The concept has been around for some time, but Simon Moores
admits to being a little awed by the arrival of the DNA
computer.
The technology sector may be at its lowest ebb for a
generation but progress, like politics, shows little respect for
economics.
In the same week I read that Sun Microsystems plans to create
processors that will increase its blade server throughput by an
impressive-sounding factor of 30 beyond 2005, I see that Israeli
scientists have devised a computer that can perform "330 trillion
operations per second". This is more than 100,000 times faster than
the quickest PC with us today, shattering Moore’s Law in the
process.
While Sun is initiating what it calls "throughput" in its
processors, combining chip multithreading to allow a single
processor to execute tens of threads simultaneously, at the
Weizmann Institute of Science in Israel, they have arrived at a new
computing paradigm and a radically different view of threading
which might sit more comfortably on the set of a Star Trek
movie.
What makes this new computer a little different is that it’s
made of jelly, DNA, in fact and, while the idea has been around for
some time - I wrote about it five years ago - it’s taken that long
for the blob to sweep past the best that Intel or anyone else can
put up against it.
The blade, or blob, if you can describe a jelly in such a way,
is a programmable molecular computing machine composed of enzymes
and DNA molecules rather than silicon processors. The big
breakthrough in the research is the harnessing of the single DNA
molecule that provides the computer with its input as a fuel source
to "power" the device; an enzyme breaks bonds in the DNA double
helix, triggering the release of enough energy for the system to be
self-sufficient.
Last week, The Guinness Book of Records recognised Israel’s DNA
computer as "the smallest biological computing device" ever
constructed in an area that is likely to transform our
understanding of computing as dramatically as the appearance of the
first IBM PC 20 years ago.
If, like me, you’re having trouble attempting to imagine
carrying your intelligent laptop supercomputer around in a
refrigerated dish 10 years from now, then I don’t blame you. But
think back 20 years and most of us then would have had trouble
imagining the wired world of 2003.
Add this kind of news to the imminent arrival of the first
terabyte hard drives, then carrying the world in the palm of your
hand becomes much less science fiction and much more of a serious
probability before the end of the decade.
Will such machines run "best with Windows"? That’s anyone’s
guess, but I wonder if the operating system argument will have
become entirely meaningless by then. Will tomorrow’s jellies, I
wonder, carry an Intel Inside logo, and what might happen if you
leave one out in the sun?
It’s all too much for me, and one really has to wonder how
enterprise IT will cope now that we are just about done with
"fourth wave" computing and are starting to picture a future where
both processor power and storage appear almost limitless by today’s
standards.
One thing you can be sure of, though. Software won’t be free,
jellies will, occasionally, crash or, dare I say it, melt down, and
there’ll still be the familiar heated arguments over standards and
open-source computing.
What do you think?
Do you closely track cutting-edge R&D?
Tell us in an e-mail >> CW360.com
reserves the right to edit and publish answers on the Web site.
Please state if your answer is not for
publication.
ZentelligenceSetting the world to rights with the collected thoughts and
opinions of the futurist writer, broadcaster and Computer Weekly
columnist Simon Moores.