A modeling approach for estimating execution time of long-running scientific applications

Sadjadi, S. Masoud; Shimizu, S.; Figueroa, Javier; Rangaswami, Raju; Delgado, Javier; Durán, H.; Collazo-Mojica, Xabriel J.

doi:10.1109/ipdps.2008.4536214

Cited by 35 publications

(27 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…via a careful manual inspection of the dynamic execution behavior of WRF. Delgado et al [8] (extending their earlier work in [20]) describe a regression-based approach for modeling WRF performance on systems with less than 256 processors, but their primary focus is on capturing the system related factors such as clock speed, network bandwidth, which they do via a multiplicative effects model.…”

Section: Performance Modeling and Predictionmentioning

confidence: 99%

A Divide and Conquer Strategy for Scaling Weather Simulations with Multiple Regions of Interest

Malakar

George

Kumar

et al. 2013

Scientific Programming

View full text Add to dashboard Cite

Abstract. Accurate and timely prediction of weather phenomena, such as hurricanes and flash floods, require high-fidelity compute intensive simulations of multiple finer regions of interest within a coarse simulation domain. Current weather applications execute these nested simulations sequentially using all the available processors, which is sub-optimal due to their sub-linear scalability. In this work, we present a strategy for parallel execution of multiple nested domain simulations based on partitioning the 2-D processor grid into disjoint rectangular regions associated with each domain. We propose a novel combination of performance prediction, processor allocation methods and topology-aware mapping of the regions on torus interconnects. Experiments on IBM Blue Gene systems using WRF show that the proposed strategies result in performance improvement of up to 33% with topology-oblivious mapping and up to additional 7% with topology-aware mapping over the default sequential strategy.

show abstract

Section: Performance Modeling and Predictionmentioning

confidence: 99%

A Divide and Conquer Strategy for Scaling Weather Simulations with Multiple Regions of Interest

Malakar

George

Kumar

et al. 2013

Scientific Programming

View full text Add to dashboard Cite

show abstract

“…via a careful manual inspection of the dynamic execution behavior of WRF. Delgado et al [7] (extending their earlier work in [8]) describe a regression-based approach for modeling WRF performance on systems with less than 256 processors, but their primary focus is on capturing the system related factors such as clock speed, network bandwidth, which they do via a multiplicative effects model.…”

Section: Related Workmentioning

confidence: 99%

A divide and conquer strategy for scaling weather simulations with multiple regions of interest

Malakar¹,

George²,

Kumar³

et al. 2012

2012 International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

Abstract-Accurate and timely prediction of weather phenomena, such as hurricanes and flash floods, require highfidelity compute intensive simulations of multiple finer regions of interest within a coarse simulation domain. Current weather applications execute these nested simulations sequentially using all the available processors, which is sub-optimal due to their sublinear scalability. In this work, we present a strategy for parallel execution of multiple nested domain simulations based on partitioning the 2-D processor grid into disjoint rectangular regions associated with each domain. We propose a novel combination of performance prediction, processor allocation methods and topology-aware mapping of the regions on torus interconnects. Experiments on IBM Blue Gene systems using WRF show that the proposed strategies result in performance improvement of up to 33% with topology-oblivious mapping and up to additional 7% with topology-aware mapping over the default sequential strategy.

show abstract

“…It only needs to be calculated when performing predictions among systems with very different CPU architectures. Our results in [71] showed that clock speed is a good indicator of the performance of systems with the same or similar CPU architectures, so it is not necessary to measure separate platform contribution parameters for systems with similar CPUs but different clock speeds.…”

Section: Extrapolating To Different Cpu Architecturesmentioning

confidence: 92%

“…The model we use is implemented in a profiling tool, Aprof, described in [71], which was developed as part of the Latin American Grid partnership. In this section, we summarize the implementation of the model and how it was applied to our work.…”

Section: Iv2 Prediction Model Overviewmentioning

confidence: 99%

See 1 more Smart Citation

Scheduling Medical Application Workloads on Virtualized Computing Systems

Adjouadi

Sadjadi

Delgado

Self Cite

View full text Add to dashboard Cite

A modeling approach for estimating execution time of long-running scientific applications

Abstract: In a Grid computing environment, resources

Cited by 35 publications

References 20 publications

A Divide and Conquer Strategy for Scaling Weather Simulations with Multiple Regions of Interest

A Divide and Conquer Strategy for Scaling Weather Simulations with Multiple Regions of Interest

A divide and conquer strategy for scaling weather simulations with multiple regions of interest

Scheduling Medical Application Workloads on Virtualized Computing Systems

Contact Info

Product

Resources

About