Manifold

A parallel simulation framework for multicore systems

Jun Wang, Jesse Beu, Rishiraj Bheda, Tom Conte, Zhenjiang Dong, Chad Kersey, Mitchelle Rasquinha, George Riley, William Jinho Song, He Xiao, Peng Xu, Sudhakar Yalamanchili

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

40 Citations (Scopus)

Abstract

This paper presents Manifold, an open-source parallel simulation framework for multicore architectures. It consists of a parallel simulation kernel, a set of microarchitecture components, and an integrated library of power, thermal, reliability, and energy models. Using the components as building blocks, users can assemble multicore architecture simulation models and perform serial or parallel simulations to study the architectural and/or the physical characteristics of the models. Users can also create new components for Manifold or port existing models. Importantly, Manifold's component-based design provides the user with the ability to easily replace a component with another for efficient explorations of the design space. It also allows components to evolve independently and making it easy for simulators to incorporate new components as they become available. The distinguishing features of Manifold include i) transparent parallel execution, ii) integration of power, thermal, reliability, and energy models, iii) full system simulation, e.g., operating system and system binaries, and iv) component-based design. In this paper we provide a description of the software architecture of Manifold, and its main elements - a parallel multicore emulator front-end and a parallel component-based back-end timing model. We describe a few simulators that are built with Manifold components to illustrate its flexibility, and present test results of the scalability obtained on full-system simulation of coherent shared-memory multicore models with 16, 32, and 64 cores executing PARSEC and SPLASH-2 benchmarks.

Original languageEnglish
Title of host publicationISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software
PublisherIEEE Computer Society
Pages106-115
Number of pages10
ISBN (Print)9781479936052
DOIs
Publication statusPublished - 2014 Jan 1
Event2014 IEEE International Symposium on Performance Analysis of Systems and Software, ISPASS 2014 - Monterey, CA, United States
Duration: 2014 Mar 232014 Mar 25

Publication series

NameISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software

Other

Other2014 IEEE International Symposium on Performance Analysis of Systems and Software, ISPASS 2014
CountryUnited States
CityMonterey, CA
Period14/3/2314/3/25

Fingerprint

Simulators
Software architecture
Scalability
Data storage equipment
Hot Temperature

All Science Journal Classification (ASJC) codes

  • Software

Cite this

Wang, J., Beu, J., Bheda, R., Conte, T., Dong, Z., Kersey, C., ... Yalamanchili, S. (2014). Manifold: A parallel simulation framework for multicore systems. In ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software (pp. 106-115). [6844466] (ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software). IEEE Computer Society. https://doi.org/10.1109/ISPASS.2014.6844466
Wang, Jun ; Beu, Jesse ; Bheda, Rishiraj ; Conte, Tom ; Dong, Zhenjiang ; Kersey, Chad ; Rasquinha, Mitchelle ; Riley, George ; Song, William Jinho ; Xiao, He ; Xu, Peng ; Yalamanchili, Sudhakar. / Manifold : A parallel simulation framework for multicore systems. ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software. IEEE Computer Society, 2014. pp. 106-115 (ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software).
@inproceedings{bcfbe24c1bc74abea262ab22dfd889a0,
title = "Manifold: A parallel simulation framework for multicore systems",
abstract = "This paper presents Manifold, an open-source parallel simulation framework for multicore architectures. It consists of a parallel simulation kernel, a set of microarchitecture components, and an integrated library of power, thermal, reliability, and energy models. Using the components as building blocks, users can assemble multicore architecture simulation models and perform serial or parallel simulations to study the architectural and/or the physical characteristics of the models. Users can also create new components for Manifold or port existing models. Importantly, Manifold's component-based design provides the user with the ability to easily replace a component with another for efficient explorations of the design space. It also allows components to evolve independently and making it easy for simulators to incorporate new components as they become available. The distinguishing features of Manifold include i) transparent parallel execution, ii) integration of power, thermal, reliability, and energy models, iii) full system simulation, e.g., operating system and system binaries, and iv) component-based design. In this paper we provide a description of the software architecture of Manifold, and its main elements - a parallel multicore emulator front-end and a parallel component-based back-end timing model. We describe a few simulators that are built with Manifold components to illustrate its flexibility, and present test results of the scalability obtained on full-system simulation of coherent shared-memory multicore models with 16, 32, and 64 cores executing PARSEC and SPLASH-2 benchmarks.",
author = "Jun Wang and Jesse Beu and Rishiraj Bheda and Tom Conte and Zhenjiang Dong and Chad Kersey and Mitchelle Rasquinha and George Riley and Song, {William Jinho} and He Xiao and Peng Xu and Sudhakar Yalamanchili",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/ISPASS.2014.6844466",
language = "English",
isbn = "9781479936052",
series = "ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software",
publisher = "IEEE Computer Society",
pages = "106--115",
booktitle = "ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software",
address = "United States",

