Analytic processor model for fast design-space exploration

Jongerius, Rik; Mariani, Giovanni; Anghel, Andreea; Dittmann, Gero; Vermij, Erik; Corporaal, Henk

doi:10.1109/iccd.2015.7357136

Cited by 11 publications

(9 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The application properties to be extrapolated are an input for these analytic hardware models. The work [6,14,22] assumes that these input properties are available and, in contrast to our work, it does not propose a methodology to retrieve those properties for exascale datasets. In the radio-astronomy case study addressed in this work, we demonstrate how an analytic hardware performance model such as the one proposed by Jongerius et al [22] can be coupled to the extrapolation models we are proposing to predict the performance of the exascale system.…”

Section: Introductionmentioning

confidence: 93%

“…The performance estimation error of the analytic hardware platform model we use is about 25 % [22]. Given the time saved by the proposed methodology, we consider this accuracy acceptable compared with the Sniper error rate of 21 % [10,22].…”

Section: Comparison To Simulation-based Approachesmentioning

confidence: 99%

“…In previous work, an analytic performance model for Xeon-based processors has been developed [22]. It takes as input the profileγ (λ) we extrapolate.…”

Section: Extrapolating For Exascale Designmentioning

confidence: 99%

“…Given the time saved by the proposed methodology, we consider this accuracy acceptable compared with the Sniper error rate of 21 % [10,22].…”

Section: Comparison To Simulation-based Approachesmentioning

confidence: 99%

“…instruction-level parallelism, instruction mix, etc.) and actual hardware performance with analytic models [6,14,22]. The application properties to be extrapolated are an input for these analytic hardware models.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Scaling Properties of Parallel Applications to Exascale

Mariani

Anghel

Jongerius

et al. 2016

Int J Parallel Prog

Self Cite

View full text Add to dashboard Cite

A detailed profile of exascale applications helps to understand the computation, communication and memory requirements for exascale systems and provides the insight necessary for fine-tuning the computing architecture. Obtaining such a profile is challenging as exascale systems will process unprecedented amounts of data. Profiling applications at the target scale would require the exascale machine itself. In this work we propose a methodology to extrapolate the exascale profile from experimental observations over datasets feasible for today's machines. Extrapolation models are carefully selected by means of statistical techniques and a high-level complexity analysis is included in the selection process to speed up the learning phase and to improve the accuracy of the final model. We extrapolate run-time properties of the target applications including information about the instruction mix, memory access pattern, instruction-level parallelism, and communication requirements. Compared to state-of-the-art techniques, the proposed methodology reduces the prediction error by an order of magnitude on the instruction count and improves the accuracy by up to 1.3× for the memory access pattern, and by more than 2× for the communication requirements.

show abstract

Section: Introductionmentioning

confidence: 93%

Section: Comparison To Simulation-based Approachesmentioning

confidence: 99%

“…In previous work, an analytic performance model for Xeon-based processors has been developed [22]. It takes as input the profileγ (λ) we extrapolate.…”

Section: Extrapolating For Exascale Designmentioning

confidence: 99%

“…Given the time saved by the proposed methodology, we consider this accuracy acceptable compared with the Sniper error rate of 21 % [10,22].…”

Section: Comparison To Simulation-based Approachesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Scaling Properties of Parallel Applications to Exascale

Mariani

Anghel

Jongerius

et al. 2016

Int J Parallel Prog

Self Cite

View full text Add to dashboard Cite

show abstract

Multicore Performance Prediction – Comparing Three Recent Approaches in a Case Study

Lüders

Arndt

Blume

2020

Euro-Par 2019: Parallel Processing Workshops

View full text Add to dashboard Cite

Relative Performance Projection on Arm Architectures

Gavoille

Taboada

Carribault

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

show abstract

Analytic processor model for fast design-space exploration

Cited by 11 publications

References 9 publications

Scaling Properties of Parallel Applications to Exascale

Scaling Properties of Parallel Applications to Exascale

Multicore Performance Prediction – Comparing Three Recent Approaches in a Case Study

Relative Performance Projection on Arm Architectures

Contact Info

Product

Resources

About