Advanced Reservation-Based Scheduling of Task Graphs on Clusters

A Task Graph (TG) is a model of a parallel program that consists of many subtasks that can be executed simultaneously on different processing elements. Subtasks exchange data via an interconnection network. The dependencies between subtasks are described by means of a Directed Acyclic Graph. Unfortunately, due to their characteristics, scheduling a TG requires dedicated or uninterruptible resources. Moreover, scheduling a TG by itself results in a low resource utilization because of the dependencies among the subtasks. Therefore, in order to solve the above problems, we propose a scheduling approach for TGs by using advance reservation in a cluster environment. In addition, to improve resource utilization, we also propose a scheduling solution by interweaving one or more TGs within the same reservation block and/or backfilling with independent jobs.

[1]  Ishfaq Ahmad,et al.  On Exploiting Task Duplication in Parallel Program Scheduling , 1998, IEEE Trans. Parallel Distributed Syst..

[2]  Daniel Gajski,et al.  Hypertool: A Programming Aid for Message-Passing Systems , 1990, IEEE Trans. Parallel Distributed Syst..

[3]  Yolanda Gil,et al.  Pegasus: Mapping Scientific Workflows onto the Grid , 2004, European Across Grids Conference.

[4]  K. Mani Chandy,et al.  A comparison of list schedules for parallel processing systems , 1974, Commun. ACM.

[5]  Emmanuel Medernach,et al.  Workload Analysis of a Cluster in a Grid Environment , 2005, JSSPP.

[6]  Ishfaq Ahmad,et al.  Link contention-constrained scheduling and mapping of tasks and messages to a network of heterogeneous processors , 1999, Proceedings of the 1999 International Conference on Parallel Processing.

[7]  Douglas Thain,et al.  Distributed computing in practice: the Condor experience , 2005, Concurr. Pract. Exp..

[8]  Francine Berman,et al.  The GrADS Project: Software Support for High-Level Grid Application Development , 2001, Int. J. High Perform. Comput. Appl..

[9]  Edward G. Coffman,et al.  Computer and job-shop scheduling theory , 1976 .

[10]  Ali Afzal,et al.  Performance Architecture within ICENI , 2004 .

[11]  Wolfram Schiffmann,et al.  Improving the efficiency of functional parallelism by means of hyper-scheduling , 2006, 2006 International Conference on Parallel Processing Workshops (ICPPW'06).

[12]  Miron Livny,et al.  Condor: a distributed job scheduler , 2001 .

[13]  Ali Afzal,et al.  Workflow Enactment in ICENI , 2004 .

[14]  Matthew R. Pocock,et al.  Taverna: a tool for the composition and enactment of bioinformatics workflows , 2004, Bioinform..

[15]  Hui Li,et al.  Workload Characteristics of a Multi-cluster Supercomputer , 2004, JSSPP.

[16]  Leonel Sousa,et al.  On Task Scheduling Accuracy: Evaluation Methodology and Results , 2004, The Journal of Supercomputing.

[17]  Edward A. Lee,et al.  A Compile-Time Scheduling Heuristic for Interconnection-Constrained Heterogeneous Processor Architectures , 1993, IEEE Trans. Parallel Distributed Syst..

[18]  Dror G. Feitelson,et al.  Utilization, Predictability, Workloads, and User Runtime Estimates in Scheduling the IBM SP2 with Backfilling , 2001, IEEE Trans. Parallel Distributed Syst..

[19]  Frank D. Anger,et al.  Scheduling Precedence Graphs in Systems with Interprocessor Communication Times , 1989, SIAM J. Comput..

[20]  Rajkumar Buyya,et al.  Grid Simulation Infrastructure Supporting Advance Reservation , 2004 .

[21]  W. Schiffmann,et al.  A COMPREHENSIVE TEST BENCH FOR THE EVALUATION OF SCHEDULING HEURISTICS , 2004 .