NASA's 10K-way supercluster

  • Thread starter Thread starter Yousuf Khan
  • Start date Start date
Y

Yousuf Khan

Yousuf said:
SGI Altix systems again. A cluster of 20 512-way Itanium2 Altixes connected
together with NUMAlink. There's also some talk of SGI re-architecting the
cluster so that it becomes 2048-way Altixes rather than 512-way Altixes.

http://www.theinquirer.net/?article=17491

Oh, BTW, it looks like NUMAlink is only used within individual Altixes. The
individual Altixes connect to each other via Infiniband.

http://news.zdnet.co.uk/hardware/0,39020351,39161899,00.htm

I'm not sure where the insightful discussion of what's really going on
here is taking place.

http://www.fcw.com/fcw/articles/2004/0726/web-nasa-07-28-04.asp

<quote>

"NASA has a long history in supercomputing dating back to the origins of
computational fluid dynamics in the 1980s," said G. Scott Hubbard,
director of the Ames Research Center. "It is exciting to join with an
industry team in this innovative venture that will change the very way
in which science and simulation are performed by providing researchers
with capabilities that, until now, they could only dream about."

</quote>

Like even the crudest meaningful estimate of the likelihood of a
catastrophic failure? To be fair, such an estimate may not be
obtainable by any existing methodology, and NASA has been performing
original field experiments in the area, albeit with the politically
explosive and ethically questionable admixture of live human subjects.

Mr. Hubbard had an opportunity to say something incisive about the real
role that computers might play and what the limitations, both real and
perceived, might be, and what, exactly, NASA is doing to make things
better, other than buying more and better boxes. Instead, he opted for
some bureaucrat blah-blah-blah that could have accompanied the roll-out
of just about anything, as long as it was a computer being used by NASA.

The stakes are high, and NASA is working under a
you'd-better-not-screw-up-again mandate. We had computers for the
previous screwups. A computer model played a direct role in the most
recent screwup. What's different now?

http://www.fcw.com/fcw/articles/2004/0726/web-nasa-07-28-04.asp

<quote>

Project Columbia, expected to give NASA's supercomputing capacity a
tenfold boost, will simulate future missions, project how humans affect
weather patterns and help design exploration vehicles.

<quote>

In other words:

Project Columbia will

1. Lead to more Top 500 press releases (more meaningless numbers).
2. Try to contribute to a methodology for failure-mode prediction that
might keep some future NASA Administrator from being hung out to dry.
3. Lead to more NASA web pages and color plot attributions on the
subject of global warming.
4. Do more detailed design work.

The one that matters is (2), which is a qualitatively different mission
from the other three. For (1), (3), and (4), more is better, almost by
definition. For (2), more is actually a trap. NASA _already_ has the
embarrassment of stupefyingly complicated failure analyses that have
proven, unfortunately only in retrospect, to have missed the critical
issues.

"Simulate future missions" means anticipating all the things that could
go wrong. That's a big space, and it's not filled with zillions of
identical grid points, all with the same state vector interacting
according to well-understood conservation laws. It's filled with
zillions of specialized widgets, each with its own specialized state
vector, and all those state vectors interact interact according to
poorly-understood ad hoc rules. I suspect that, if NASA could have put
all 10K nodes under the same system image, they would have; as it is,
the meganodes at least minimize the artificial complications due to
computational (as opposed to real-world system) boundaries.

http://www.computerworld.com/hardwaretopics/hardware/story/0,10801,94862,00.html

<quote>

The reliance of supercomputers built on clusters, commodity products
that operate in parallel, has been criticized before congressional
panels by some corporate IT executives, including officials at Ford
Motor Co. They see the use of clusters as a setback for supercomputing
because the hardware doesn't push technological advances (see story).

[http://www.computerworld.com/hardwaretopics/hardware/story/0,10801,94607,00.html]

SGI officials said that criticism doesn't apply to their system, which
uses a single image of the Linux operating system in each of its large
nodes and has shared memory.

David Parry, senior vice president of SGI's server platform group, said
the system is "intended to solve very large complex problems that
require the sharing of capabilities of all of those processors on a
single problem at the same time."

