An Opportunity Cost Approach for Job Assignment in a Scalable Computing Cluster
IEEE Transactions on Parallel and Distributed Systems
Scheduling Divisible Loads in Parallel and Distributed Systems
Scheduling Divisible Loads in Parallel and Distributed Systems
The ANL/IBM SP Scheduling System
IPPS '95 Proceedings of the Workshop on Job Scheduling Strategies for Parallel Processing
Pfair scheduling: beyond periodic task systems
RTCSA '00 Proceedings of the Seventh International Conference on Real-Time Systems and Applications
On-line scheduling of scalable real-time tasks on multiprocessor systems
Journal of Parallel and Distributed Computing
A taxonomy of market-based resource management systems for utility-driven cluster computing
Software—Practice & Experience
Real-Time Divisible Load Scheduling for Cluster Computing
RTAS '07 Proceedings of the 13th IEEE Real Time and Embedded Technology and Applications Symposium
Adaptive job scheduling via predictive job resource allocation
JSSPP'06 Proceedings of the 12th international conference on Job scheduling strategies for parallel processing
Feedback scheduling of real-time divisible loads in clusters
ACM SIGBED Review - Special issue on the the 14th IEEE real-time and embedded technology and applications symposium (RTAS'08) WIP session
Multi-round real-time divisible load scheduling for clusters
HiPC'08 Proceedings of the 15th international conference on High performance computing
Scheduling real-time divisible loads with advance reservations
Real-Time Systems
Hi-index | 0.00 |
Providing QoS and performance guarantees for arbitrarily divisible loads in a cluster has become a significant problem. While progress is being made in scheduling arbitrarily divisible loads, some of the proposed approaches may cause Inserted Idle Times (IITs) that are detrimental to system performance. Two contributions are made in addressing this problem. First, we propose two constraints that, when satisfied, lead to an optimal partitioning in utilizing IITs. Second, we integrate the new partitioning method with a previous approach and develop an enhanced algorithm that better utilizes IITs. Simulation results demonstrate the advantages of our new approach.