The KOALA grid scheduler

Omer Ozan Sonmez, Technische Universiteit Delft, The Netherlands

KOALA is a grid scheduler that has been designed, implemented, and deployed by the PDS group in TUDelft on the DAS2 multicluster system in the context of the Virtual Lab for e-science (VL-e) research project. The main feature of KOALA is its support for co-allocation, that is, the simultaneous allocation of resources in multiple clusters of the DAS to single applications which consist of multiple components. Currently, KOALA supports processor and data co-allocation in that it starts the components of a single application on different clusters at the same time, and it transfers the input files of the components to their proper locations prior to application execution. KOALA uses some of the components of the Globus toolkit, and interfaces on the DAS to the local SGE schedulers. Currently, we are working on adapting KOALA for Grid'5000. We target to conduct research on efficient scheduling policies that would dispatch jobs both to DAS and Grid'5000 testbeds.