Output details
11 - Computer Science and Informatics
University of Derby
Exploring decentralized dynamic scheduling for grids and clouds using the community-aware scheduling algorithm
<7>
This work led to a PhD thesis (Dr. Ye, 2011) and was carried out in the context of the SmartGrid project (Swiss Hasler Foundation and Marie Curie Knowledge Transfer Programme). MaSim, a new MaGate simulator supporting real grid workload trace archive, simplified dynamic scheduling algorithm, and graphically intuitive tracing system was released in 2010. It is been used from the CoreGrid WP6 community. The produced algorithm is a follow-up of two best paper awards (INTERNET-2009; FGIT/GDC-2009). An evaluation with a real workload trace demonstrates a 30%–61% better average job slowdown, and a 68%–86% shorter average job waiting time.