The hybrid dynamic parallel scheduling algorithm for load balancing on Chained-Cubic Tree interconnection networks

The Chained-Cubic Tree (CCT) interconnection network topology was recently proposed as a continuation for the extended efforts in the area of interconnection networks’ performance improvement. This topology, which promises to exhibit the best properties of the hypercube and tree topologies, needs to be deeply investigated in order to evaluate its performance among other interconnection networks’ topologies. This work comes as a complementary effort, in which the load balancing technique is investigated as one of the most important aspects of performance improvement. This paper proposes a new load balancing algorithm on CCT interconnection networks. The proposed algorithm, which is called Hybrid Dynamic Parallel Scheduling Algorithm (HD-PSA), is a combination of two common load balancing strategies; dynamic load balancing and parallel scheduling. The performance of the proposed algorithm is evaluated both, analytically and experimentally, in terms of various performance metrics; including, execution time, load balancing accuracy, communication cost, number of tasks hops, and tasks locality.

[1]  Wei Shi,et al.  Hierarchical star: a new two level interconnection network , 2005, J. Syst. Archit..

[2]  Min-You Wu,et al.  A load-balancing algorithm for N-cubes , 1996, Proceedings of the 1996 ICPP Workshop on Challenges for Parallel Processing.

[3]  Sartaj Sahni,et al.  Programming a hypercube multicomputer , 1988, IEEE Software.

[4]  Yong Qin,et al.  Hybrid Diffusion Schemes for Load Balancing on OTIS-Networks , 2007, ICA3PP.

[5]  Deron Liang,et al.  Concentrations, Load Balancing, Multicasting and Partial Permutation Routing on Hypercube Parallel Computers , 2002, J. Inf. Sci. Eng..

[6]  Basel A. Mahafzah,et al.  The load balancing problem in OTIS-Hypercube interconnection networks , 2008, The Journal of Supercomputing.

[7]  Emad Abuelrub,et al.  The chained-cubic tree interconnection network , 2011, Int. Arab J. Inf. Technol..

[8]  Min-You Wu,et al.  An Incremental Parallel Scheduling Approach to Solving Dynamic and Irregular Problems , 1995, ICPP.

[9]  Min-You Wu,et al.  On Runtime Parallel Scheduling for Processor Load Balancing , 1997, IEEE Trans. Parallel Distributed Syst..

[10]  Eunseuk Oh,et al.  Hyper-Star Graph: A New Interconnection Network Improving the Network Cost of the Hypercube , 2002, EurAsia-ICT.

[11]  Ce-Kuen Shieh,et al.  A Group-Based Load Balance Scheme for Software Distributed Shared Memory Systems , 2004, The Journal of Supercomputing.

[12]  Gang Qu,et al.  An area-efficient high-throughput hybrid interconnection network for single-chip parallel processing , 2008, 2008 45th ACM/IEEE Design Automation Conference.

[13]  Eunseuk Oh,et al.  Matrix-Star Graphs: A New Interconnection Network Based on Matrix Operations , 2005, Asia-Pacific Computer Systems Architecture Conference.

[14]  Min-You Wu,et al.  Runtime Incremental Parallel Scheduling (RIPS) on Distributed Memory Computers , 1996, IEEE Trans. Parallel Distributed Syst..

[15]  Laurence T. Yang,et al.  Comparative study of the server-initiated lowest algorithm using a load balancing index based on the process behavior for heterogeneous environment , 2006, Cluster Computing.

[16]  Han Jungang,et al.  Double-Loop Hypercube: A New Scalable Interconnection Network for Massively Parallel Computing , 2008, 2008 ISECS International Colloquium on Computing, Communication, Control, and Management.

[17]  Subir Bandyopadhyay,et al.  A New Network Topology with Multiple Three-Dimensional Meshes , 2004, IWDC.

[18]  Gene Eu Jan,et al.  Effective load balancing on highly parallel multicomputers based on superconcentrators , 1994, Proceedings of 1994 International Conference on Parallel and Distributed Systems.

[19]  Gene Eu Jan,et al.  An Efficient Algorithm for Perfect Load Balancing on Hypercube Multiprocessors , 2004, The Journal of Supercomputing.

[20]  Hwakyung Rim,et al.  A simple reduction of non-uniformity in dynamic load balancing of quantized loads on hypercube multiprocessors and hiding balancing overheads , 2003, J. Comput. Syst. Sci..

[21]  Yong Qin,et al.  GDED-X Schemes for Load Balancing on Heterogeneous OTIS-Networks , 2007, ICA3PP.

[22]  Bharadwaj Veeravalli,et al.  Design and Performance Analysis of Divisible Load Scheduling Strategies on Arbitrary Graphs , 2004, Cluster Computing.

[23]  Bhabani P. Sinha,et al.  A New Network Topology with Multiple Meshes , 1999, IEEE Trans. Computers.

[24]  Hwakyung Rim,et al.  An efficient dynamic load balancing using the dimension exchange method for balancing of quantized loads on hypercube multiprocessors , 1999, Proceedings 13th International Parallel Processing Symposium and 10th Symposium on Parallel and Distributed Processing. IPPS/SPDP 1999.