Grid computing platforms and components are subject to a great variability. For instance, CPU speeds and link bandwidths are expected to change over time, as the execution proceeds, and this evolution is likely to take the form of abrupt changes followed by epochs of relative stability. Also, processors and network links may fail and then be repaired during execution. Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous, and subject to modifications and failures. Accounting for the dynamic nature of platform parameters is a key component for the design of robust scheduling algorithms. This workshop aims at bringing together researchers working in the areas of scheduling, performance evaluation and fault tolerance to share their experiences and latest research results on these hot topics. We welcome submission of original, unpublished papers, as well as work-in-progress contributions, related to these research areas. Topics Topics of interest for the workshop include but are not limited to: * Scheduling algorithms for heterogeneous platforms * Performance models * Models of platform/application failures * Fault tolerance issues * Resource discovery and management * Task and communication scheduling * Task coordination and workflow * Job scheduling * Stochastic scheduling * Scheduling applications for clusters and grids
RobSched’08: International Workshop on Robust Scheduling – part of ICPADS’08 (Int. Conf. on Parallel and Distributed Systems), Melbourne, Australia.
ALDINUCCI, MARCO
2008-01-01
Abstract
Grid computing platforms and components are subject to a great variability. For instance, CPU speeds and link bandwidths are expected to change over time, as the execution proceeds, and this evolution is likely to take the form of abrupt changes followed by epochs of relative stability. Also, processors and network links may fail and then be repaired during execution. Mapping applications onto parallel platforms is a challenging problem, that becomes even more difficult when platforms are heterogeneous, and subject to modifications and failures. Accounting for the dynamic nature of platform parameters is a key component for the design of robust scheduling algorithms. This workshop aims at bringing together researchers working in the areas of scheduling, performance evaluation and fault tolerance to share their experiences and latest research results on these hot topics. We welcome submission of original, unpublished papers, as well as work-in-progress contributions, related to these research areas. Topics Topics of interest for the workshop include but are not limited to: * Scheduling algorithms for heterogeneous platforms * Performance models * Models of platform/application failures * Fault tolerance issues * Resource discovery and management * Task and communication scheduling * Task coordination and workflow * Job scheduling * Stochastic scheduling * Scheduling applications for clusters and grids
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
