UTCS & ECE FACULTY CANDIDATE: James Balfour/Stanford University: "ELM - Parallel Architectures for Efficient Embedded Computing" ACES 2.302, Monday, April 6, 2009 2:00 p.m.

Contact Name: 
Jenna Whitney
Apr 6, 2009 2:00pm - 3:00pm

There is a <a target="UTCSsignupSchedule" hr


how.cgi?person=JamesBalfour-CSandECEFACULTYCANDIDATE">signup sche

dule</a> for this event (UT EID required).

Type of Talk:&nbsp


Speaker/Affiliation:  Jame

s Balfour/Stanford University

Date/Time:  Monday, April 6, 20

09  2:00 p.m.

Location:  ACES 2.302

Host:  S

teve Keckler and Yale Patt

Talk Title:  "ELM - Parallel A

rchitectures for Efficient Embedded Computing"

Talk Abstract:

Embedded systems are ubiquitous, appearing in applications as diverse
as mobile phones, digital television, automobiles, communications syste

ms, sensor networks, and medical devices.  Satisfying the demanding
computational and efficiency requirements of such systems presently requir

es the use of complex application-specific integrated circuits (ASICs). Des

igning ASICs is time consuming and expensive, often involving tasks such a

s lowering reference implementations from C to equivalent gate-level repres

entations. The significant design and verification efforts required impose

large non-recurring engineering costs in the development of new systems, d

eterring innovation and limiting the viability of new applications.

Programmable processors offer flexibility, improved productivity

, and  reduced cost. Unfortunately, even low-power embedded process

ors are  inefficient compared to ASICs.  The inefficiency arise

s because  programmable processors devote significantly more energy a

nd chip area  to delivering instructions and data to functional units

.  To become viable ASIC replacements, programmable processors must

deliver high  computational performance with significantly greater ef


In this talk, I will describe ELM, an energy-e

fficient programmable  processor for high-performance embedded applic

ations.  ELM is 10x more  efficient that conventional embedded

processors, and approaches the  efficiency of ASICs on compute-inten

sive tasks.  ELM achieves this  efficiency by using distributed
and hierarchical register and memory  organizations that allow softw

are to better exploit parallelism and  instruction/data locality in a

pplications.  This talk will focus on  novel computer architect

ure, compiler, and circuit techniques used in  ELM, and discuss ho

w the systems-based approach used to design ELM can  be used to impro

ve the efficiency of embedded systems more broadly.