Corpus ID: 12654268

ASSESSING TRENDS IN THE ELECTRICAL EFFICIENCY OF COMPUTATION OVER TIME

@inproceedings{Koomey2009ASSESSINGTI,
  title={ASSESSING TRENDS IN THE ELECTRICAL EFFICIENCY OF COMPUTATION OVER TIME},
  author={Jonathan G. Koomey and Stephen Berard and Marla Sanchez and Henry Wong},
  year={2009}
}
Information technology (IT) has captured the popular imagination, in part because of the tangible benefits IT brings, but also because the underlying technological trends proceed at easily measurable, remarkably predictable, and unusually rapid rates. The number of transistors on a chip has doubled more or less every two years for decades, a trend that is popularly (but often imprecisely) encapsulated as " Moore's law ". This article explores the relationship between the performance of… Expand
Efficient LU Factorization for Texas Instruments Keystone Architecture Digital Signal Processors
TLDR
This work aims to provide a high-quality tool for the HPC industry standard high-performance LINPACK benchmark (HPL) for the eight-core Texas Instruments TMS320C6678 digitalsignal processor (DSP) with optimized implementation of the embedded generalized matrix-matrix multiplication (GEMM). Expand
Beyond Power over Ethernet: the development of Digital Energy Networks for Buildings
Alternating current power distribution using analogue control and safety devices has been the dominant process of power distribution within our buildings since the electricity industry began in theExpand
Modelling energy efficiency for computation
TLDR
It is argued that the strands of the energy problem are inextricably entangled and cannot be solved in isolation, and unifying these diverse approaches into one " Theory of Energy " for computing. Expand
A High Speed Low Power CAM With a Parity Bit and Power-Gated ML Sensing
TLDR
This paper introduces a parity bit that leads to 39% sensing delay reduction at a cost of less than 1% area and power overhead and proposes an effective gated-power technique to reduce the peak and average power consumption and enhance the robustness of the design against process variations. Expand
A Historical Perspective on Hardware AI Inference, Charge-Based Computational Circuits and an 8 bit Charge-Based Multiply-Add Core in 16 nm FinFET CMOS
  • Kayode Sanni, A. Andreou
  • Computer Science
  • IEEE Journal on Emerging and Selected Topics in Circuits and Systems
  • 2019
TLDR
This paper presents the architecture, implementation and measurements from a mixed-signal, charge-based 8-bit analog multiplier for limited precision linear algebra in AI systems, and argues that if done right, analog computation will have a role in future hardware AI systems. Expand
Range Scaling of Wirelessly Powered Sensor Systems
TLDR
It is argued that energy efficiency scaling, when combined with RF path loss, implies that the range at which a particular fixed computational workload can be wirelessly powered is increasing exponentially, but with a “shallow” scaling exponent that is one half of the underlyingEnergy efficiency scaling exponent. Expand
Ternary circuits for Null Convention Logic
TLDR
Low-power circuits for basic components such as DATA/NULL Detector circuit and a Ternary Register are described based on the characteristics of a commercial silicon-on-sapphire MOS process that offers multiple simultaneous transistor thresholds. Expand
TECHNOLOGY Optimize Parity Encoding for Power Reduction in Content Addressable Memory
TLDR
The main challenge in CAM design is to reduce power while maintaining speed and low area, and this paper presents the CAM low power techniques at architecture level. Expand
Optimize Parity Encoding for Power Reduction in Content Addressable Memory
Most memory devices store and retrieve data by addressing specific memory locations. As a result, this path often becomes the limiting factor for systems that rely on fast memory accesses. The timeExpand
Bio-inspired system architecture for energy efficient, BIGDATA computing with application to wide area motion imagery
TLDR
A brain-inspired system architecture for real-time big velocity BIGDATA processing that originates in large format tiled imaging arrays used in wide area motion imagery ubiquitous surveillance. Expand
...
1
2
3
4
...

References

SHOWING 1-10 OF 62 REFERENCES
Establishing Moore's Law
  • Ethan Mollick
  • Computer Science
  • IEEE Annals of the History of Computing
  • 2006
TLDR
It is the entry of foreign competition that seems to have played a critical role in maintaining the pace of Moore's law in the early VLSI transition, as many different kinds of chips used many competing logic families. Expand
Bell's Law for the Birth and Death of Computer Classes: A theory of the Computer's Evolution
  • G. Bell
  • Computer Science
  • IEEE Solid-State Circuits Newsletter
  • 2008
TLDR
In this new paper Gordon Bell explains the history of the computing industry, positing a general theory ("Bell's Law) for the creation, evolution, and death of computer classes since 1951, and predicts that the powerful microprocessor will be the basis for nearly all computer classes in 2010. Expand
Bell's Law for the Birth and Death of Computer Classes: A theory of the Computer's Evolution
TLDR
In this new paper Gordon Bell explains the history of the computing industry, positing a general theory ("Bell's Law) for the creation, evolution, and death of computer classes since 1951, and predicts that the powerful microprocessor will be the basis for nearly all computer classes in 2010. Expand
Cramming More Components Onto Integrated Circuits
  • G. Moore
  • Computer Science, Engineering
  • Proceedings of the IEEE
  • 1998
The future of integrated electronics is the future of electronics itself. The advantages of integration will bring about a proliferation of electronics, pushing this science into many new areas.Expand
The Landscape of Parallel Computing Research: A View from Berkeley
TLDR
The parallel landscape is frame with seven questions, and the following are recommended to explore the design space rapidly: • The overarching goal should be to make it easy to write programs that execute efficiently on highly parallel computing systems • The target should be 1000s of cores per chip, as these chips are built from processing elements that are the most efficient in MIPS (Million Instructions per Second) per watt, MIPS per area of silicon, and MIPS each development dollar. Expand
ASSESSING TRENDS OVER TIME IN PERFORMANCE , COSTS , AND ENERGY USE FOR SERVERS
Data centers are the heart of the global economy. In the mid-1990s, the costs of these large computing facilities were dominated by the costs of the information technology (IT) equipment that theyExpand
A 30 Year Retrospective on Dennard's MOSFET Scaling Paper
  • M. Bohr
  • Engineering
  • IEEE Solid-State Circuits Newsletter
  • 2007
The MOSFET scaling principles for obtaining simultaneous improvements in transistor density, switching speed, and power dissipation described by Robert H. Dennard and others in "Design ofExpand
Power levels in office equipment: Measurements of new monitors and personal computers
TLDR
Findings include a trend among monitor manufacturers to provide a single very low low-power level, and the need to standardize methods for measuring monitor on power, to more accurately estimate the annual energy consumption of office equipment, as well as actual and potential energy savings from power management. Expand
The CRAY-1 computer system
TLDR
The CRAY-1 is the only computer to have been built to date that satisfies ERDA's Class VI requirement (a computer capable of processing from 20 to 60 million floating point operations per second) and its Fortran compiler (CFT) is designed to give the scientific user immediate access to the benefits of the Cray-1's vector processing architecture. Expand
The business of scaling
  • Rakesh Kumar
  • Engineering
  • IEEE Solid-State Circuits Newsletter
  • 2007
Managing the economics of scaling and increasing demand have been key factors in the growth of the semiconductor industry to nearly $250B over more than four decades. During the same period,Expand
...
1
2
3
4
5
...