This paper proposes an on-chip memory-path architecture employing the dynamically variable line-size (D-VLS) cache for high performance and low energy consumption. The D-VLS cache exploits the high on-chip memory bandwidth attainable on merged DRAM/logic LSIs by replacing a whole large cache line in one cycle. At the same time, it attempts to avoid frequent evictions by decreasing the cache-line size when programs have poor spatial locality. Activating only on-chip DRAM subarrays corresponding to a replaced cache-line size produces a significant energy reduction. In our simulation, it is observed that our proposed on-chip memory-path architecture, which employs a direct-mapped D-VLS cache, improves the ED (Energy Delay) product by more than 75% over a conventional memory-path model.
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
Koji INOUE, Koji KAI, Kazuaki MURAKAMI, "A High-Performance/Low-Power On-Chip Memory-Path Architecture with Variable Cache-Line Size" in IEICE TRANSACTIONS on Electronics,
vol. E83-C, no. 11, pp. 1716-1723, November 2000, doi: .
Abstract: This paper proposes an on-chip memory-path architecture employing the dynamically variable line-size (D-VLS) cache for high performance and low energy consumption. The D-VLS cache exploits the high on-chip memory bandwidth attainable on merged DRAM/logic LSIs by replacing a whole large cache line in one cycle. At the same time, it attempts to avoid frequent evictions by decreasing the cache-line size when programs have poor spatial locality. Activating only on-chip DRAM subarrays corresponding to a replaced cache-line size produces a significant energy reduction. In our simulation, it is observed that our proposed on-chip memory-path architecture, which employs a direct-mapped D-VLS cache, improves the ED (Energy Delay) product by more than 75% over a conventional memory-path model.
URL: https://global.ieice.org/en_transactions/electronics/10.1587/e83-c_11_1716/_p
Copy
@ARTICLE{e83-c_11_1716,
author={Koji INOUE, Koji KAI, Kazuaki MURAKAMI, },
journal={IEICE TRANSACTIONS on Electronics},
title={A High-Performance/Low-Power On-Chip Memory-Path Architecture with Variable Cache-Line Size},
year={2000},
volume={E83-C},
number={11},
pages={1716-1723},
abstract={This paper proposes an on-chip memory-path architecture employing the dynamically variable line-size (D-VLS) cache for high performance and low energy consumption. The D-VLS cache exploits the high on-chip memory bandwidth attainable on merged DRAM/logic LSIs by replacing a whole large cache line in one cycle. At the same time, it attempts to avoid frequent evictions by decreasing the cache-line size when programs have poor spatial locality. Activating only on-chip DRAM subarrays corresponding to a replaced cache-line size produces a significant energy reduction. In our simulation, it is observed that our proposed on-chip memory-path architecture, which employs a direct-mapped D-VLS cache, improves the ED (Energy Delay) product by more than 75% over a conventional memory-path model.},
keywords={},
doi={},
ISSN={},
month={November},}
Copy
TY - JOUR
TI - A High-Performance/Low-Power On-Chip Memory-Path Architecture with Variable Cache-Line Size
T2 - IEICE TRANSACTIONS on Electronics
SP - 1716
EP - 1723
AU - Koji INOUE
AU - Koji KAI
AU - Kazuaki MURAKAMI
PY - 2000
DO -
JO - IEICE TRANSACTIONS on Electronics
SN -
VL - E83-C
IS - 11
JA - IEICE TRANSACTIONS on Electronics
Y1 - November 2000
AB - This paper proposes an on-chip memory-path architecture employing the dynamically variable line-size (D-VLS) cache for high performance and low energy consumption. The D-VLS cache exploits the high on-chip memory bandwidth attainable on merged DRAM/logic LSIs by replacing a whole large cache line in one cycle. At the same time, it attempts to avoid frequent evictions by decreasing the cache-line size when programs have poor spatial locality. Activating only on-chip DRAM subarrays corresponding to a replaced cache-line size produces a significant energy reduction. In our simulation, it is observed that our proposed on-chip memory-path architecture, which employs a direct-mapped D-VLS cache, improves the ED (Energy Delay) product by more than 75% over a conventional memory-path model.
ER -