Information -Aware Scheduling Strategies for Desktop Grid Environment
In this paper, the authors will show how it is possible to build Information-aware schedulers able to outperform The Work Queue with Replication-Fault Tolerant Scheduler(WQR-FT). They propose different scheduling policies considering information about resources and applications. They will discuss two task selection policies and four machine selection policies that when combined give rise to 8 different scheduling algorithms. As a matter of fact, the results obtained shows that it is possible to achieve better performance than WQR-FT and reduce the wasted CPU cycles.