Application-aware scheduling of a magnetohydrodynamics application in the Legion metasystem

Dail, Holly; Obertelli, Graziano; Berman, Francine; Wolski, Richard; Grimshaw, Andrew S.

doi:10.1109/hcw.2000.843746

Cited by 23 publications

(18 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This raises the need for the development of techniques that make applications robust for the continuously changing circumstances of the grid environment. Those techniques strongly rely on the effectiveness of prediction methods: significant speedups can be obtained by good scheduling schemes based on accurate predictions, as is demonstrated in an experimental setting in [2,4,13]. Similar observations are made in Dobber et al [6] where we investigate the impact of fluctuations in processing speeds on running times of grid applications.…”

Section: Introductionsupporting

confidence: 56%

A prediction method for job runtimes on shared processors: Survey, statistical analysis and new avenues

Dobber

Mei

Koole

2007

Performance Evaluation

View full text Add to dashboard Cite

Grid computing is an emerging technology by which huge numbers of processors over the world create a global source of processing power. Their collaboration makes it possible to perform computations that are too extensive to perform on a single processor. On a grid processors may connect and disconnect at any time, and the load on the computers can be highly bursty. Those characteristics raise the need for the development of techniques that make grid applications robust against the dynamics of the grid environment. In particular, applications that use significant amounts of processor power for running jobs need effective predictions of the expected computation times of those jobs on remote hosts. Currently, there are no effective prediction methods available that cope with the ever-changing running times of jobs on a grid environment. Motivated by this, we develop the Dynamic Exponential Smoothing (DES) method to predict running times in a grid environment. The main idea behind DES is that it dynamically adapts its prediction strategy to the height of the fluctuations in those running times. We have performed extensive experiments in a real global-scale grid environment to compare the effectiveness of DES. The results demonstrate that DES strongly and consistently outperforms existing prediction methods.

show abstract

Section: Introductionsupporting

confidence: 56%

A prediction method for job runtimes on shared processors: Survey, statistical analysis and new avenues

Dobber

Mei

Koole

2007

Performance Evaluation

View full text Add to dashboard Cite

show abstract

“…In future work, we will add Nimrod/G's notion of resource cost to our current scheduling model (see Section 6). Finally, the AppLeS described in this paper builds upon other previous AppLeS work [18][19][20]4] in its strategies for resource selection and work allocation. These AppLeS have focused on improving the performance of applications with fixed configurations.…”

Section: Related Workmentioning

confidence: 99%

“…The scheduler makes predictions of the performance the application may experience on prospective resources at execution time. Using these predictions, a potentially performance-efficient schedule for the application is identified and deployed [18][19][20]4]. In the case of GTOMO, an embarrassingly parallel application, it is natural to use self-scheduling [21].…”

Section: Off-line Parallel Tomographymentioning

confidence: 99%

See 1 more Smart Citation

Applying Scheduling and Tuning to On‐line Parallel Tomography

Smallen

Casanova

Berman

2002

Scientific Programming

View full text Add to dashboard Cite

Abstract. Tomography is a popular technique to reconstruct the three-dimensional structure of an object from a series of twodimensional projections. Tomography is resource-intensive and deployment of a parallel implementation onto Computational Grid platforms has been studied in previous work. In this work, we address on-line execution of the application where computation is performed as data is collected from an on-line instrument. The goal is to compute incremental 3-D reconstructions that provide quasi-real-time feedback to the user. We model on-line parallel tomography as a tunable application: trade-offs between resolution of the reconstruction and frequency of feedback can be used to accommodate various resource availabilities. We demonstrate that application scheduling/tuning can be framed as multiple constrained optimization problems and evaluate our methodology in simulation. Our results show that prediction of dynamic network performance is key to efficient scheduling and that tunability allows for production runs of on-line parallel tomography in Computational Grid environments.

show abstract

“…Huh et al [5] present a method for unifying dynamic resource requirements among heterogeneous hosts of a RMS, using a system resource model with initial profiles for applications. A scheduler for a RMS is presented by Dail et al [2] along with its performance model, which requires "good predictions of megabytes transferred, number of messages initiated, overhead factor, benchmarks for program CPU and memory utilization over the different target architectures".…”

Section: Introductionmentioning

confidence: 99%