Adaptive methods for irregular parallel discrete event simulation workloads

Eric Mikida, Laxmikant Kale

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Parallel Discrete Event Simulations (PDES) running at large scales involve the coordination of billions of very ne grain events distributed across a large number of processes. At such large scales optimistic synchronization protocols, such as TimeWarp, allow for a high degree of parallelism between processes, but with the additional complexity of managing event rollback and cancellation. This can become especially problematic in models that exhibit imbalance resulting in low event eciency, which increases the total amount of work required to run a simulation to completion. Managing this complexity becomes key to achieving a high degree of performance across a wide range of models. In this paper, we address this issue by analyzing the relationship between synchronization cost and event eciency. We rst look at how these two characteristics are coupled via the computation of Global Virtual Time (GVT). We then introduce dynamic load balancing, and show how, when combined with low overhead GVT computation, we can achieve higher e-ciency with less synchronization cost. In doing so, we achieve up to 2 better performance on a variety of benchmarks and models of practical importance.

Original languageEnglish (US)
Title of host publicationSIGSIM-PADS 2018 - Proceedings of the 2018 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation
PublisherAssociation for Computing Machinery, Inc
Pages189-200
Number of pages12
ISBN (Electronic)9781450350921
DOIs
StatePublished - May 14 2018
Event2018 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation, SIGSIM-PADS 2018 - Rome, Italy
Duration: May 23 2018May 25 2018

Publication series

NameSIGSIM-PADS 2018 - Proceedings of the 2018 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation

Other

Other2018 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation, SIGSIM-PADS 2018
CountryItaly
CityRome
Period5/23/185/25/18

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Modeling and Simulation

Fingerprint Dive into the research topics of 'Adaptive methods for irregular parallel discrete event simulation workloads'. Together they form a unique fingerprint.

Cite this