Designing vertical processors in monolithic 3D

Bhargava Gopireddy, Josep Torrellas

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

Abstract

A processor laid out vertically in stacked layers can benefit from reduced wire delays, low energy consumption, and a small footprint. Such a design can be enabled by Monolithic 3D (M3D), a technology that provides short wire lengths, good thermal properties, and high integration. In current M3D technology, due to manufacturing constraints, the layers in the stack are asymmetric: the bottom-most one has a relatively higher performance. In this paper, we examine how to partition a processor for M3D. We partition logic and storage structures into two layers, taking into account that the top layer has lower-performance transistors. In logic structures, we place the critical paths in the bottom layer. In storage structures, we partition the hardware unequally, assigning to the top layer fewer ports with larger access transistors, or a shorter bitcell subarray with larger bitcells. We find that, with conservative assumptions on M3D technology, an M3D core executes applications on average 25% faster than a 2D core, while consuming 39% less energy. With aggressive technology assumptions, the M3D core performs even better: it is on average 38% faster than a 2D core and consumes 41% less energy. Further, under a similar power budget, an M3D multicore can use twice as many cores as a 2D multicore, executing applications on average 92% faster with 39% less energy. Finally, an M3D core is thermally efficient.

Original languageEnglish (US)
Title of host publicationISCA 2019 - Proceedings of the 2019 46th International Symposium on Computer Architecture
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages643-656
Number of pages14
ISBN (Electronic)9781450366694
DOIs
StatePublished - Jun 22 2019
Event46th International Symposium on Computer Architecture, ISCA 2019 - Phoenix, United States
Duration: Jun 22 2019Jun 26 2019

Publication series

NameProceedings - International Symposium on Computer Architecture
ISSN (Print)1063-6897

Conference

Conference46th International Symposium on Computer Architecture, ISCA 2019
CountryUnited States
CityPhoenix
Period6/22/196/26/19

Fingerprint

Transistors
Wire
Thermodynamic properties
Energy utilization
Hardware

Keywords

  • 3D integration
  • Monolithic 3D
  • Processor architecture

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

Gopireddy, B., & Torrellas, J. (2019). Designing vertical processors in monolithic 3D. In ISCA 2019 - Proceedings of the 2019 46th International Symposium on Computer Architecture (pp. 643-656). (Proceedings - International Symposium on Computer Architecture). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1145/3307650.3322233

Designing vertical processors in monolithic 3D. / Gopireddy, Bhargava; Torrellas, Josep.

ISCA 2019 - Proceedings of the 2019 46th International Symposium on Computer Architecture. Institute of Electrical and Electronics Engineers Inc., 2019. p. 643-656 (Proceedings - International Symposium on Computer Architecture).

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

Gopireddy, B & Torrellas, J 2019, Designing vertical processors in monolithic 3D. in ISCA 2019 - Proceedings of the 2019 46th International Symposium on Computer Architecture. Proceedings - International Symposium on Computer Architecture, Institute of Electrical and Electronics Engineers Inc., pp. 643-656, 46th International Symposium on Computer Architecture, ISCA 2019, Phoenix, United States, 6/22/19. https://doi.org/10.1145/3307650.3322233
Gopireddy B, Torrellas J. Designing vertical processors in monolithic 3D. In ISCA 2019 - Proceedings of the 2019 46th International Symposium on Computer Architecture. Institute of Electrical and Electronics Engineers Inc. 2019. p. 643-656. (Proceedings - International Symposium on Computer Architecture). https://doi.org/10.1145/3307650.3322233
Gopireddy, Bhargava ; Torrellas, Josep. / Designing vertical processors in monolithic 3D. ISCA 2019 - Proceedings of the 2019 46th International Symposium on Computer Architecture. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 643-656 (Proceedings - International Symposium on Computer Architecture).
@inproceedings{95ecadce07f44b45b7b309a3643cc9c6,
title = "Designing vertical processors in monolithic 3D",
abstract = "A processor laid out vertically in stacked layers can benefit from reduced wire delays, low energy consumption, and a small footprint. Such a design can be enabled by Monolithic 3D (M3D), a technology that provides short wire lengths, good thermal properties, and high integration. In current M3D technology, due to manufacturing constraints, the layers in the stack are asymmetric: the bottom-most one has a relatively higher performance. In this paper, we examine how to partition a processor for M3D. We partition logic and storage structures into two layers, taking into account that the top layer has lower-performance transistors. In logic structures, we place the critical paths in the bottom layer. In storage structures, we partition the hardware unequally, assigning to the top layer fewer ports with larger access transistors, or a shorter bitcell subarray with larger bitcells. We find that, with conservative assumptions on M3D technology, an M3D core executes applications on average 25{\%} faster than a 2D core, while consuming 39{\%} less energy. With aggressive technology assumptions, the M3D core performs even better: it is on average 38{\%} faster than a 2D core and consumes 41{\%} less energy. Further, under a similar power budget, an M3D multicore can use twice as many cores as a 2D multicore, executing applications on average 92{\%} faster with 39{\%} less energy. Finally, an M3D core is thermally efficient.",
keywords = "3D integration, Monolithic 3D, Processor architecture",
author = "Bhargava Gopireddy and Josep Torrellas",
year = "2019",
month = "6",
day = "22",
doi = "10.1145/3307650.3322233",
language = "English (US)",
series = "Proceedings - International Symposium on Computer Architecture",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "643--656",
booktitle = "ISCA 2019 - Proceedings of the 2019 46th International Symposium on Computer Architecture",
address = "United States",

