The increase of recent non-recurrent engineering cost (design, mask and test cost) have made large System-on-Chip (SoC) difficult to develop especially with advanced technology. We radically explore an approach for cheap and flexible chip stacking by using Inductive coupling ThruChip Interface (TCI). In order to connect a large number of small chips for building a large scale system, novel chip stacking methods called the linear stacking and staggered stacking are proposed. They enable the system to be extended to x or/and y dimensions, not only to z dimension. Here, a novel chip staking layout, and its deadlock-free routing design for the case using single-core chips and multi-core chips are shown. The network with 256 nodes formed by the proposed stacking improves the latency of 2D mesh by 13.8% and the performance of NAS Parallel Benchmarks by 5.4% on average compared to that of 2D mesh.
Hiroshi NAKAHARA
Keio University
Tomoya OZAKI
Keio University
Hiroki MATSUTANI
Keio University
Michihiro KOIBUCHI
National Institute of Informatics
Hideharu AMANO
Keio University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Hiroshi NAKAHARA, Tomoya OZAKI, Hiroki MATSUTANI, Michihiro KOIBUCHI, Hideharu AMANO, "Novel Chip Stacking Methods to Extend Both Horizontally and Vertically for Many-Core Architectures with ThrouChip Interface" in IEICE TRANSACTIONS on Information,
vol. E99-D, no. 12, pp. 2871-2880, December 2016, doi: 10.1587/transinf.2016PAP0033.
Abstract: The increase of recent non-recurrent engineering cost (design, mask and test cost) have made large System-on-Chip (SoC) difficult to develop especially with advanced technology. We radically explore an approach for cheap and flexible chip stacking by using Inductive coupling ThruChip Interface (TCI). In order to connect a large number of small chips for building a large scale system, novel chip stacking methods called the linear stacking and staggered stacking are proposed. They enable the system to be extended to x or/and y dimensions, not only to z dimension. Here, a novel chip staking layout, and its deadlock-free routing design for the case using single-core chips and multi-core chips are shown. The network with 256 nodes formed by the proposed stacking improves the latency of 2D mesh by 13.8% and the performance of NAS Parallel Benchmarks by 5.4% on average compared to that of 2D mesh.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2016PAP0033/_p
Copy
@ARTICLE{e99-d_12_2871,
author={Hiroshi NAKAHARA, Tomoya OZAKI, Hiroki MATSUTANI, Michihiro KOIBUCHI, Hideharu AMANO, },
journal={IEICE TRANSACTIONS on Information},
title={Novel Chip Stacking Methods to Extend Both Horizontally and Vertically for Many-Core Architectures with ThrouChip Interface},
year={2016},
volume={E99-D},
number={12},
pages={2871-2880},
abstract={The increase of recent non-recurrent engineering cost (design, mask and test cost) have made large System-on-Chip (SoC) difficult to develop especially with advanced technology. We radically explore an approach for cheap and flexible chip stacking by using Inductive coupling ThruChip Interface (TCI). In order to connect a large number of small chips for building a large scale system, novel chip stacking methods called the linear stacking and staggered stacking are proposed. They enable the system to be extended to x or/and y dimensions, not only to z dimension. Here, a novel chip staking layout, and its deadlock-free routing design for the case using single-core chips and multi-core chips are shown. The network with 256 nodes formed by the proposed stacking improves the latency of 2D mesh by 13.8% and the performance of NAS Parallel Benchmarks by 5.4% on average compared to that of 2D mesh.},
keywords={},
doi={10.1587/transinf.2016PAP0033},
ISSN={1745-1361},
month={December},}
Copy
TY - JOUR
TI - Novel Chip Stacking Methods to Extend Both Horizontally and Vertically for Many-Core Architectures with ThrouChip Interface
T2 - IEICE TRANSACTIONS on Information
SP - 2871
EP - 2880
AU - Hiroshi NAKAHARA
AU - Tomoya OZAKI
AU - Hiroki MATSUTANI
AU - Michihiro KOIBUCHI
AU - Hideharu AMANO
PY - 2016
DO - 10.1587/transinf.2016PAP0033
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E99-D
IS - 12
JA - IEICE TRANSACTIONS on Information
Y1 - December 2016
AB - The increase of recent non-recurrent engineering cost (design, mask and test cost) have made large System-on-Chip (SoC) difficult to develop especially with advanced technology. We radically explore an approach for cheap and flexible chip stacking by using Inductive coupling ThruChip Interface (TCI). In order to connect a large number of small chips for building a large scale system, novel chip stacking methods called the linear stacking and staggered stacking are proposed. They enable the system to be extended to x or/and y dimensions, not only to z dimension. Here, a novel chip staking layout, and its deadlock-free routing design for the case using single-core chips and multi-core chips are shown. The network with 256 nodes formed by the proposed stacking improves the latency of 2D mesh by 13.8% and the performance of NAS Parallel Benchmarks by 5.4% on average compared to that of 2D mesh.
ER -