Resource Allocation for Steerable Parallel Parameter Searches

Computational Grids lend themselves well to parameter sweep applications, in which independent tasks calculate results for points in a parameter space. It is possible for a parameter space to become so large as to pose prohibitive system requirements. In these cases, user-directed steering promises to reduce overall computation time. In this paper, we address an interesting challenge posed by these user-directed searches: how should compute resources be allocated to application tasks as the overall computation is being steered by the user? We present a model for user-directed searches, and then propose a number of resource allocation strategies and evaluate them in simulation. We find that prioritizing the assignments of tasks to compute resources throughout the search can lead to substantial performance improvements.