</quote>

The criticism doesn't apply to their system? Say what? A single system
image sounds nice for a gigantic blob of a problem with so many unique
pieces that just getting the crudest screwups out of the simulation
(never mind how well it corresponds to reality) sounds like a major
challenge. Make the problem big enough, though, and a shuttle mission
is about as big as it gets in terms of what people are actually going to
try, and you still have to cluster.

RM
 
Robert said:
<quote>

The reliance of supercomputers built on clusters, commodity products
that operate in parallel, has been criticized before congressional
panels by some corporate IT executives, including officials at Ford
Motor Co. They see the use of clusters as a setback for supercomputing
because the hardware doesn't push technological advances (see story).

[http://www.computerworld.com/hardwaretopics/hardware/story/0,10801,94607,00.html]

Wow! And here I thought we were trying to use cost-effective hardware to
solve the problems. I thought the technical advances came from solving
the problems. Sounds like sales talk to me, we would get the same answer
using more expensive computers?

Is this the same theory which holds that getting from point A to point B
is somehow better for your business if you go first class? 'Scuse me if
I don't buy it, if a computer vendor wants to "push technological
advances" they can do it on their dime, I want to push my bottom line.
SGI officials said that criticism doesn't apply to their system, which
uses a single image of the Linux operating system in each of its large
nodes and has shared memory.

David Parry, senior vice president of SGI's server platform group, said
the system is "intended to solve very large complex problems that
require the sharing of capabilities of all of those processors on a
single problem at the same time."

</quote>

The criticism doesn't apply to their system? Say what? A single system
image sounds nice for a gigantic blob of a problem with so many unique
pieces that just getting the crudest screwups out of the simulation
(never mind how well it corresponds to reality) sounds like a major
challenge. Make the problem big enough, though, and a shuttle mission
is about as big as it gets in terms of what people are actually going to
try, and you still have to cluster.

I can find an emoticon which means "I'm looking for your point," but I
am. It doesn't matter how you solve a problem as long as you get the
right answer. And using different computers would not help NASA (or
anyone else) if they ask the wrong questions.

This is a group for discussion of Intel compatible hardware, I think we
can leave the politics out of it, thanks.
 
Bill said:
Robert said:
<quote>

The reliance of supercomputers built on clusters, commodity products
that operate in parallel, has been criticized before congressional
panels by some corporate IT executives, including officials at Ford
Motor Co. They see the use of clusters as a setback for supercomputing
because the hardware doesn't push technological advances (see story).

[http://www.computerworld.com/hardwaretopics/hardware/story/0,10801,94607,00.html]



Wow! And here I thought we were trying to use cost-effective hardware to
solve the problems. I thought the technical advances came from solving
the problems. Sounds like sales talk to me, we would get the same answer
using more expensive computers?

The critic from Ford isn't a computer salesman...he's a buyer, and his
comments are far from unique.

The question isn't, will you get the right answer, but will you get an
answer in time. There are zillions of different ways of using computers
for applications, but I have most often seen computers used successfully
in ways that are very little different from the mythical envelope: as an
aid to visualization and thought. In that mode, it's more important to
get the drawing scratched out when the ideas are hot than it is to make
the lines beautiful. Workstations meet that need nicely. Much of the
space that was once filled by more powerful boxes (say, from SGI) has
been taken over by workstations as workstations have become more powerful.

What do you do, though, when the workstation won't give you the answer
fast enough? You put together a bunch of workstations into a cluster.
Getting the bunch of workstations to produce the answer faster is often
possible, but it takes significant work, and you are definitely out of
back of the envelope territory.

The government has mostly been funding clusters. What industrial users
could most readily use is a bigger, faster envelope.
Is this the same theory which holds that getting from point A to point B
is somehow better for your business if you go first class? 'Scuse me if
I don't buy it, if a computer vendor wants to "push technological
advances" they can do it on their dime, I want to push my bottom line.

