ACM Home Page
Please provide us with feedback. Feedback
Single-vDD and single-vT super-drowsy techniques for low-leakage high-performance instruction caches
Full text PdfPdf (162 KB)
Source
International Symposium on Low Power Electronics and Design archive
Proceedings of the 2004 international symposium on Low power electronics and design table of contents
Newport Beach, California, USA
POSTER SESSION: Cache and bus design table of contents
Pages: 54 - 57  
Year of Publication: 2004
ISBN:1-58113-929-2
Authors
Nam Sung Kim  Intel Corp.
Krisztián Flautner  ARM Ltd.
David Blaauw  University of Michigan
Trevor Mudge  University of Michigan
Sponsors
ACM: Association for Computing Machinery
SIGDA: ACM Special Interest Group on Design Automation
Publisher
ACM  New York, NY, USA
Bibliometrics
Downloads (6 Weeks): 2,   Downloads (12 Months): 18,   Citation Count: 10
Additional Information:

abstract   references   cited by   index terms   collaborative colleagues  

Tools and Actions: Request Permissions Request Permissions    Review this Article  
DOI Bookmark: Use this link to bookmark this Article: http://doi.acm.org/10.1145/1013235.1013254
What is a DOI?

ABSTRACT

In this paper, we present a circuit technique that supports a super-drowsy mode with a single-V DD . In addition, we perform a detailed working set analysis for various cache line update policies for placing lines in a drowsy state. The analysis presents a policy for an instruction cache and shows it is as good as or better than more complex schemes proposed in the past. Furthermore, as an alternative to using high-threshold devices to reduce the bitline leakage through access transistors in drowsy caches, we propose a gated bitline precharge technique. A single threshold process is now sufficient. The gated precharge employs a simple but effective predictor that almost completely hides any performance loss incurred by the transitions between sub-banks. A 64-entry predictor with 3 bits per entry reduces the run-time increase by 78%, which is as effective as previous proposals that used content addressable predictors with 40 bits per entry. Overall, the combination of the proposed techniques reduces the leakage power by 72% with negligible (0.4%) run-time increase.


REFERENCES

Note: OCR errors may be found in this Reference List extracted from the full text article. ACM has opted to expose the complete List rather than only correct and linked references.

 
1
T. Douseki, et al., "A 0.5-1V MTCMOS/SIMOX SRAM macro with multi-V TH memory cells," IEEE SOI Conf., 2000.
2
 
3
4
 
5
K. Flautner, et al., "Drowsy caches," ISCA, 2002.
 
6
N. Kim, et al., "Drowsy instruction caches," MICRO, 2002.
 
7
8
9
 
10
 
11

CITED BY  10

Collaborative Colleagues:
Nam Sung Kim: colleagues
Krisztián Flautner: colleagues
David Blaauw: colleagues
Trevor Mudge: colleagues