Steady-state scheduling of task graphs on heterogeneous computing platforms

In this paper, we consider the execution of a complex application on a heterogeneous "grid" computing platform. The complex application consists of a suite of identical, independent problems to be solved. In turn, each problem consists of a set of tasks. There are dependences (precedence constraints) between these tasks. A typical example is the repeated execution of the same algorithm on several distinct data samples. We use a non-oriented graph to model the grid platform, where resources have different speeds of computation and communication. We show how to determine the optimal steady-state scheduling strategy for each processor (the fraction of time spent computing and the fraction of time spent communicating with each neighbor) and how to build such as schedule. This result holds for a quite general framework, allowing for cycles and multiple paths in the platform graph.

[1]  Ali R. Hurson,et al.  Scheduling and Load Balancing in Parallel and Distributed Systems , 1995 .

[2]  Sachin S. Sapatnekar,et al.  A Framework for Exploiting Task and Data Parallelism on Distributed Memory Multicomputers , 1997, IEEE Trans. Parallel Distributed Syst..

[3]  Z Liu,et al.  Scheduling Theory and its Applications , 1997 .

[4]  Joel H. Saltz,et al.  Optimizing execution of component-based applications using group instances , 2001, Proceedings First IEEE/ACM International Symposium on Cluster Computing and the Grid.

[5]  Larry Carter,et al.  Bandwidth-centric allocation of independent tasks on heterogeneous platforms , 2002, Proceedings 16th International Parallel and Distributed Processing Symposium.

[6]  Thomas R. Gross,et al.  Exploiting task and data parallelism on a multicomputer , 1993, PPOPP '93.

[7]  David S. Johnson,et al.  Computers and Intractability: A Guide to the Theory of NP-Completeness , 1978 .

[8]  Francine Berman,et al.  Adaptive scheduling of master/worker applications on distributed computational resources , 2001 .

[9]  Joel H. Saltz,et al.  Executing Multiple Pipelined Data Analysis Operations in the Grid , 2002, ACM/IEEE SC 2002 Conference (SC'02).

[10]  Yves Robert,et al.  Scheduling Strategies for Master-Slave Tasking on Heterogeneous Processor Grids , 2002, PARA.

[11]  Larry Carter,et al.  Scheduling strategies for master-slave tasking on heterogeneous processor platforms , 2004, IEEE Transactions on Parallel and Distributed Systems.

[12]  Alexander Schrijver,et al.  Combinatorial optimization. Polyhedra and efficiency. , 2003 .

[13]  J. Dongarra,et al.  Algorithmic Issues on Heterogeneous Computing Platforms Algorithmic Issues on Heterogeneous Computing Platforms Algorithmic Issues on Heterogeneous Computing Platforms , 1998 .

[14]  Yves Robert,et al.  The Master-Slave Paradigm with Heterogeneous Processors , 2001, CLUSTER.

[15]  Larry Carter,et al.  Autonomous protocols for bandwidth-centric scheduling of independent-task applications , 2003, Proceedings International Parallel and Distributed Processing Symposium.

[16]  Matthew Haines,et al.  Approaches for integrating task and data parallelism , 1998, IEEE Concurr..

[17]  Jeff T. Linderoth,et al.  An enabling framework for master-worker applications on the Computational Grid , 2000, Proceedings the Ninth International Symposium on High-Performance Distributed Computing.

[18]  Jaspal Subhlok,et al.  Optimal Use of Mixed Task and Data Parallelism for Pipelined Computations , 2000, J. Parallel Distributed Comput..

[19]  Jon B. Weissman,et al.  Scheduling multi-component applications in heterogeneous wide-area networks , 2000, Proceedings 9th Heterogeneous Computing Workshop (HCW 2000) (Cat. No.PR00556).

[20]  James Demmel,et al.  Models and Scheduling Algorithms for Mixed Data and Task Parallel Programs , 1997, J. Parallel Distributed Comput..

[21]  Francine Berman,et al.  Master/slave computing on the Grid , 2000, Proceedings 9th Heterogeneous Computing Workshop (HCW 2000) (Cat. No.PR00556).

[22]  Arjen K. Lenstra,et al.  A World Wide Number Field Sieve Factoring Record: On to 512 Bits , 1996, ASIACRYPT.

[23]  Miron Livny,et al.  Adaptive Scheduling for Master-Worker Applications on the Computational Grid , 2000, GRID.