On Simulation and Design of Parallel-Systems Schedulers: Are We Doing the Right Thing?
- 22 August 2008
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 20 (7) , 983-996
Abstract
It is customary to use open-system trace-driven simulations to evaluate the performance of parallel-system schedulers. As a consequence, all schedulers have evolved to optimize the packing of jobs in the schedule, as a means to improve a number of performance metrics that are conjectured to be correlated with user satisfaction, with the premise that this will result in a higher productivity in reality. We argue that these simulations suffer from severe limitations that lead to suboptimal scheduler designs and to even dismissing potentially good design alternatives. We propose an alternative simulation methodology called site-level simulation, in which the workload for the evaluation is generated dynamically by user models that interact with the system. We present a novel scheduler called CREASY that exploits knowledge on user behavior to directly improve user satisfaction and compare its performance to the original packing-based EASY scheduler. We show that user productivity improves by up to 50 percent under the user-aware design, while according to the conventional metrics, performance may actually degrade.Keywords
This publication has 24 references indexed in Scilit:
- On the User–Scheduler Dialogue: Studies of User-Provided Runtime Estimates and Utility FunctionsThe International Journal of High Performance Computing Applications, 2006
- Backfilling with lookahead to optimize the packing of parallel jobsJournal of Parallel and Distributed Computing, 2005
- Unfairness Metrics for Space-Sharing Parallel Job Schedulers2005
- The workload on parallel supercomputers: modeling the characteristics of rigid jobsJournal of Parallel and Distributed Computing, 2003
- The Impact of More Accurate Requested Runtimes on Production Job Scheduling Performance2002
- Selective Reservation Strategies for Backfill Job Scheduling2002
- Multiple-queue backfilling scheduling with priorities and reservations for parallel systemsACM SIGMETRICS Performance Evaluation Review, 2002
- Utilization, predictability, workloads, and user runtime estimates in scheduling the IBM SP2 with backfillingIEEE Transactions on Parallel and Distributed Systems, 2001
- Characterizing Web user sessionsACM SIGMETRICS Performance Evaluation Review, 2000
- Packing schemes for gang scheduling1996