Optimal periodic remapping of dynamic bulk synchronous computations

Fong, Ngo-Tai; Xu, Cheng‐Zhong; Wang, Le Yi

doi:10.1016/j.jpdc.2003.07.004

Cited by 8 publications

(4 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Some works have focused on stochastic task scheduling [16]- [21], [24], [25], [35]. Dong et al proposed a mechanism to estimate the probability distribution of a task's execution time based on resource load, and a resource load based stochastic scheduling algorithm for grid environments was developed [25].…”

Section: Motivationmentioning

confidence: 99%

Scheduling Precedence Constrained Stochastic Tasks on Heterogeneous Cluster Systems

Veeravalli

et al. 2015

IEEE Trans. Comput.

178

View full text Add to dashboard Cite

Generally, a parallel application consists of precedence constrained stochastic tasks, where task processing times and intertask communication times are random variables following certain probability distributions. Scheduling such precedence constrained stochastic tasks with communication times on a heterogeneous cluster system with processors of different computing capabilities to minimize a parallel application's expected completion time is an important but very difficult problem in parallel and distributed computing. In this paper, we present a model of scheduling stochastic parallel applications on heterogeneous cluster systems. We discuss stochastic scheduling attributes and methods to deal with various random variables in scheduling stochastic tasks. We prove that the expected makespan of scheduling stochastic tasks is greater than or equal to the makespan of scheduling deterministic tasks, where all processing times and communication times are replaced by their expected values. To solve the problem of scheduling precedence constrained stochastic tasks efficiently and effectively, we propose a stochastic dynamic level scheduling (SDLS) algorithm, which is based on stochastic bottom levels and stochastic dynamic levels. Our rigorous performance evaluation results clearly demonstrate that the proposed stochastic task scheduling algorithm significantly outperforms existing algorithms in terms of makespan, speedup, and makespan standard deviation.

show abstract

Section: Motivationmentioning

confidence: 99%

Scheduling Precedence Constrained Stochastic Tasks on Heterogeneous Cluster Systems

Veeravalli

et al. 2015

IEEE Trans. Comput.

178

View full text Add to dashboard Cite

show abstract

“…O remapeamento a cada k etapas tem sido aplicado a muitas aplicac ¸ões paralelas. Em particular, esta técnica é usada em aplicac ¸ões que exibem alterac ¸ão gradual na carga de trabalho [7]. MigBSP usa uma variável denotada α que é atualizada a cada chamada de reescalonamento, indicando o intervalo da próxima.…”

Section: Oferecendo Tratamento Dinâmico Para O Reescalonamento De Pro...unclassified

Process Migration: Controlling Application and Resource Dynamics by Combining Computation, Communication and Memory Metrics

Graebin

Righi

Navaux

2011

Anais Do XII Simpósio Em Sistemas Computacionais De Alto Desempenho (SSCAD 2011)

View full text Add to dashboard Cite

In this paper we present MigBSP, a rescheduling model that acts on Bulk Synchronous Parallel applications. It combines the metrics Computation, Communication and Memory to make migration decisions on Computation Grids. MigBSP also offers efficient adaptations to reduce its own overhead. Additionally, MigBSP is infrastructure and application independent and tries to handle dynamicity on both levels. MigBSP’s results show performance gains of up to 16% on dynamic environments while maintaining a small overhead when migrations do not take place.

show abstract

“…In [5,6], the authors addressed the above problem with homogeneous processors and derived accurate closed form approximations of T . In this paper, we further address the problem in heterogeneous computing systems where processors have different speeds.…”

Section: Introductionmentioning

confidence: 99%

Optimal Period of Workload Redistribution for Dynamic Bulk Synchronous Computations in Heterogeneous Computing Systems

2006

J Supercomput

View full text Add to dashboard Cite

In dynamic bulk synchronous computations, processors may change their workloads from phase to phase. Such workload change will possibly increase the duration of a phase and the overall parallel execution time. Therefore, it is necessary to redistribute workload at runtime to reduce the parallel time. However, such workload redistribution at runtime can be expensive and the overhead of frequent runtime workload redistribution may exceed the benefit of workload redistribution and balancing. The problem of finding the optimal period of runtime workload redistribution for dynamic bulk synchronous computations is motivated by the combined consideration of parallel execution time and system overhead for workload redistribution and balancing. We develop an analytical method to solve the problem in heterogeneous computing systems. We also demonstrate numerical data of the analytical method and simulation results that verify the analytical data.

show abstract

Optimal periodic remapping of dynamic bulk synchronous computations

Cited by 8 publications

References 25 publications

Scheduling Precedence Constrained Stochastic Tasks on Heterogeneous Cluster Systems

Scheduling Precedence Constrained Stochastic Tasks on Heterogeneous Cluster Systems

Process Migration: Controlling Application and Resource Dynamics by Combining Computation, Communication and Memory Metrics

Optimal Period of Workload Redistribution for Dynamic Bulk Synchronous Computations in Heterogeneous Computing Systems

Contact Info

Product

Resources

About