162
click to vote
JSSPP
15 years 11 months ago
2004 Springer 
As we continue to evolve into large-scale parallel systems, many of them employing hundreds of computing engines to take on mission-critical roles, it is crucial to design those s...
187
click to vote
JSSPP
15 years 11 months ago
2004 Springer 
The self-tuning dynP scheduler for modern cluster resource management systems switches between different basic scheduling policies dynamically during run time. This allows to reac...
158
Voted
JSSPP
15 years 11 months ago
2004 Springer 
Computer system batch schedulers typically require information from the user upon job submission, including a runtime estimate. Inaccuracy of these runtime estimates, relative to ...
146
click to vote
JSSPP
15 years 11 months ago
2004 Springer 
Using a single traditional gang scheduling algorithm cannot provide the best performance for all workloads and parallel architectures. A solution for this problem is the use of...
145
click to vote
JSSPP
15 years 11 months ago
2004 Springer 
This paper presents a comprehensive characterization of a multi-cluster supercomputer1 workload using twelvemonth scientific research traces. Metrics that we characterize include...
|
Distributed And Parallel Computing