Giraf: General purpose in-storage resistive associative framework

Leonid Yavits; Roman Kaplan; Ran Ginosar

doi:10.1109/PACT.2019.00053

Giraf: General purpose in-storage resistive associative framework

Leonid Yavits, Roman Kaplan, Ran Ginosar

Technion-Israel Institute of Technology

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

Abstract

GIRAF is an in-storage architecture and algorithm framework based on Resistive Content Addressable Memory (RCAM). GIRAF functions simultaneously as a storage and a massively parallel associative processor. GIRAF alleviates the bandwidth wall by connecting every memory bit to processing transistors and keeping computing inside the storage arrays, thus implementing in-data, rather than near-data, processing. We show that GIRAF outperforms a reference computer architecture with a bandwidth-limited external storage access on a variety of data intensive workloads. The performance of GIRAF Euclidean distance, dot product and histogram implementation, exceeds the attainable performance of a reference architecture by up to four orders of magnitude, depending on the dataset size. The performance of GIRAF SpMV exceeds the attainable performance of such reference architecture by more than two orders of magnitude.

Original language	English
Title of host publication	Proceedings - 2019 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	476-477
Number of pages	2
ISBN (Electronic)	9781728136134
DOIs	https://doi.org/10.1109/PACT.2019.00053
State	Published - Sep 2019
Externally published	Yes
Event	28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019 - Seattle, United States Duration: 21 Sep 2019 → 25 Sep 2019

Publication series

Name	Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT
Volume	2019-September
ISSN (Print)	1089-795X

Conference

Conference	28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019
Country/Territory	United States
City	Seattle
Period	21/09/19 → 25/09/19

Bibliographical note

Publisher Copyright:
© 2019 IEEE.

Keywords

processing in meemory
processing in storage
resistive memory

Access to Document

10.1109/PACT.2019.00053

Cite this

Yavits, L., Kaplan, R., & Ginosar, R. (2019). Giraf: General purpose in-storage resistive associative framework. In Proceedings - 2019 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019 (pp. 476-477). Article 8891629 (Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT; Vol. 2019-September). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PACT.2019.00053

Yavits, Leonid ; Kaplan, Roman ; Ginosar, Ran. / Giraf : General purpose in-storage resistive associative framework. Proceedings - 2019 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 476-477 (Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT).

@inproceedings{9e1e5780f5d2420691cc3d8c0b5d8b82,

title = "Giraf: General purpose in-storage resistive associative framework",

abstract = "GIRAF is an in-storage architecture and algorithm framework based on Resistive Content Addressable Memory (RCAM). GIRAF functions simultaneously as a storage and a massively parallel associative processor. GIRAF alleviates the bandwidth wall by connecting every memory bit to processing transistors and keeping computing inside the storage arrays, thus implementing in-data, rather than near-data, processing. We show that GIRAF outperforms a reference computer architecture with a bandwidth-limited external storage access on a variety of data intensive workloads. The performance of GIRAF Euclidean distance, dot product and histogram implementation, exceeds the attainable performance of a reference architecture by up to four orders of magnitude, depending on the dataset size. The performance of GIRAF SpMV exceeds the attainable performance of such reference architecture by more than two orders of magnitude.",

keywords = "processing in meemory, processing in storage, resistive memory",

author = "Leonid Yavits and Roman Kaplan and Ran Ginosar",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019 ; Conference date: 21-09-2019 Through 25-09-2019",

year = "2019",

month = sep,

doi = "10.1109/PACT.2019.00053",

language = "אנגלית",

series = "Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "476--477",

booktitle = "Proceedings - 2019 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019",

address = "ארצות הברית",

}

Yavits, L, Kaplan, R & Ginosar, R 2019, Giraf: General purpose in-storage resistive associative framework. in Proceedings - 2019 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019., 8891629, Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT, vol. 2019-September, Institute of Electrical and Electronics Engineers Inc., pp. 476-477, 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019, Seattle, United States, 21/09/19. https://doi.org/10.1109/PACT.2019.00053

Giraf: General purpose in-storage resistive associative framework. / Yavits, Leonid; Kaplan, Roman; Ginosar, Ran.
Proceedings - 2019 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 476-477 8891629 (Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT; Vol. 2019-September).

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

TY - GEN

T1 - Giraf

T2 - 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019

AU - Yavits, Leonid

AU - Kaplan, Roman

AU - Ginosar, Ran

PY - 2019/9

Y1 - 2019/9

N2 - GIRAF is an in-storage architecture and algorithm framework based on Resistive Content Addressable Memory (RCAM). GIRAF functions simultaneously as a storage and a massively parallel associative processor. GIRAF alleviates the bandwidth wall by connecting every memory bit to processing transistors and keeping computing inside the storage arrays, thus implementing in-data, rather than near-data, processing. We show that GIRAF outperforms a reference computer architecture with a bandwidth-limited external storage access on a variety of data intensive workloads. The performance of GIRAF Euclidean distance, dot product and histogram implementation, exceeds the attainable performance of a reference architecture by up to four orders of magnitude, depending on the dataset size. The performance of GIRAF SpMV exceeds the attainable performance of such reference architecture by more than two orders of magnitude.

AB - GIRAF is an in-storage architecture and algorithm framework based on Resistive Content Addressable Memory (RCAM). GIRAF functions simultaneously as a storage and a massively parallel associative processor. GIRAF alleviates the bandwidth wall by connecting every memory bit to processing transistors and keeping computing inside the storage arrays, thus implementing in-data, rather than near-data, processing. We show that GIRAF outperforms a reference computer architecture with a bandwidth-limited external storage access on a variety of data intensive workloads. The performance of GIRAF Euclidean distance, dot product and histogram implementation, exceeds the attainable performance of a reference architecture by up to four orders of magnitude, depending on the dataset size. The performance of GIRAF SpMV exceeds the attainable performance of such reference architecture by more than two orders of magnitude.

KW - processing in meemory

KW - processing in storage

KW - resistive memory

UR - http://www.scopus.com/inward/record.url?scp=85075438487&partnerID=8YFLogxK

U2 - 10.1109/PACT.2019.00053

DO - 10.1109/PACT.2019.00053

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:85075438487

T3 - Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT

SP - 476

EP - 477

BT - Proceedings - 2019 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 21 September 2019 through 25 September 2019

ER -

Yavits L, Kaplan R, Ginosar R. Giraf: General purpose in-storage resistive associative framework. In Proceedings - 2019 28th International Conference on Parallel Architectures and Compilation Techniques, PACT 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 476-477. 8891629. (Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT). doi: 10.1109/PACT.2019.00053

Giraf: General purpose in-storage resistive associative framework

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this