, LJS and CJS perform similarly and MJS performs worse than both of them.
KeywordsSchedule Strategy Data Grid Queue Size Replication Strategy Average Queue Size
Unable to display preview. Download preview PDF.
- 2.Foster, Kesselman, C.: The Globus Project: A Status Report. In: Proc. IPPS/SPDP 1998 Heterogeneous Computing Workshop, pp. 4–18 (1998)Google Scholar
- 3.Casanova, H., Obertelli, G., Berman, F., Wolski, R.: The AppLeS Parameter Sweep Template: User-Level Middleware for the Grid. In: Proceedings of SuperComputing 2000 (2000)Google Scholar
- 4.Alhusaini, A.H., Prasanna, V.K., Raghavendra, C.S.: A Unified Resource Scheduling Framework for Heterogeneous Computing Environments. In: Eighth Heterogeneous Computing Workshop (1999)Google Scholar