IEEE Spectrum October, 2012 - 13

US $1230 Per kIlogrAm

Cost of extracting uranium from seawater using a recently improved technique.
It's twice as productive as the old method but is easily outstripped by mining.

Could Supercomputing Turn to
Signal Processors (Again)?

texas team says digital signal processors could compete
in high-performance computing

B

top: gluekit; right: cornell university

uilding highperformance computers
used to be all about
maximizing flops, or floatingpoint operations per second. But
the engineers designing today's
high-performance systems are
keeping a close eye not just on
the number of flops but also
on flops per watt. Judged by
that energy-efficiency metric,
some digital-signal processing
(DSP) chips-the sophisticated
signal conditioners that run
our wireless networks, among
other things-might make
promising building blocks for
future supercomputers, recent
research suggests.
The DSPs that might make
the jump to supercomputing
come from Texas Instruments,
which originally designed them
for relatively modest applications. "We hadn't even thought
to look at high-performance
computing or supercomputers," says Arnon Friedmann,
multicore business manager for
TI. "It wasn't on our radar." TI's
DSP chips are typically used

spectrum.ieee.org

in embedded systems, most
prominently cellular base stations. For such applications,
power efficiency is vital, but
until recently these systems
didn't require floating-point
calculations, making do instead
with just integer arithmetic.
The advent of 4G cellular networks, however, increased
the computing burden within
base stations, making floatingpoint calculations essential.
TI engineers added
floating-point hardware to the
TMS320C66 family of multicore
DSPs late in 2010 without
appreciably slowing these
processors down or increasing
the power consumed. But it was
only after the new chips came
out that some forward thinkers
at TI realized that the eight-core
C6678 DSP, which can perform
as many as 12.8 gigaflops per
watt running flat out, might
be useful for general-purpose
high-performance computing.
"The question was whether
we'd be able to extract that
potential in the real world,"

says Francisco D. Igual, now
a postdoctoral researcher at
Universidad Complutense de
Madrid. He was working at the
University of Texas at Austin
with engineering professor
Robert A. van de Geijn when
TI approached them for help.
Collaborating with TI, they
wrote code for the new DSP to
perform general matrix-matrix
multiplication, something they
felt would be representative of
the kind of numerical weight
lifting that high-performance
systems are often asked to do.
With that code in hand, the
team compared the new DSP
chip against some common
supercomputer architectures.
The DSP came out on top, at
7.4 gigaflops per watt. "We
were happy, but we were not
that surprised," says Igual.
But not everyone is swayed
by those results. "It's very
impressive, but not a fair
comparison," says John Shalf of
the National Energy Research
Scientific Computing Center
at Lawrence Berkeley National

Superthin
Circuits
Researchers at Cornell
University, in Ithaca,
N.Y., have found a way
to get graphene, the
nanoelectronics wonder
material, to knit together
with also-ran boron nitride.
The result is a striped
pattern of interlocking
graphene conductors and
boron nitride insulators
that forms an array of wires
just one atom thick. Sheets
of the arrays could possibly
be laid atop each other
to form complex circuits,
the researchers say. The
conductive graphene can
also knit together with a
semiconducting version of
itself; the Cornell team is
going to try integrating it
with the two-dimensional
semiconductor molybdenum disulfide, too.

OCTOBER 2012 * IEEE SpECTRum * NA

13
http://spectrum.ieee.org
Table of Contents for the Digital Edition of IEEE Spectrum October, 2012
