Tuyere: Enabling scalable memory workloads for system exploration

Ivy Bo Peng; Shirley V. Moore; Jeffrey S. Vetter; Seyong Lee

doi:10.1145/3208040.3208057

Tuyere: Enabling scalable memory workloads for system exploration

Ivy Bo Peng, Shirley V. Moore, Jeffrey S. Vetter, Seyong Lee

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

3 Scopus citations

Abstract

Memory technologies are under active development. Meanwhile, workloads on contemporary computing systems are increasing rapidly in size and diversity. Such dynamics in hardware and software further widen the gap between memory system design and performance evaluation. In this work, we propose a data-centric abstraction of high-performance computing applications for fast exploration of new memory technologies. We also provide a framework that uses a formal modeling language to describe the abstraction, automatically translates abstractions into memory traffic, and directly interfaces with cycle-accurate simulators. We evaluated the framework using 20 workloads and validated the memory traffic profile, the simulation results, and the relative memory changes of four memory technologies. Our results show that the data-centric abstraction can accurately capture application behavior adaptable to different input problems and can expedite system exploration.

Original language	English
Title of host publication	HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing
Publisher	Association for Computing Machinery, Inc
Pages	180-191
Number of pages	12
ISBN (Electronic)	9781450357852
DOIs	https://doi.org/10.1145/3208040.3208057
State	Published - Jun 11 2018
Event	27th ACM International Symposium on High-Performance Parallel and Distributed Computing, HPDC 2018 - Tempe, United States Duration: Jun 11 2018 → Jun 15 2018

Publication series

Name	HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing

Conference

Conference	27th ACM International Symposium on High-Performance Parallel and Distributed Computing, HPDC 2018
Country/Territory	United States
City	Tempe
Period	06/11/18 → 06/15/18

Funding

This material is based upon work supported by the U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research, under contract number DE-AC05-00OR22725 This manuscript has been co-authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, worldwide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan. This research was supported in part by an appointment to the Oak Ridge National Laboratory ASTRO Program, sponsored by the U.S. Department of Energy and administered by the Oak Ridge Institute for Science and Education.

Funders	Funder number
U.S. Department of Energy
Office of Science

Keywords

Application Abstraction
Data-centric Modeling
Memory Simulation

Access to Document

10.1145/3208040.3208057

Cite this

Peng, I. B., Moore, S. V., Vetter, J. S., & Lee, S. (2018). Tuyere: Enabling scalable memory workloads for system exploration. In HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing (pp. 180-191). (HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing). Association for Computing Machinery, Inc. https://doi.org/10.1145/3208040.3208057

Peng, Ivy Bo ; Moore, Shirley V. ; Vetter, Jeffrey S. et al. / Tuyere : Enabling scalable memory workloads for system exploration. HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing. Association for Computing Machinery, Inc, 2018. pp. 180-191 (HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing).

@inproceedings{534f67f588c440fd986d7a35333a88a5,

title = "Tuyere: Enabling scalable memory workloads for system exploration",

abstract = "Memory technologies are under active development. Meanwhile, workloads on contemporary computing systems are increasing rapidly in size and diversity. Such dynamics in hardware and software further widen the gap between memory system design and performance evaluation. In this work, we propose a data-centric abstraction of high-performance computing applications for fast exploration of new memory technologies. We also provide a framework that uses a formal modeling language to describe the abstraction, automatically translates abstractions into memory traffic, and directly interfaces with cycle-accurate simulators. We evaluated the framework using 20 workloads and validated the memory traffic profile, the simulation results, and the relative memory changes of four memory technologies. Our results show that the data-centric abstraction can accurately capture application behavior adaptable to different input problems and can expedite system exploration.",

keywords = "Application Abstraction, Data-centric Modeling, Memory Simulation",

author = "Peng, {Ivy Bo} and Moore, {Shirley V.} and Vetter, {Jeffrey S.} and Seyong Lee",

note = "Publisher Copyright: {\textcopyright} 2018 Association for Computing Machinery.; 27th ACM International Symposium on High-Performance Parallel and Distributed Computing, HPDC 2018 ; Conference date: 11-06-2018 Through 15-06-2018",

year = "2018",

month = jun,

day = "11",

doi = "10.1145/3208040.3208057",

language = "English",

series = "HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing",

publisher = "Association for Computing Machinery, Inc",

pages = "180--191",

booktitle = "HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing",

}

Peng, IB, Moore, SV, Vetter, JS & Lee, S 2018, Tuyere: Enabling scalable memory workloads for system exploration. in HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing. HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing, Association for Computing Machinery, Inc, pp. 180-191, 27th ACM International Symposium on High-Performance Parallel and Distributed Computing, HPDC 2018, Tempe, United States, 06/11/18. https://doi.org/10.1145/3208040.3208057

Tuyere: Enabling scalable memory workloads for system exploration. / Peng, Ivy Bo; Moore, Shirley V.; Vetter, Jeffrey S. et al.
HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing. Association for Computing Machinery, Inc, 2018. p. 180-191 (HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Tuyere

T2 - 27th ACM International Symposium on High-Performance Parallel and Distributed Computing, HPDC 2018

AU - Peng, Ivy Bo

AU - Moore, Shirley V.

AU - Vetter, Jeffrey S.

AU - Lee, Seyong

PY - 2018/6/11

Y1 - 2018/6/11

N2 - Memory technologies are under active development. Meanwhile, workloads on contemporary computing systems are increasing rapidly in size and diversity. Such dynamics in hardware and software further widen the gap between memory system design and performance evaluation. In this work, we propose a data-centric abstraction of high-performance computing applications for fast exploration of new memory technologies. We also provide a framework that uses a formal modeling language to describe the abstraction, automatically translates abstractions into memory traffic, and directly interfaces with cycle-accurate simulators. We evaluated the framework using 20 workloads and validated the memory traffic profile, the simulation results, and the relative memory changes of four memory technologies. Our results show that the data-centric abstraction can accurately capture application behavior adaptable to different input problems and can expedite system exploration.

AB - Memory technologies are under active development. Meanwhile, workloads on contemporary computing systems are increasing rapidly in size and diversity. Such dynamics in hardware and software further widen the gap between memory system design and performance evaluation. In this work, we propose a data-centric abstraction of high-performance computing applications for fast exploration of new memory technologies. We also provide a framework that uses a formal modeling language to describe the abstraction, automatically translates abstractions into memory traffic, and directly interfaces with cycle-accurate simulators. We evaluated the framework using 20 workloads and validated the memory traffic profile, the simulation results, and the relative memory changes of four memory technologies. Our results show that the data-centric abstraction can accurately capture application behavior adaptable to different input problems and can expedite system exploration.

KW - Application Abstraction

KW - Data-centric Modeling

KW - Memory Simulation

UR - http://www.scopus.com/inward/record.url?scp=85050129806&partnerID=8YFLogxK

U2 - 10.1145/3208040.3208057

DO - 10.1145/3208040.3208057

M3 - Conference contribution

AN - SCOPUS:85050129806

T3 - HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing

SP - 180

EP - 191

BT - HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing

PB - Association for Computing Machinery, Inc

Y2 - 11 June 2018 through 15 June 2018

ER -

Peng IB, Moore SV, Vetter JS , Lee S. Tuyere: Enabling scalable memory workloads for system exploration. In HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing. Association for Computing Machinery, Inc. 2018. p. 180-191. (HPDC 2018 - Proceedings of the 2018 International Symposium on High-Performance Parallel and Distributed Computing). doi: 10.1145/3208040.3208057

Tuyere: Enabling scalable memory workloads for system exploration

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this