loading...
 This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Mixed-Signal Approximate Computation: A Neural Predictor Case Study
January/February 2009 (vol. 29 no. 1)
pp. 104-115
ASCII Text
x 
 
Renée St. Amant, Daniel A. Jiménez, Doug Burger, "Mixed-Signal Approximate Computation: A Neural Predictor Case Study," IEEE Micro, vol. 29, no. 1, pp. 104-115, January/February, 2009.
 
 
BibTex
x
 
@article{ 10.1109/MM.2009.10,
author = {Renée St. Amant and Daniel A. Jiménez and Doug Burger},
title = {Mixed-Signal Approximate Computation: A Neural Predictor Case Study},
journal ={IEEE Micro},
volume = {29},
number = {1},
issn = {0272-1732},
year = {2009},
pages = {104-115},
doi = {http://doi.ieeecomputersociety.org/10.1109/MM.2009.10},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
}
 
 
RefWorks Procite/RefMan/Endnote
x 
 
TY - MGZN
JO - IEEE Micro
TI - Mixed-Signal Approximate Computation: A Neural Predictor Case Study
IS - 1
SN - 0272-1732
SP104
EP115
EPD - 104-115
A1 - Renée St. Amant,
A1 - Daniel A. Jiménez,
A1 - Doug Burger,
PY - 2009
KW - computer architecture
KW - low power
KW - mixed signal
KW - programmable
KW - analog circuits
KW - approximate computation
KW - imprecise
KW - neural predictor
VL - 29
JA - IEEE Micro
ER -
 
Renée St. Amant, University of Texas at Austin
Daniel A. Jiménez, University of Texas at San Antonio
Doug Burger, Microsoft Research

As transistors shrink and processors trend toward low power, maintaining precise digital behavior grows more expensive. Replacing digital units with analog equivalents sometimes allows similar computation to be performed at higher speed using less power. As a case study in mixed-signal approximate computation, the authors describe an enhanced neural prediction algorithm and its efficient analog implementation.

1. A. Seznec, "A 256 Kbits L-Tage Branch Predictor," J. Instruction-Level Parallelism (JILP), vol. 9, May 2007; http://www.jilp.org/vol9v9paper6.pdf.
2. A. Seznec, "Redundant History Skewed Perceptron Predictors: Pushing Limits on Global History Branch Predictors," IRISA tech. report 1554, 2003.
3. R. St. Amant, D.A. Jiménez, and D. Burger, "Low-Power, High-Performance Analog Neural Branch Prediction," Proc. 41st Ann. IEEE/ACM Int'l Symp. Microarchitecture (Micro 41), IEEE CS Press, 2008, pp. 447-458.
4. D.A. Johns and K. Martin, Analog Integrated Circuit Design, John Wiley &Sons, 1997.
5. D.A. Jiménez, "Piecewise Linear Branch Prediction," Proc. 32nd Ann. Int'l Symp. Computer Architecture (ISCA 32), IEEE CS Press, 2005, pp. 382-393.
6. D. Tarjan, S. Thoziyoor, and N.P. Jouppi, "Cacti 4.0," HP Laboratories Palo Alto, tech. report HPL-2006-86, 2006.
7. S. Wey and C.-L. Krishman, "Built-in Self-test (BIST) Structures for Analog Circuit Fault Diagnosis with Current Test Data," IEEE Trans. Instrumentation and Measurement, vol. 41, no. 4, Aug. 1992, pp. 535-539.
9. Y. Miura, "Real-Time Current Testing for a/d Converters," IEEE Design &Test, vol. 13, no. 2, Mar./Apr. 1996, pp. 34-41.

Index Terms:
computer architecture, low power, mixed signal, programmable, analog circuits, approximate computation, imprecise, neural predictor
Citation:
Renée St. Amant, Daniel A. Jiménez, Doug Burger, "Mixed-Signal Approximate Computation: A Neural Predictor Case Study," IEEE Micro, vol. 29, no. 1, pp. 104-115, Jan./Feb. 2009, doi:10.1109/MM.2009.10
Usage of this product signifies your acceptance of the Terms of Use.