By the same logic, we don't need that national labs playing in the
sandbox to build clusters. They are easy enough to build. If the DoE
is going to claim to advance technology, then they should be trying
things that really would advance technology. The DoE's stated desire is
to help make HPC available as a competitive tool for industry. If they
really want to do that, then I would think that what potential
industrial users think would matter a great deal.
I can find an emoticon which means "I'm looking for your point," but I
am. It doesn't matter how you solve a problem as long as you get the
right answer. And using different computers would not help NASA (or
anyone else) if they ask the wrong questions.

It matters a great deal how you get the answer if you've already cut
metal by the time the answer is available.

Some HPC problems are highly-predictable with lots of structure. Those
problems are at least conceptually suited to being partitioned onto a
slew of identical nodes. Big system simulations aren't likely to have
that kind of conveniently crystalline structure, and getting the work
distributed over a large number of nodes in a reasonable fashion could
be comparable in difficulty to the problem itself. Ideal for a big SMP
box, but the problem is so big that you wind up with a cluster, anyway.
In the end, it _is_ a cluster, and the whole simulation could be
bottlenecked by box-to-box communications required by only a small
subset of the simulation.
This is a group for discussion of Intel compatible hardware, I think we
can leave the politics out of it, thanks.

Altix uses Itanium, and purchases of very large computers are inherently
political.

RM
 
On Fri, 30 Jul 2004 14:43:22 GMT, Robert Myers

<snip>

What an alarmingly wellinformed response.
cheers!

Now, when will we see some usable quantum computers?
 
Loke said:
On Fri, 30 Jul 2004 14:43:22 GMT, Robert Myers

<snip>

What an alarmingly wellinformed response.
cheers!

Thank you.
Now, when will we see some usable quantum computers?

A promising resource for those aspiring to be well-informed would be

http://qubit.nist.gov/qiset-PDF/Williams.QISET2004.pdf

which projects 2025 for "Quantum Computation."

It's easy enough to find predictions that quantum computers will never
be used for general-purpose computing. I wouldn't indulge such a
prediction, but quantum computers will be used for applications like
cryptography long before they are used for general purpose computation.
I'm hoping that we can find some more cleverness with classical
computers in the meantime.

RM
 
Robert said:
[...]

A promising resource for those aspiring to be well-informed would be

http://qubit.nist.gov/qiset-PDF/Williams.QISET2004.pdf

which projects 2025 for "Quantum Computation."

It's easy enough to find predictions that quantum computers will never
be used for general-purpose computing. I wouldn't indulge such a
prediction, but quantum computers will be used for applications like
cryptography long before they are used for general purpose computation.
I'm hoping that we can find some more cleverness with classical
computers in the meantime.

RM

Isn't Quantum Computation just a paper exercise? Certain quantum principles
are inherently parallel, but how to access those states when a quantum
measurement collapses those states? Secondly, a transistor already rely on
quantum principles that lead to movements of a charge in controlled energy
bands for the crystal structures.
 
Johannes said:
Isn't Quantum Computation just a paper exercise?

If there is credible information that quantum computing can never be
made practical, I'm not aware of it.
Certain quantum principles
are inherently parallel, but how to access those states when a quantum
measurement collapses those states?

http://www.cs.caltech.edu/~westside/quantum-intro.html

tackles the question you are asking head-on and describes actual
experiments that illustrate the proposed answer to your objection, which
is to use redundancy to allow non-destructive exploitation of quantum
information.
Secondly, a transistor already rely on
quantum principles that lead to movements of a charge in controlled energy
bands for the crystal structures.

Thermodynamics pretty do a pretty good job scrambling the quantum state
of individual objects quickly enough that it can't be exploited for
quantum computing in a room-temperature semiconductor device.

It is possible to preserve and to exploit quantum state information in a
macroscopic device, as in a SQUID, but the geometry is engineered to
make the quantum state manifest, and the device operates at a low enough
temperature that the quantum state has a usefully long decoherence time.

RM
 
Robert said:
If there is credible information that quantum computing can never be
made practical, I'm not aware of it.


http://www.cs.caltech.edu/~westside/quantum-intro.html

tackles the question you are asking head-on and describes actual
experiments that illustrate the proposed answer to your objection, which
is to use redundancy to allow non-destructive exploitation of quantum
information.


Thermodynamics pretty do a pretty good job scrambling the quantum state
of individual objects quickly enough that it can't be exploited for
quantum computing in a room-temperature semiconductor device.

