An Alternative Model for Scheduling on a Computational Grid

Abstract In this paper we discuss scheduling on a computational grid ofau­tonomous nodes. A twolevel process involyes a meta-schedulerwhoseobjective 1s to generate an optimal schedule and reduce the total ex­ecution time of an work-flow,and local schedulers whose objective isto optimize the utilization of local resources. We introduce severalresource allocation and consumption models and discuss issues perti­nent to decision makingwithincomplete and/oroutdated information.Then we present a stock market model for scheduliug on a computergrid with autonomous nodes. 1 Introduction A computationalgrid is a large-scale,heterogeneous collection ofautonomoussystems, geographically distributed and interconnected by low latency andhigh bandwidth networks. The mission of a grid is to provide dependableservices at a low cost for a large community of users and to support collabo­rative work. The term l'computationalgrid" is based on the analogy with the"power grid", it reflects the desire to view the network as a pool of resourcesand to provide a commodity very much like the electric power and it does

[1]  K. Mani Chandy,et al.  A world-wide distributed system using Java and the Internet , 1996, Proceedings of 5th IEEE International Symposium on High Performance Distributed Computing.

[2]  Thomas M. Cover,et al.  Elements of Information Theory , 2005 .

[3]  Andrew S. Grimshaw,et al.  Legion-a view from 50,000 feet , 1996, Proceedings of 5th IEEE International Symposium on High Performance Distributed Computing.

[4]  Ami Marowka,et al.  The GRID: Blueprint for a New Computing Infrastructure , 2000, Parallel Distributed Comput. Pract..

[5]  Ladislau Bölöni,et al.  Seamless Metacomputing with Bond , 1998 .

[6]  Dan C. Marinescu,et al.  Models and Algorithms for Coscheduling Compute-Intensive Tasks on a Network of Workstations , 1992, J. Parallel Distributed Comput..

[7]  Jean-Charles Fabre,et al.  A Metaobject Architecture for Fault-Tolerant Distributed Systems: The FRIENDS Approach , 1998, IEEE Trans. Computers.

[8]  Octavian F. Carbunar,et al.  Dynamic scheduling of process groups , 1998, Concurr. Pract. Exp..

[9]  Ian T. Foster,et al.  The Globus project: a status report , 1998, Proceedings Seventh Heterogeneous Computing Workshop (HCW'98).

[10]  P. Wilmott,et al.  The Mathematics of Financial Derivatives: Contents , 1995 .

[11]  Dan C. Marinescu,et al.  A scheduling expert advisor for heterogeneous environments , 1997, Proceedings Sixth Heterogeneous Computing Workshop (HCW'97).