IEEE Transactions on Parallel and Distributed Systems
PROC: Process ReOrdering-Based Coscheduling on Workstation Clusters
IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Papers - Volume 01
On the Scalability of Centralized Control
IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Workshop 18 - Volume 19
Token-ordered LRU: an effective page replacement policy and its implementation in Linux systems
Performance Evaluation - Performance modelling and evaluation of high-performance parallel and distributed systems
Another approach to backfilled jobs: applying virtual malleability to expired windows
Proceedings of the 19th annual international conference on Supercomputing
Adaptive Parallel Job Scheduling with Flexible Coscheduling
IEEE Transactions on Parallel and Distributed Systems
LOMARC: Lookahead Matchmaking for Multiresource Coscheduling on Hyperthreaded CPUs
IEEE Transactions on Parallel and Distributed Systems
A runtime resolution scheme for priority boost conflict in implicit coscheduling
The Journal of Supercomputing
Adaptive time/space sharing with SCOJO
International Journal of High Performance Computing and Networking
Cooperating coscheduling: a coscheduling proposal aimed at non-dedicated heterongeneous NOWs
Journal of Computer Science and Technology
Moldable parallel job scheduling using job efficiency: an iterative approach
JSSPP'06 Proceedings of the 12th international conference on Job scheduling strategies for parallel processing
The Impact of noise on the scaling of collectives: the nearest neighbor model
HiPC'07 Proceedings of the 14th international conference on High performance computing
Dynamic load balancing in MPI jobs
ISHPC'05/ALPS'06 Proceedings of the 6th international symposium on high-performance computing and 1st international conference on Advanced low power systems
New challenges of parallel job scheduling
JSSPP'07 Proceedings of the 13th international conference on Job scheduling strategies for parallel processing
Xen-OSCAR for cluster virtualization
ISPA'06 Proceedings of the 2006 international conference on Frontiers of High Performance Computing and Networking
The impact of noise on the scaling of collectives: a theoretical approach
HiPC'05 Proceedings of the 12th international conference on High Performance Computing
Parallel job scheduling — a status report
JSSPP'04 Proceedings of the 10th international conference on Job Scheduling Strategies for Parallel Processing
Reconfigurable gang scheduling algorithm
JSSPP'04 Proceedings of the 10th international conference on Job Scheduling Strategies for Parallel Processing
LOMARC — lookahead matchmaking for multi-resource coscheduling
JSSPP'04 Proceedings of the 10th international conference on Job Scheduling Strategies for Parallel Processing
ScoPred–scalable user-directed performance prediction using complexity modeling and historical data
JSSPP'05 Proceedings of the 11th international conference on Job Scheduling Strategies for Parallel Processing
AnthillSched: a scheduling strategy for irregular and iterative I/O-intensive parallel jobs
JSSPP'05 Proceedings of the 11th international conference on Job Scheduling Strategies for Parallel Processing
Multi-domain job coscheduling for leadership computing systems
The Journal of Supercomputing
Hi-index | 0.00 |
Fine-grained parallel applications require all their processes to run simultaneously on distinct processors to achieve good efficiency. This is typically accomplished by space slicing, wherein nodes are dedicated for the duration of the run, or by gang scheduling, wherein time slicing is coordinated across processors. Both schemes suffer from fragmentation, whereprocessors are left idle because jobs cannot be packed with perfect efficiency. Obviously, this leads to reduced utilization and sub-optimal performance. Flexible coscheduling (FCS) solves this problem by monitoring each job's granularity and communication activity, and using gang scheduling only for those jobs that require it. Processes from other jobs, which can be scheduled without any constraints, are used as filler to reduce fragmentation. In addition, inefficiencies due to load imbalance and hardware heterogeneity are also reduced because the classification is done on a per-process basis. FCS has been fully implemented as part of the STORM resource manager, and shown to be competitive with gang scheduling and implicit coscheduling.