It is possible to preserve and to exploit quantum state information in a
macroscopic device, as in a SQUID, but the geometry is engineered to
make the quantum state manifest, and the device operates at a low enough
temperature that the quantum state has a usefully long decoherence time.

RM

Thanks for info. Now I probably know a little more, or think that I know.
I guess the type of algorithms they're looking for are massively parallel
algorithms. However, NP algorithms are parallel by virtue of just trying
out every possibility (brute force) and see which number will fit a given
problem. Nature itself does parallel computations to find the optimal path
of a quantum wave function. But with so much emphasis on the algorithmic
side of quantum computing, it would be nice to see at least a simple
hardware device in operation. Secondly, as I mentioned, quantum principles
are not unique to quantum computing, transistors use them too.
 
Johannes H Andersen wrote:

But with so much emphasis on the algorithmic
side of quantum computing, it would be nice to see at least a simple
hardware device in operation.

Well, let's see, how are we doing? The transistor was invented in 1947,
and the first fully transistorized computer, TRADIC, appeared in 1955,
with recognizable computing elements having appeared at Bell Labs c. 1950.

http://www.infoworld.com/article/03...F/6pmakDp3es5w=&ERIGHTS=1091979647605_rbmyers

(requires registration)

<quote>

October 29, 2003

A research team in Japan says it has successfully demonstrated for the
first time in the world in a solid-state device one of the two basic
building blocks that will be needed to construct a viable quantum computer.

The team has built a controlled NOT (CNOT) gate, a fundamental building
block for quantum computing in the same way that a NAND gate is for
classical computing.

Research into quantum computers is still in its early days and experts
predict it will be at least 10 years before a viable quantum computer is
developed.

</quote>

When you consider that the key ideas had already been worked out with
vaccum tubes, and that building a solid-state computer was a fairly
straightforward translation from vacuum tubes to solid-state devices,
there seems to be nothing about the way things are progressing in the
development of actual quantum computing devices to say that it is slow
by historical standards or that there is any indication that the
enterprise will not end in success. Surprise, of course, is an
inevitable accompinament to discovery.
Secondly, as I mentioned, quantum principles
are not unique to quantum computing, transistors use them too.

If state is represented as a scalar value chosen from the distinct
possibilities 0 and 1, you have a classical computer, no matter how much
quantum mechanics has to be conjured in explaining its operation. If
state is represented as a vector a0*|0> + a1*|1>, where |0> and |1> are
quantum states and where a0**2+a1**2==1, you have a quantum computer.

RM
 
Robert said:
Johannes H Andersen wrote:
When you consider that the key ideas had already been worked out with
vaccum tubes, and that building a solid-state computer was a fairly
straightforward translation from vacuum tubes to solid-state devices,
there seems to be nothing about the way things are progressing in the
development of actual quantum computing devices to say that it is slow
by historical standards or that there is any indication that the
enterprise will not end in success. Surprise, of course, is an
inevitable accompinament to discovery.


If state is represented as a scalar value chosen from the distinct
possibilities 0 and 1, you have a classical computer, no matter how much
quantum mechanics has to be conjured in explaining its operation. If
state is represented as a vector a0*|0> + a1*|1>, where |0> and |1> are
quantum states and where a0**2+a1**2==1, you have a quantum computer.

RM

__ __
a0*a0+a1*a1=1 :-)

I am as curious as anybody, of course. But there is the usual line of:
"historic performance does not guarantee future results...". This is
neither a negative or a positive statement.

I such computers were made, I would imagine that quantum processors
will be very specialised, at least in the beginning. My guess is that
they will be hardwired machines that can solve particular problems; a
kind of extension of the old analog computing, but with exact results.
Even if such devices were not programmable, they could still be important
for solving certain types of problems e.g. large factorsation and large
combinatorial optimisation problems that are inaccessible by today's
machinery. Other problems may be cast into a few fundamental solvers.
Our pentium or AMD of the future would then run programs as usual, but
occasionally access a specialised quantum device for solving a particular
problem. All this is my own extrapolation and not based on any official
source on quantum computing.
 
Back
Top