Scaling energy per operation via an asynchronous pipeline

Authors:
Bo Marr;Brian Degnan;Paul Hasler;David Anderson
Affiliations:
Raytheon Company, El Segundo, CA;School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA;School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA;School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA
Venue:
IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Year:
2013

Citing 10
Cited 1

Self-timed rings and their application to division

Self-timed rings and their application to division
An analytical MOS transistor model valid in all regions of operation and dedicated to low-voltage and low-current applications

Analog Integrated Circuits and Signal Processing - Special issue: low-voltage low-power analog integrated circuits
Analog VLSI: Circuits and Principles

Analog VLSI: Circuits and Principles
How to Achieve Worst-Case Performance

ASYNC '01 Proceedings of the 7th International Symposium on Asynchronous Circuits and Systems
Energy-Efficient Pipelines

ASYNC '02 Proceedings of the 8th International Symposium on Asynchronus Circuits and Systems
Computer Architecture: A Quantitative Approach

Computer Architecture: A Quantitative Approach
Ultralow-voltage, minimum-energy CMOS

IBM Journal of Research and Development - Advanced silicon technology
The design of high-performance dynamic asynchronous pipelines: high-capacity style

IEEE Transactions on Very Large Scale Integration (VLSI) Systems
Extending Amdahl's Law for Energy-Efficient Computing in the Many-Core Era

Computer
Circuit techniques for dynamic variation tolerance

Proceedings of the 46th Annual Design Automation Conference

2013 Special Issue: Computing with networks of spiking neurons on a biophysically motivated floating-gate based neuromorphic integrated circuit

Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

Statistical analysis of computations per unit energy in processors over the last 30 years is given that illustrates a sharp reduction in the rate of energy efficiency improvements over the last several years resulting in the formation of an asymptotic "wall" with our dataset; we use the measure of giga multiply accumulates per Joule. We have developed an energy model which takes into account the realities of scaling, specifically for asynchronous systems. Studies of an energy efficient asynchronous pipeline show fabricated results of 17 Giga Operations per Joule in 0.6 µm at subthreshold when fully pipelined, and simulations at a more modern 65 nm process show a further order of magnitude improvement on that.