A scalable 3D heterogeneous multicore with an inductive ThruChip interface

Noriyuki Miura, Yusuke Koizumi, Yasuhiro Take, Hiroki Matsutani, Tadahiro Kuroda, Hideharu Amano, Ryuichi Sakamoto, Mitaro Namiki, Kimiyoshi Usami, Masaaki Kondo, Hiroshi Nakamura

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

The authors developed a scalable heterogeneous multicore processor. 3D heterogeneous chip stacking of a general-purpose CPU and reconfigurable multicore accelerators enables various trade-offs between performance and energy consumption. The stacked chips interconnect through a scalable 3D network on a chip (NoC). By simply changing the number of stacked accelerator chips, processor parallelism can be widely scaled. No design change is needed, and hence, no additional nonrecurring engineering (NRE) cost is required. An inductive-coupling ThruChip Interface (TCI) is applied to stacked-chip communications, forming a low-cost and robust high-speed 3D NoC. The authors developed a prototype system called Cube-1 with 65-nm CMOS test chips, and confirmed successful system operations, including 10 hours of continuous Linux OS operation. Simple filters and a streaming application were implemented on Cube-1 and performance acceleration up to about three times was achieved.

Original languageEnglish
Article number6684194
Pages (from-to)6-15
Number of pages10
JournalIEEE Micro
Volume33
Issue number6
DOIs
Publication statusPublished - 2013 Nov

Fingerprint

Particle accelerators
Program processors
Costs
Energy utilization
Communication
Linux

Keywords

  • Heterogeneous multicore system
  • Inductive coupling through chip interface
  • Network-on-chip

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Hardware and Architecture
  • Software

Cite this

Miura, N., Koizumi, Y., Take, Y., Matsutani, H., Kuroda, T., Amano, H., ... Nakamura, H. (2013). A scalable 3D heterogeneous multicore with an inductive ThruChip interface. IEEE Micro, 33(6), 6-15. [6684194]. https://doi.org/10.1109/MM.2013.112

A scalable 3D heterogeneous multicore with an inductive ThruChip interface. / Miura, Noriyuki; Koizumi, Yusuke; Take, Yasuhiro; Matsutani, Hiroki; Kuroda, Tadahiro; Amano, Hideharu; Sakamoto, Ryuichi; Namiki, Mitaro; Usami, Kimiyoshi; Kondo, Masaaki; Nakamura, Hiroshi.

In: IEEE Micro, Vol. 33, No. 6, 6684194, 11.2013, p. 6-15.

Research output: Contribution to journalArticle

Miura, N, Koizumi, Y, Take, Y, Matsutani, H, Kuroda, T, Amano, H, Sakamoto, R, Namiki, M, Usami, K, Kondo, M & Nakamura, H 2013, 'A scalable 3D heterogeneous multicore with an inductive ThruChip interface', IEEE Micro, vol. 33, no. 6, 6684194, pp. 6-15. https://doi.org/10.1109/MM.2013.112
Miura N, Koizumi Y, Take Y, Matsutani H, Kuroda T, Amano H et al. A scalable 3D heterogeneous multicore with an inductive ThruChip interface. IEEE Micro. 2013 Nov;33(6):6-15. 6684194. https://doi.org/10.1109/MM.2013.112
Miura, Noriyuki ; Koizumi, Yusuke ; Take, Yasuhiro ; Matsutani, Hiroki ; Kuroda, Tadahiro ; Amano, Hideharu ; Sakamoto, Ryuichi ; Namiki, Mitaro ; Usami, Kimiyoshi ; Kondo, Masaaki ; Nakamura, Hiroshi. / A scalable 3D heterogeneous multicore with an inductive ThruChip interface. In: IEEE Micro. 2013 ; Vol. 33, No. 6. pp. 6-15.
@article{06a55469660f40378856be6b3a1f3b26,
title = "A scalable 3D heterogeneous multicore with an inductive ThruChip interface",
abstract = "The authors developed a scalable heterogeneous multicore processor. 3D heterogeneous chip stacking of a general-purpose CPU and reconfigurable multicore accelerators enables various trade-offs between performance and energy consumption. The stacked chips interconnect through a scalable 3D network on a chip (NoC). By simply changing the number of stacked accelerator chips, processor parallelism can be widely scaled. No design change is needed, and hence, no additional nonrecurring engineering (NRE) cost is required. An inductive-coupling ThruChip Interface (TCI) is applied to stacked-chip communications, forming a low-cost and robust high-speed 3D NoC. The authors developed a prototype system called Cube-1 with 65-nm CMOS test chips, and confirmed successful system operations, including 10 hours of continuous Linux OS operation. Simple filters and a streaming application were implemented on Cube-1 and performance acceleration up to about three times was achieved.",
keywords = "Heterogeneous multicore system, Inductive coupling through chip interface, Network-on-chip",
author = "Noriyuki Miura and Yusuke Koizumi and Yasuhiro Take and Hiroki Matsutani and Tadahiro Kuroda and Hideharu Amano and Ryuichi Sakamoto and Mitaro Namiki and Kimiyoshi Usami and Masaaki Kondo and Hiroshi Nakamura",
year = "2013",
month = "11",
doi = "10.1109/MM.2013.112",
language = "English",
volume = "33",
pages = "6--15",
journal = "IEEE Micro",
issn = "0272-1732",
publisher = "IEEE Computer Society",
number = "6",

