Scheduling Tightly-Coupled Applications on Heterogeneous Desktop Grids
Platforms that comprise volatile processors, such as desktop grids, have been traditionally used for executing independent-task applications. In this paper, the authors examine the scheduling of tightly-coupled iterative master-worker applications onto volatile processors. The main challenge is that workers must be simultaneously available for the application to make progress. They consider three additional complications: one should take into account that workers can become temporarily reclaimed and, for data-intensive applications; one should account for the limited bandwidth between the master and the workers; and workers are strongly heterogeneous, with different computing speeds and availability probability distributions.