Computer Architecture Techniques for Power-Efficiency by Stefanos Kaxiras

By Stefanos Kaxiras

Within the previous couple of years, energy dissipation has turn into a tremendous layout constraint, on par with functionality, within the layout of recent computers. while long ago, the first task of the pc architect was once to translate advancements in working frequency and transistor count number into functionality, now strength potency has to be taken under consideration at each step of the layout technique. whereas for it slow, architects were profitable in providing forty% to 50% annual development in processor functionality, expenses that have been formerly disregarded ultimately stuck up. the main serious of those charges is the inexorable elevate in strength dissipation and tool density in processors. strength dissipation matters have catalyzed new subject parts in desktop structure, leading to a considerable physique of labor on extra power-efficient architectures. strength dissipation coupled with diminishing functionality earnings, was once additionally the most reason for the change from single-core to multi-core architectures and a slowdown in frequency bring up. This e-book goals to rfile the most very important architectural concepts that have been invented, proposed, and utilized to minimize either dynamic energy and static strength dissipation in processors and reminiscence hierarchies. an important variety of suggestions were proposed for quite a lot of events and this ebook synthesizes these suggestions by means of targeting their universal features.

Show description

Read or Download Computer Architecture Techniques for Power-Efficiency PDF

Similar design & architecture books

Mastering JXTA: Building Java Peer-to-Peer Applications

A entire, code-intensive consultant to construction commercial-quality peer-to-peer purposes with JXTA and Java millions of individuals use peer-to-peer (P2P) functions akin to KaZaA, AOL fast Messenger, and disbursed. web. those purposes harness the idle CPU cycles in their host desktops to provide huge, immense databases of data, construct robust processing engines, and permit communique and file-sharing between clients world wide.

Network Architecture & Design ''A Field Guide for IT Professionals'' (Sams White Book)

Community structure and layout takes readers via each section of a brand new venture from buyer conferences, web site surveys, facts assortment and interpretation, documentation to really designing and enforcing the community in response to spec. The dialogue includes:An assessment of LAN and WAN topologiesCoverage of NOS (Novell working System)Integration of the customer working approach (this 50% of community structure is usually ignored in related titles)ProtocolsConnectivity DevicesImplementing distant AccessSecurityInternet connectivityNetwork MonitoringIn addition, the writer has ready a pattern of patron documentation, a thesaurus of phrases and a bother taking pictures fast reference consultant.

Computer Organization and Design: The Hardware Software Interface, 3rd Edition

A revised printing for this publication could be on hand in June 2007! what is New within the 3rd variation, Revised Printing a similar nice publication will get larger! The revised printing good points all the unique content material in addition to those extra features:. Appendix A (Assemblers, Linkers, and the SPIM Simulator) has been moved from the CD-ROM into the published publication.

Load Distribution: Implementation for the Mach Microkernel

J iirgen N ehmer Load distribution is an important proposal for dispensed platforms so one can in attaining higher functionality, source usage and reaction occasions. delivering effi cient mechanisms for the obvious help of load distribution has confirmed to be an exceptionally tricky project.

Additional resources for Computer Architecture Techniques for Power-Efficiency

Sample text

Another similar question regards whether continuous settings of (V , f ) pairs are possible, or whether these values can only be changed in fixed, discrete steps. ” As a result, simple online techniques might have difficulty finding global optima, and more complicated or offline analysis again becomes warranted. Because DVFS is available for experimentation on real systems [111, 112, 2], and because it offers such high leverage in power/energy savings, it has been widely studied in a variety of communities.

OPT, FUTURE, and PAST: OPT is a simplified Oracle algorithm that perfectly eliminates idle time in every quantum by stretching the run times in a trace. It can look arbitrarily far into the future. It provides a reference point for scheduling all work in a power-optimal way. However, it makes several over-simplifications. First, it does not make a distinction between “soft” and “hard” idle time. , for I/O) that should not be stretched or compressed. In addition, it does not care on how long a job is delayed, as long as it finishes by the end of the trace.

In which units) the power dissipation is most or least prominent. Such per-unit power attribution is useful both for guiding powerefficient design optimizations, as well as for guiding thermal models of on-chip hotspots. For example, a total power estimate will merely tell me if I am near or exceeding the overall chip power budget or thermal capacity; it cannot tell me about whether I have one particular hotspot on the chip that is nearing its local thermal limit. Accurate and efficient per-unit power estimators can be built by exploiting the specific hardware performance counters provided on nearly all high-performance microprocessors today.

Download PDF sample

Rated 4.47 of 5 – based on 46 votes