Balancing Load versus Decreasing Communication: Parameterizing the Tradeoff

Taylor, Valerie; Schwabe, Eric J.; Holmer, Bruce K.; Hribar, Michelle R.

doi:10.1006/jpdc.2001.1701

Cited by 11 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This decrease in t max occurs together with an increase in the vertex weight imbalance. This confirms observations made in [10], that allowing some imbalance in the partition can lead to lower estimated execution times, and that the benefit of this imbalance increases with the number of processors and with the communication to computation ratio. Even though the PaGrid partitions are imbalanced in terms of vertex weights, they are all balanced in terms of t p , with i t < 1.03 in all cases.…”

Section: Homogeneous Clusterssupporting

confidence: 90%

Partitioning and Mapping of Mesh-Based Applications onto Computational Grids

Wanschoor

Aubanel

Fifth IEEE/ACM International Workshop on Grid Computing

View full text Add to dashboard Cite

Mesh-based applications, such as those that involve the numerical solution of partial differential equations, may be able to take advantage of the performance of computational grids. To do so will require mesh partitioners that take the heterogeneity of the computational platform into account. Recent work in our group led to the creation of a heterogeneous mesh partitioner, PaGrid. We present a redesigned version of PaGrid, which uses estimated execution time as a cost function in all levels of multilevel refinement. It takes into account the characteristics of the application (computational complexity and size of messages) and of the computing platform (processor and network speeds), and balances the estimated execution time of processors. This results in partitions with up to 60% lower estimated execution times than METIS, a homogeneous partitioner, and similar improvements over JOSTLE, a heterogeneous partitioner. PaGrid achieves this in a reasonable amount of time, taking only two to three times longer than METIS.

show abstract

Section: Homogeneous Clusterssupporting

confidence: 90%

Partitioning and Mapping of Mesh-Based Applications onto Computational Grids

Wanschoor

Aubanel

Fifth IEEE/ACM International Workshop on Grid Computing

View full text Add to dashboard Cite

show abstract

“…For the future work, we plan to improve the 338 Parallel Simulations of Dynamic Earthquake Rupture Along Geometrically Complex Faults on CMP Systems Figure 18. Speedup of the hybrid implementation on Hydra memory requirements of the hybrid simulation code by localizing large global arrays and data structures, and consider load balancing strategies [18] so that the hybrid earthquake simulation with large production datasets can be run on up to 10,000 processors.…”

Section: Discussionmentioning

confidence: 99%

Parallel Simulations of Dynamic Earthquake Rupture along Geometrically Complex Faults on CMP Systems

Duan

Taylor

2011

Journal of Algorithms & Computational Technology

View full text Add to dashboard Cite

Chip multiprocessors (CMP) are widely used for high performance computing and are being configured in a hierarchical manner to compose a CMP compute node in a CMP system. Such a CMP system provides a natural programming paradigm for hybrid MPI/OpenMP applications. In this paper, we use OpenMP to parallelize a sequential earthquake simulation code for modeling spontaneous earthquake rupture along geometrically complex faults on two CMP systems, IBM POWER5+ system and SUN Opteron server. The experimental results indicate that the OpenMP implementation has the accurate output results and the good scalability on the two CMP systems. We apply the optimization techniques such as large page and processor binding to the OpenMP implementation to achieve up to 7.05% performance improvement on the CMP systems without any code modification. Further, we illustrate an element-based partitioning scheme for explicit finite element methods. Based on the partitioning scheme and what we learn from the OpenMP implementation, we discuss how efficiently to use hybrid MPI/OpenMP to parallelize the sequential earthquake rupture simulation code in order to not only achieve multiple levels of parallelism of the code but also to reduce the communication overhead of MPI within a CMP node by taking advantage of the shared address space and on-chip high inter-core Parallel Simulations of Dynamic Earthquake Rupture Along Geometrically Complex Faults on CMP Systems bandwidth and low inter-core latency. Our initial experimental results indicate that the hybrid MPI/OpenMP implementation obtains the accurate output results and has good scalability on CMP systems.

show abstract

“…Since in distributed systems, the overhead time consumed in processors communication is a significant factor, some approaches include this parameter in the balancing decision. Taylor et al [35] in their work presented a trade-off function that indicates whether or not load imbalance should be explored. This function is the ratio of the decrease in communication requirements to the increase in computational load for load imbalance.…”

Section: Load Information Collectionmentioning

confidence: 98%

A Dynamic Load Balancing Framework for Real-time Applications in Message Passing Systems

Kabbany

Wanas²,

Hegazi

et al. 2010

Int J Parallel Prog

View full text Add to dashboard Cite

Load balancing algorithms are designed essentially to equally distribute the load on processors and maximize their utilities while minimizing the total task execution time. In order to achieve these goals, the load-balancing mechanism should be "fair" in distributing the load across the different processors. This implies that the difference between the heaviest-loaded and the lightest-loaded processors should be minimized. Therefore, the load information on each processor must be updated such that the load-balancing mechanism can be more effective. In this work, we present an application independent dynamic algorithm for scheduling tasks and loadbalancing in message passing systems. We propose a DAG-based Dynamic Load Balancing algorithm for Real time applications (DAG-DLBR) that is designed to work dynamically to cope with possible changes in the load that might occur during runtime. This algorithm addresses the challenge of devising a load balancing scheme which judicially deals with the hybrid execution of existing real-time application (represented by a Direct Acyclic Graph (DAG)) together with newly arriving jobs. The main objective of this algorithm is to reduce response times of the newly arriving jobs while maintaining the time constrains of the existing DAG. To evaluate the performance of the DAG-DLBR algorithm, a comparison with the performance of two common dynamic load balancing algorithms is presented. This comparison is performed by evaluating, experimentally, the execution time of different load balancing algorithms on a homogenous real parallel machine. In addition, the values of load imbalance, the execution time, and the communication overhead time are evaluated analytically using different benchmarks as test-bed workloads. These workloads cover a wide range of dynamic applications with different task types. Experimental results illustrate the improved performance of the DAG-DLBR algorithm compared to both distributed and hierarchal based algorithms by at least 12 and 19%, respectively. This improvement is true for all workloads, even with highly dependent workload. The DAG-DLBR algorithm achieves lower computation time than its corresponding values of both the distributed and the hierarchical-based algorithms for 4, 8, 12 and 16 processors.

show abstract

Balancing Load versus Decreasing Communication: Parameterizing the Tradeoff

Cited by 11 publications

References 19 publications

Partitioning and Mapping of Mesh-Based Applications onto Computational Grids

Partitioning and Mapping of Mesh-Based Applications onto Computational Grids

Parallel Simulations of Dynamic Earthquake Rupture along Geometrically Complex Faults on CMP Systems

A Dynamic Load Balancing Framework for Real-time Applications in Message Passing Systems

Contact Info

Product

Resources

About