}

TY - GEN

T1 - Designing vertical processors in monolithic 3D

AU - Gopireddy, Bhargava

AU - Torrellas, Josep

PY - 2019/6/22

Y1 - 2019/6/22

N2 - A processor laid out vertically in stacked layers can benefit from reduced wire delays, low energy consumption, and a small footprint. Such a design can be enabled by Monolithic 3D (M3D), a technology that provides short wire lengths, good thermal properties, and high integration. In current M3D technology, due to manufacturing constraints, the layers in the stack are asymmetric: the bottom-most one has a relatively higher performance. In this paper, we examine how to partition a processor for M3D. We partition logic and storage structures into two layers, taking into account that the top layer has lower-performance transistors. In logic structures, we place the critical paths in the bottom layer. In storage structures, we partition the hardware unequally, assigning to the top layer fewer ports with larger access transistors, or a shorter bitcell subarray with larger bitcells. We find that, with conservative assumptions on M3D technology, an M3D core executes applications on average 25% faster than a 2D core, while consuming 39% less energy. With aggressive technology assumptions, the M3D core performs even better: it is on average 38% faster than a 2D core and consumes 41% less energy. Further, under a similar power budget, an M3D multicore can use twice as many cores as a 2D multicore, executing applications on average 92% faster with 39% less energy. Finally, an M3D core is thermally efficient.

AB - A processor laid out vertically in stacked layers can benefit from reduced wire delays, low energy consumption, and a small footprint. Such a design can be enabled by Monolithic 3D (M3D), a technology that provides short wire lengths, good thermal properties, and high integration. In current M3D technology, due to manufacturing constraints, the layers in the stack are asymmetric: the bottom-most one has a relatively higher performance. In this paper, we examine how to partition a processor for M3D. We partition logic and storage structures into two layers, taking into account that the top layer has lower-performance transistors. In logic structures, we place the critical paths in the bottom layer. In storage structures, we partition the hardware unequally, assigning to the top layer fewer ports with larger access transistors, or a shorter bitcell subarray with larger bitcells. We find that, with conservative assumptions on M3D technology, an M3D core executes applications on average 25% faster than a 2D core, while consuming 39% less energy. With aggressive technology assumptions, the M3D core performs even better: it is on average 38% faster than a 2D core and consumes 41% less energy. Further, under a similar power budget, an M3D multicore can use twice as many cores as a 2D multicore, executing applications on average 92% faster with 39% less energy. Finally, an M3D core is thermally efficient.

KW - 3D integration

KW - Monolithic 3D

KW - Processor architecture

UR - http://www.scopus.com/inward/record.url?scp=85069477550&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069477550&partnerID=8YFLogxK

U2 - 10.1145/3307650.3322233

DO - 10.1145/3307650.3322233

M3 - Conference contribution

AN - SCOPUS:85069477550

T3 - Proceedings - International Symposium on Computer Architecture

SP - 643

EP - 656

BT - ISCA 2019 - Proceedings of the 2019 46th International Symposium on Computer Architecture

PB - Institute of Electrical and Electronics Engineers Inc.

ER -