Scheduling Heterogeneous Systems Using Relative Latencies

Scheduling of parallel modules of an application may produce a significant impact on the performance. The problem of finding optimal schedule is, however, NP-complete. The diversity of the processing elements adds to the complexity that is addressed by presenting heuristic algorithms. This paper presents a novel heuristic Relative Latency-based Scheduling (RLS) for producing schedules for heterogeneous systems. The suggested approach makes use of processing capabilities of the existing processors for diverse number of successors in a task graph while making scheduling decisions. The experiments have been performed on a large number of graphs using different topologies including fork–join, Laplace equation solver, Cholesky decomposition and random graphs. The RLS variants produce better results with a significant difference in schedule length as compared to the well-known HEFT and LMT strategies used for scheduling heterogeneous systems.

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

[2]  Boontee Kruatrachue,et al.  Grain size determination for parallel processing , 1988, IEEE Software.

[3]  Jack J. Dongarra,et al.  Scheduling workflow applications on processors with different capabilities , 2006, Future Gener. Comput. Syst..

[4]  Howard Jay Siegel,et al.  A dynamic matching and scheduling algorithm for heterogeneous computing systems , 1998, Proceedings Seventh Heterogeneous Computing Workshop (HCW'98).

[5]  Zhiqing Shao,et al.  DRSCRO: A Metaheuristic Algorithm for Task Scheduling on Heterogeneous Systems , 2015 .

[6]  Philippe Chrétienne.  Task scheduling with interprocessor communication delays , 1992 .

[7]  Emmanuel Jeannot,et al.  Triplet: A clustering scheduling algorithm for heterogeneous systems , 2001, Proceedings International Conference on Parallel Processing Workshops.

[8]  Hussein M. Abdel-Wahab,et al.  Scheduling to Minimize Maximum Cumulative Cost Subject to Series-Parallel Precedence Constraints , 1978, Oper. Res..

[9]  Ishfaq Ahmad,et al.  Exploiting duplication to minimize the execution times of parallel programs on message-passing systems , 1994, Proceedings of 1994 6th IEEE Symposium on Parallel and Distributed Processing.

[10]  D. P. Agrawal,et al.  SDBS: a task duplication based optimal scheduling algorithm , 1994, Proceedings of IEEE Scalable High Performance Computing Conference.

[11]  Dharma P. Agrawal,et al.  Optimal Scheduling Algorithm for Distributed-Memory Machines , 1998, IEEE Trans. Parallel Distributed Syst..

[12]  E.L. Lawler,et al.  Optimization and Approximation in Deterministic Sequencing and Scheduling: a Survey , 1977 .

[13]  Behrooz Shirazi,et al.  DFRN: a new approach for duplication based scheduling for distributed memory multiprocessor systems , 1997, Proceedings 11th International Parallel Processing Symposium.

[14]  Tao Yang,et al.  A Comparison of Clustering Heuristics for Scheduling Directed Acycle Graphs on Multiprocessors , 1992, J. Parallel Distributed Comput..

[15]  Mayez A. Al-Mouhamed,et al.  Lower Bound on the Number of Processors and Time for Scheduling Precedence Graphs with Communication Costs , 1990, IEEE Trans. Software Eng..

[16]  Ishfaq Ahmad,et al.  Dynamic Critical-Path Scheduling: An Effective Technique for Allocating Task Graphs to Multiprocessors , 1996, IEEE Trans. Parallel Distributed Syst..

[17]  B. J. Lageweg,et al.  Multiprocessor scheduling with communication delays , 1990, Parallel Comput..

[18]  Marinus Veldhorst.  A linear time algorithm to schedule trees with communication delays optimally on two machines , 1993 .

[19]  Salim Hariri,et al.  Task scheduling algorithms for heterogeneous processors , 1999, Proceedings. Eighth Heterogeneous Computing Workshop (HCW'99).

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

[21]  Ümit V. Çatalyürek,et al.  Compaction of Schedules and a Two-Stage Approach for Duplication-Based DAG Scheduling , 2009, IEEE Transactions on Parallel and Distributed Systems.

[22]  Henri Casanova,et al.  From Heterogeneous Task Scheduling to Heterogeneous Mixed Parallel Scheduling , 2004, Euro-Par.

[23]  Hesham El-Rewini,et al.  Scheduling Parallel Program Tasks onto Arbitrary Target Machines , 1990, J. Parallel Distributed Comput..

[24]  Jing-Chiou Liou,et al.  An Efficient Task Clustering Heuristic for Scheduling DAGs on Multiprocessors , 2007 .

[25]  Tao Yang,et al.  DSC: Scheduling Parallel Tasks on an Unbounded Number of Processors , 1994, IEEE Trans. Parallel Distributed Syst..

[26]  Kenli Li,et al.  A DAG scheduling scheme on heterogeneous computing systems using double molecular structure-based chemical reaction optimization , 2013, J. Parallel Distributed Comput..

[27]  Jing-Jang Hwang,et al.  Multiprocessor scheduling with interprocessor communication delays , 1988 .

[28]  Atakan Dogan,et al.  LDBS: a duplication based scheduling algorithm for heterogeneous computing systems , 2002, Proceedings International Conference on Parallel Processing.

[29]  Minhaj Ahmad Khan.  Task scheduling for heterogeneous systems using an incremental approach , 2016, The Journal of Supercomputing.

[30]  Santosh Pande,et al.  Efficient Program Partitioning Based on Compiler Controlled Communication , 1999, IPPS/SPDP Workshops.

[31]  Mihalis Yannakakis,et al.  Towards an architecture-independent analysis of parallel algorithms , 1990, STOC '88.

[32]  Virgílio A. F. Almeida,et al.  Using random task graphs to investigate the potential benefits of heterogeneity in parallel systems , 1992, Proceedings Supercomputing '92.

[33]  Qing Guo,et al.  A Lower Extremity Exoskeleton: Human-Machine Coupled Modeling, Robust Control Design, Simulation, and Overload-Carrying Experiment , 2015 .