< Back to previous page


Economic and network aware grid resource management.

The introduction of economic principles in Grid computing, Grid Economics, has two separate meanings. On the one hand the use of economically inspired principles to develop effective forms of resource management, and on the other hand to enable supplying resources and services as an economic activity.The fact that a substantial number of Grid applications is very data intensive has led to the research and propositions of scheduling algorithms that take into account the effects of data transport, so called network aware scheduling. An approach like this can both increase efficiency of computational as well as network resources and decrease responsetimes for jobs.In this project, we aim to combine Grid economics and network aware scheduling. The objective is the design of algorithms and protocols that allow the co-allocation of network and computational resources when using a grid resource management system based on economic markets. With regards to co-allocation, already a lot of work has been done. On a more limited scale research is being done on the use of market mechanisms for the allocation of network paths. The combination of a market and co-allocation of network and computational resources has not been explored before however. This combination leads to new possibilities when creating allocations and schedules. It creates a more relevant description of the value a user associates with a certain allocation. Because of this, economic and network aware scheduling will be a unique contribution to the domain of Grid economics. This project also fits nicely into the current research being done in the CoMP group into different market mechanisms and their applicability in economic scheduling of computational resources in grids.
Date:1 Oct 2010  →  30 Sep 2012
Disciplines:Applied mathematics in specific fields, Computer architecture and networks, Distributed computing, Information sciences, Information systems, Programming languages, Scientific computing, Theoretical computer science, Visual computing, Other information and computing sciences