Performance analysis techniques for the exascale co-design process

Martin Schulz, Jim Belak, Abhinav Bhatele, Peer Timo Bremer, Greg Bronevetsky, Marc Casas, Todd Gamblin, Katherine E. Isaacs, Ignacio Laguna, Joshua Levine, Valerio Pascucci, David Richards, Barry Rountree

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

3 Citations (Scopus)

Abstract

Efficient and effective performance analysis techniques are critical for the development of future generation systems. They are the drivers behind the required co-design process that helps establish the principles needed for their design. In this paper, we will highlight two such approaches: PAVE, a project that investigates mapping of performance data to more intuitive domains and uses advanced visualization techniques to expose problems, and GREMLIN, a system evaluation environment capable of emulating expected properties of exascale architectures on petascale machines. Combined with other approaches in system modeling and simulation, these projects enable us to provide a meaningful introspection into a target application's characteristics as well as its expected behavior and, more importantly, likely bottlenecks on future generation machines.

Original languageEnglish (US)
Title of host publicationParallel Computing
Subtitle of host publicationAccelerating Computational Science and Engineering (CSE)
PublisherIOS Press BV
Pages19-32
Number of pages14
ISBN (Print)9781614993803
DOIs
StatePublished - Jan 1 2014
Externally publishedYes

Publication series

NameAdvances in Parallel Computing
Volume25
ISSN (Print)0927-5452

Fingerprint

Visualization

Keywords

  • Architecture Emulation
  • Co-Design
  • Performance Analysis
  • Performance Visualization

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Schulz, M., Belak, J., Bhatele, A., Bremer, P. T., Bronevetsky, G., Casas, M., ... Rountree, B. (2014). Performance analysis techniques for the exascale co-design process. In Parallel Computing: Accelerating Computational Science and Engineering (CSE) (pp. 19-32). (Advances in Parallel Computing; Vol. 25). IOS Press BV. https://doi.org/10.3233/978-1-61499-381-0-19

Performance analysis techniques for the exascale co-design process. / Schulz, Martin; Belak, Jim; Bhatele, Abhinav; Bremer, Peer Timo; Bronevetsky, Greg; Casas, Marc; Gamblin, Todd; Isaacs, Katherine E.; Laguna, Ignacio; Levine, Joshua; Pascucci, Valerio; Richards, David; Rountree, Barry.

Parallel Computing: Accelerating Computational Science and Engineering (CSE). IOS Press BV, 2014. p. 19-32 (Advances in Parallel Computing; Vol. 25).

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

Schulz, M, Belak, J, Bhatele, A, Bremer, PT, Bronevetsky, G, Casas, M, Gamblin, T, Isaacs, KE, Laguna, I, Levine, J, Pascucci, V, Richards, D & Rountree, B 2014, Performance analysis techniques for the exascale co-design process. in Parallel Computing: Accelerating Computational Science and Engineering (CSE). Advances in Parallel Computing, vol. 25, IOS Press BV, pp. 19-32. https://doi.org/10.3233/978-1-61499-381-0-19
Schulz M, Belak J, Bhatele A, Bremer PT, Bronevetsky G, Casas M et al. Performance analysis techniques for the exascale co-design process. In Parallel Computing: Accelerating Computational Science and Engineering (CSE). IOS Press BV. 2014. p. 19-32. (Advances in Parallel Computing). https://doi.org/10.3233/978-1-61499-381-0-19
Schulz, Martin ; Belak, Jim ; Bhatele, Abhinav ; Bremer, Peer Timo ; Bronevetsky, Greg ; Casas, Marc ; Gamblin, Todd ; Isaacs, Katherine E. ; Laguna, Ignacio ; Levine, Joshua ; Pascucci, Valerio ; Richards, David ; Rountree, Barry. / Performance analysis techniques for the exascale co-design process. Parallel Computing: Accelerating Computational Science and Engineering (CSE). IOS Press BV, 2014. pp. 19-32 (Advances in Parallel Computing).
@inproceedings{390170a14e654218820bdf46755cd5d2,
title = "Performance analysis techniques for the exascale co-design process",
abstract = "Efficient and effective performance analysis techniques are critical for the development of future generation systems. They are the drivers behind the required co-design process that helps establish the principles needed for their design. In this paper, we will highlight two such approaches: PAVE, a project that investigates mapping of performance data to more intuitive domains and uses advanced visualization techniques to expose problems, and GREMLIN, a system evaluation environment capable of emulating expected properties of exascale architectures on petascale machines. Combined with other approaches in system modeling and simulation, these projects enable us to provide a meaningful introspection into a target application's characteristics as well as its expected behavior and, more importantly, likely bottlenecks on future generation machines.",
keywords = "Architecture Emulation, Co-Design, Performance Analysis, Performance Visualization",
author = "Martin Schulz and Jim Belak and Abhinav Bhatele and Bremer, {Peer Timo} and Greg Bronevetsky and Marc Casas and Todd Gamblin and Isaacs, {Katherine E.} and Ignacio Laguna and Joshua Levine and Valerio Pascucci and David Richards and Barry Rountree",
year = "2014",
month = "1",
day = "1",
doi = "10.3233/978-1-61499-381-0-19",
language = "English (US)",
isbn = "9781614993803",
series = "Advances in Parallel Computing",
publisher = "IOS Press BV",
pages = "19--32",
booktitle = "Parallel Computing",

}

TY - GEN

T1 - Performance analysis techniques for the exascale co-design process

AU - Schulz, Martin

AU - Belak, Jim

AU - Bhatele, Abhinav

AU - Bremer, Peer Timo

AU - Bronevetsky, Greg

AU - Casas, Marc

AU - Gamblin, Todd

AU - Isaacs, Katherine E.

AU - Laguna, Ignacio

AU - Levine, Joshua

AU - Pascucci, Valerio

AU - Richards, David

AU - Rountree, Barry

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Efficient and effective performance analysis techniques are critical for the development of future generation systems. They are the drivers behind the required co-design process that helps establish the principles needed for their design. In this paper, we will highlight two such approaches: PAVE, a project that investigates mapping of performance data to more intuitive domains and uses advanced visualization techniques to expose problems, and GREMLIN, a system evaluation environment capable of emulating expected properties of exascale architectures on petascale machines. Combined with other approaches in system modeling and simulation, these projects enable us to provide a meaningful introspection into a target application's characteristics as well as its expected behavior and, more importantly, likely bottlenecks on future generation machines.

AB - Efficient and effective performance analysis techniques are critical for the development of future generation systems. They are the drivers behind the required co-design process that helps establish the principles needed for their design. In this paper, we will highlight two such approaches: PAVE, a project that investigates mapping of performance data to more intuitive domains and uses advanced visualization techniques to expose problems, and GREMLIN, a system evaluation environment capable of emulating expected properties of exascale architectures on petascale machines. Combined with other approaches in system modeling and simulation, these projects enable us to provide a meaningful introspection into a target application's characteristics as well as its expected behavior and, more importantly, likely bottlenecks on future generation machines.

KW - Architecture Emulation

KW - Co-Design

KW - Performance Analysis

KW - Performance Visualization

UR - http://www.scopus.com/inward/record.url?scp=84902290240&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84902290240&partnerID=8YFLogxK

U2 - 10.3233/978-1-61499-381-0-19

DO - 10.3233/978-1-61499-381-0-19

M3 - Conference contribution

AN - SCOPUS:84902290240

SN - 9781614993803

T3 - Advances in Parallel Computing

SP - 19

EP - 32

BT - Parallel Computing

PB - IOS Press BV

ER -