}

TY - JOUR

T1 - A scalable 3D heterogeneous multicore with an inductive ThruChip interface

AU - Miura, Noriyuki

AU - Koizumi, Yusuke

AU - Take, Yasuhiro

AU - Matsutani, Hiroki

AU - Kuroda, Tadahiro

AU - Amano, Hideharu

AU - Sakamoto, Ryuichi

AU - Namiki, Mitaro

AU - Usami, Kimiyoshi

AU - Kondo, Masaaki

AU - Nakamura, Hiroshi

PY - 2013/11

Y1 - 2013/11

N2 - The authors developed a scalable heterogeneous multicore processor. 3D heterogeneous chip stacking of a general-purpose CPU and reconfigurable multicore accelerators enables various trade-offs between performance and energy consumption. The stacked chips interconnect through a scalable 3D network on a chip (NoC). By simply changing the number of stacked accelerator chips, processor parallelism can be widely scaled. No design change is needed, and hence, no additional nonrecurring engineering (NRE) cost is required. An inductive-coupling ThruChip Interface (TCI) is applied to stacked-chip communications, forming a low-cost and robust high-speed 3D NoC. The authors developed a prototype system called Cube-1 with 65-nm CMOS test chips, and confirmed successful system operations, including 10 hours of continuous Linux OS operation. Simple filters and a streaming application were implemented on Cube-1 and performance acceleration up to about three times was achieved.

AB - The authors developed a scalable heterogeneous multicore processor. 3D heterogeneous chip stacking of a general-purpose CPU and reconfigurable multicore accelerators enables various trade-offs between performance and energy consumption. The stacked chips interconnect through a scalable 3D network on a chip (NoC). By simply changing the number of stacked accelerator chips, processor parallelism can be widely scaled. No design change is needed, and hence, no additional nonrecurring engineering (NRE) cost is required. An inductive-coupling ThruChip Interface (TCI) is applied to stacked-chip communications, forming a low-cost and robust high-speed 3D NoC. The authors developed a prototype system called Cube-1 with 65-nm CMOS test chips, and confirmed successful system operations, including 10 hours of continuous Linux OS operation. Simple filters and a streaming application were implemented on Cube-1 and performance acceleration up to about three times was achieved.

KW - Heterogeneous multicore system

KW - Inductive coupling through chip interface

KW - Network-on-chip

UR - http://www.scopus.com/inward/record.url?scp=84890940213&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84890940213&partnerID=8YFLogxK

U2 - 10.1109/MM.2013.112

DO - 10.1109/MM.2013.112

M3 - Article

AN - SCOPUS:84890940213

VL - 33

SP - 6

EP - 15

JO - IEEE Micro

JF - IEEE Micro

SN - 0272-1732

IS - 6

M1 - 6684194

ER -