Platform independent software analysis for near memory computing

Stefano Corda; Gagandeep Singh; Ahsan Javed Awan; Roel Jordans; Henk Corporaal

doi:10.1109/DSD.2019.00093

Platform independent software analysis for near memory computing

Stefano Corda, Gagandeep Singh, Ahsan Javed Awan, Roel Jordans, Henk Corporaal

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

6 Citations (Scopus)

75 Downloads (Pure)

Abstract

Near-memory Computing (NMC) promises improved performance for the applications that can exploit the features of emerging memory technologies such as 3D-stacked memory. However, it is not trivial to find such applications and specialized tools are needed to identify them. In this paper, we present PISA-NMC, which extends a state-of-the-art hardware agnostic profiling tool with metrics concerning memory and parallelism, which are relevant for NMC. The metrics include memory entropy, spatial locality, data-level, and basic-block-level parallelism. By profiling a set of representative applications and correlating the metrics with the application's performance on a simulated NMC system, we verify the importance of those metrics. Finally, we demonstrate which metrics are useful in identifying applications suitable for NMC architectures.

Original language	English
Title of host publication	Proceedings of the 22nd Euromicro Conference on Digital System Design
Editors	Nikos Konofaos, Paris Kitsos
Place of Publication	Piscataway
Publisher	Institute of Electrical and Electronics Engineers
Pages	606-609
Number of pages	4
ISBN (Electronic)	978-1-7281-2862-7
DOIs	https://doi.org/10.1109/DSD.2019.00093
Publication status	Published - 24 Jun 2019
Event	22nd Euromicro Conference on Digital System Design, DSD 2019 - Kallithea, Kallithea, Chalkidiki, Greece Duration: 28 Aug 2019 → 30 Aug 2019 Conference number: 22 http://dsd-seaa2019.csd.auth.gr/

Conference

Conference	22nd Euromicro Conference on Digital System Design, DSD 2019
Abbreviated title	DSD 2019
Country/Territory	Greece
City	Kallithea, Chalkidiki
Period	28/08/19 → 30/08/19
Internet address	http://dsd-seaa2019.csd.auth.gr/

Keywords

cs.PF
cs.ET
Memory Entropy
LLVM
NMC
Spatial Locality
Data-Level Parallelism

Access to Document

10.1109/DSD.2019.00093

arXiv 1906.10037v1Final author version , 475 KB

Cite this

@inproceedings{39593e33d64341609f795328ce3ac87f,

title = "Platform independent software analysis for near memory computing",

abstract = "Near-memory Computing (NMC) promises improved performance for the applications that can exploit the features of emerging memory technologies such as 3D-stacked memory. However, it is not trivial to find such applications and specialized tools are needed to identify them. In this paper, we present PISA-NMC, which extends a state-of-the-art hardware agnostic profiling tool with metrics concerning memory and parallelism, which are relevant for NMC. The metrics include memory entropy, spatial locality, data-level, and basic-block-level parallelism. By profiling a set of representative applications and correlating the metrics with the application's performance on a simulated NMC system, we verify the importance of those metrics. Finally, we demonstrate which metrics are useful in identifying applications suitable for NMC architectures. ",

keywords = "cs.PF, cs.ET, Memory Entropy, LLVM, NMC, Spatial Locality, Data-Level Parallelism",

author = "Stefano Corda and Gagandeep Singh and Awan, {Ahsan Javed} and Roel Jordans and Henk Corporaal",

year = "2019",

month = jun,

day = "24",

doi = "10.1109/DSD.2019.00093",

language = "English",

pages = "606--609",

editor = "Nikos Konofaos and Paris Kitsos",

booktitle = "Proceedings of the 22nd Euromicro Conference on Digital System Design",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "22nd Euromicro Conference on Digital System Design, DSD 2019, DSD 2019 ; Conference date: 28-08-2019 Through 30-08-2019",

url = "http://dsd-seaa2019.csd.auth.gr/",

}

Corda, S, Singh, G, Awan, AJ, Jordans, R & Corporaal, H 2019, Platform independent software analysis for near memory computing. in N Konofaos & P Kitsos (eds), Proceedings of the 22nd Euromicro Conference on Digital System Design., 8875040, Institute of Electrical and Electronics Engineers, Piscataway, pp. 606-609, 22nd Euromicro Conference on Digital System Design, DSD 2019, Kallithea, Chalkidiki, Greece, 28/08/19. https://doi.org/10.1109/DSD.2019.00093

Platform independent software analysis for near memory computing. / Corda, Stefano; Singh, Gagandeep; Awan, Ahsan Javed et al.
Proceedings of the 22nd Euromicro Conference on Digital System Design. ed. / Nikos Konofaos; Paris Kitsos. Piscataway: Institute of Electrical and Electronics Engineers, 2019. p. 606-609 8875040.

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

TY - GEN

T1 - Platform independent software analysis for near memory computing

AU - Corda, Stefano

AU - Singh, Gagandeep

AU - Awan, Ahsan Javed

AU - Jordans, Roel

AU - Corporaal, Henk

N1 - Conference code: 22

PY - 2019/6/24

Y1 - 2019/6/24

N2 - Near-memory Computing (NMC) promises improved performance for the applications that can exploit the features of emerging memory technologies such as 3D-stacked memory. However, it is not trivial to find such applications and specialized tools are needed to identify them. In this paper, we present PISA-NMC, which extends a state-of-the-art hardware agnostic profiling tool with metrics concerning memory and parallelism, which are relevant for NMC. The metrics include memory entropy, spatial locality, data-level, and basic-block-level parallelism. By profiling a set of representative applications and correlating the metrics with the application's performance on a simulated NMC system, we verify the importance of those metrics. Finally, we demonstrate which metrics are useful in identifying applications suitable for NMC architectures.

AB - Near-memory Computing (NMC) promises improved performance for the applications that can exploit the features of emerging memory technologies such as 3D-stacked memory. However, it is not trivial to find such applications and specialized tools are needed to identify them. In this paper, we present PISA-NMC, which extends a state-of-the-art hardware agnostic profiling tool with metrics concerning memory and parallelism, which are relevant for NMC. The metrics include memory entropy, spatial locality, data-level, and basic-block-level parallelism. By profiling a set of representative applications and correlating the metrics with the application's performance on a simulated NMC system, we verify the importance of those metrics. Finally, we demonstrate which metrics are useful in identifying applications suitable for NMC architectures.

KW - cs.PF

KW - cs.ET

KW - Memory Entropy

KW - LLVM

KW - NMC

KW - Spatial Locality

KW - Data-Level Parallelism

UR - http://www.scopus.com/inward/record.url?scp=85070604637&partnerID=8YFLogxK

U2 - 10.1109/DSD.2019.00093

DO - 10.1109/DSD.2019.00093

M3 - Conference contribution

SP - 606

EP - 609

BT - Proceedings of the 22nd Euromicro Conference on Digital System Design

A2 - Konofaos, Nikos

A2 - Kitsos, Paris

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

T2 - 22nd Euromicro Conference on Digital System Design, DSD 2019

Y2 - 28 August 2019 through 30 August 2019

ER -

Platform independent software analysis for near memory computing

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

NeMeCo

Cite this

Platform independent software analysis for near memory computing

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Projects

NeMeCo

Cite this