RESOURCE ALLOCATION STRATEGIES FOR GUIDED PARAMETER SPACE SEARCHES

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