}

Wang, J, Beu, J, Bheda, R, Conte, T, Dong, Z, Kersey, C, Rasquinha, M, Riley, G, Song, WJ, Xiao, H, Xu, P & Yalamanchili, S 2014, Manifold: A parallel simulation framework for multicore systems. in ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software., 6844466, ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software, IEEE Computer Society, pp. 106-115, 2014 IEEE International Symposium on Performance Analysis of Systems and Software, ISPASS 2014, Monterey, CA, United States, 14/3/23. https://doi.org/10.1109/ISPASS.2014.6844466

Manifold : A parallel simulation framework for multicore systems. / Wang, Jun; Beu, Jesse; Bheda, Rishiraj; Conte, Tom; Dong, Zhenjiang; Kersey, Chad; Rasquinha, Mitchelle; Riley, George; Song, William Jinho; Xiao, He; Xu, Peng; Yalamanchili, Sudhakar.

ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software. IEEE Computer Society, 2014. p. 106-115 6844466 (ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software).

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

TY - GEN

T1 - Manifold

T2 - A parallel simulation framework for multicore systems

AU - Wang, Jun

AU - Beu, Jesse

AU - Bheda, Rishiraj

AU - Conte, Tom

AU - Dong, Zhenjiang

AU - Kersey, Chad

AU - Rasquinha, Mitchelle

AU - Riley, George

AU - Song, William Jinho

AU - Xiao, He

AU - Xu, Peng

AU - Yalamanchili, Sudhakar

PY - 2014/1/1

Y1 - 2014/1/1

N2 - This paper presents Manifold, an open-source parallel simulation framework for multicore architectures. It consists of a parallel simulation kernel, a set of microarchitecture components, and an integrated library of power, thermal, reliability, and energy models. Using the components as building blocks, users can assemble multicore architecture simulation models and perform serial or parallel simulations to study the architectural and/or the physical characteristics of the models. Users can also create new components for Manifold or port existing models. Importantly, Manifold's component-based design provides the user with the ability to easily replace a component with another for efficient explorations of the design space. It also allows components to evolve independently and making it easy for simulators to incorporate new components as they become available. The distinguishing features of Manifold include i) transparent parallel execution, ii) integration of power, thermal, reliability, and energy models, iii) full system simulation, e.g., operating system and system binaries, and iv) component-based design. In this paper we provide a description of the software architecture of Manifold, and its main elements - a parallel multicore emulator front-end and a parallel component-based back-end timing model. We describe a few simulators that are built with Manifold components to illustrate its flexibility, and present test results of the scalability obtained on full-system simulation of coherent shared-memory multicore models with 16, 32, and 64 cores executing PARSEC and SPLASH-2 benchmarks.

AB - This paper presents Manifold, an open-source parallel simulation framework for multicore architectures. It consists of a parallel simulation kernel, a set of microarchitecture components, and an integrated library of power, thermal, reliability, and energy models. Using the components as building blocks, users can assemble multicore architecture simulation models and perform serial or parallel simulations to study the architectural and/or the physical characteristics of the models. Users can also create new components for Manifold or port existing models. Importantly, Manifold's component-based design provides the user with the ability to easily replace a component with another for efficient explorations of the design space. It also allows components to evolve independently and making it easy for simulators to incorporate new components as they become available. The distinguishing features of Manifold include i) transparent parallel execution, ii) integration of power, thermal, reliability, and energy models, iii) full system simulation, e.g., operating system and system binaries, and iv) component-based design. In this paper we provide a description of the software architecture of Manifold, and its main elements - a parallel multicore emulator front-end and a parallel component-based back-end timing model. We describe a few simulators that are built with Manifold components to illustrate its flexibility, and present test results of the scalability obtained on full-system simulation of coherent shared-memory multicore models with 16, 32, and 64 cores executing PARSEC and SPLASH-2 benchmarks.

UR - http://www.scopus.com/inward/record.url?scp=84904504588&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84904504588&partnerID=8YFLogxK

U2 - 10.1109/ISPASS.2014.6844466

DO - 10.1109/ISPASS.2014.6844466

M3 - Conference contribution

SN - 9781479936052

T3 - ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software

SP - 106

EP - 115

BT - ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software

PB - IEEE Computer Society

ER -

Wang J, Beu J, Bheda R, Conte T, Dong Z, Kersey C et al. Manifold: A parallel simulation framework for multicore systems. In ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software. IEEE Computer Society. 2014. p. 106-115. 6844466. (ISPASS 2014 - IEEE International Symposium on Performance Analysis of Systems and Software). https://doi.org/10.1109/ISPASS.2014.6844466