On the Supercomputer Technology Crisis

scoobrs writes "Experts claim America has been eating our 'supercomputer feed corn' by developing clusters rather than new supercomputer processors and interconnects. Forbes says America is playing catch-up and that the new federal budget items are too little too late. Cray is laying people off due to decreased federal spending and claims lower margin products have forced them to create products based on commodity parts. Red Storm, one of their new Linux-based products, is being delayed to next year."

Comments

I Need A RAIS

[grunt107]

Random Array of Inexpensive Servers.

If the 'supercomputers' of today are increasing performance, does it really matter the design?

Maybe that is a signal that monolithic computer tasks are best handled in a hive mentality - have the Queen issue the big orders, have the warriors performing security, have the workers transporting the goodies (data), and have the requisite extra daughters and suitors to grow the hive and assure its viability (redundancy).

The fact that it is cost-effective is even better.

You are all missing the point

[Anonymous Coward]

Its the fact that clusters require higher skill to program efficiently for than do single processor systems. Plus you have all of the wasted processing power used for communication between the nodes. Granted, many problems lend themselves well to distributed computing (essentially what a cluster is, but the nodes are closer and communicate faster), but there are also problems that are handled better by a smaller amount of specialized hardware. The other point is that by using off the shelf parts, we are not really innovating in this space like we should be. We are allowing the commodity computer market determine the direction of the supercomputer market.

Re:it makes sense

[badboy_tw2002]

Then trust the fact that not all problems are easily attacked from a parallel perspective. This means problems where working on one section of the dataset affects large amounts of data in other sections. There's a lot of locking and waiting for tasks in other parts of the system to be completed; and a lot of data transfer/need for shared memory, which if you're bussing between cluster components, its going to be slow.

This doesn't mean that clusters don't have some use in these regards, it just means that for these types of problems no one has figured out an efficient parallel algorithm to use on them.

What tasks require high-speed interconnects?

[Bruce Perens]

One of the nice things about clusters is that they encourage people to consider how to decompose a problem so that it can work without a large high-speed shared data memory. Some of the older supercomputers were important because scientists hadn't done this work because there wasn't the economic incentive back then. Now there is one.

So, what tasks still require a high-speed shared data memory? Answer that, and you'll understand where you can still sell a supercomputer.

Bruce

Complex issues that have to be solved

[Anonymous Coward]

I've been in this field over 25 years, been in public position at a major lab now for 8.

If this was a simple issue, the HPC community would already have completely moved to clusters and never looked back 3 or 4 years ago. But it's not kiddies.

Want to run a physics projection for more than 1 microsecond? Takes real horsepower that clusters cannot provide even distributed. Just too much damn data. Chem codes that include REAL data for useable time slices? too slow for clustered memory. Every auto maker in the world (almost) has been whining about the lack of BIG horsepower for a few years now.(crash codes and FEA) I could go on forever. Sure, some problems work awesome on clusters, which is why we have them. But definately not all of them.

The problem is partly diminishing returns, partly the pathetic ammount of useable memory on a cluster and its joke for memory throughput, partly the growth in power of the low end and clustered networking, partly the ridiculously long development cycles invloved in High Performance Computing and the low $ returns,

One of the biggest things congress sees is that this country will more than likely NEVER again lead the world in computing power for defense and research.

And thats something we ought to do as the last real Superpower.

The national labs TRIED clusters, they don't get all the jobs done they wanted. (see testimony before congress, writings in HPC jounals, and the last couple RFPs from US gov. labs,heck every auto maker in the world) People in HPC _know_ it now, but having let what little there was of the supercomputer industry die out, there isn't mcuh of an industry left to turn to now. It just may be too darned late. HPC hasn't been a money making industry since the early 80s.
Heck, even Intel abandoned their clustered machine they custom built for the government.

Most folks in HPC will readily admit the Top500 is kind of a joke. The HPC-challenge #s are a little more realistic for the tests, but we really do need something that approximately real world applications, not just a 70s cpu benchmark.

For those that think this is a 'Linux wins' issue,
consider that mostly it was fast interconnect networks that allowed clustering, not the OS. Examine the history of clusters and you'll see this is true. Btw, the last few SC companies are already mostly moving to linux anyway.(nec,fujitsu,cray;ibm dabbles in hpc)

Hopefully the industry will survive long enough to allow for even better mergers of supercomputing power with low end cost, but at this point I doubt it. Cray has been on the ropes since 96, fujitsu's sc division is a loss leader, and NEC has been trying to get out of it for a while for something with a margin.

Ed -gov labs HPC research punk
-former Cray-on
-former CDC type

Clusters and supercomputers...

[gillbates]

I've seen a lot of naive comments suggesting that supercomputers are being replaced by clusters. The truth is, anyone who can replace their supercomputer with a cluster didn't need a supercomputer in the first place:

1. (compared to a supercomputer):
2. The prime advantage of an x86-based server is that it is cheap, and it has a fast processor. It is only fast for applications in which the whole dataset resides in memory - and even then, it is still the slowest of the group.
3. Clusters are a little better, but suffer from severe scalability problems when driving IO-bound processes. As with the x86 server, if you can't put the full dataset into memory, you might as well forget using a cluster. The node to node throughput is several orders of magnitude slower than the processor bus in multiple CPU systems. (6.4GB/s vs 17MB/s for regular ethernet, or 170MB/s for Gigabit)
4. Multiple CPU servers do better, but still lack the massive storage capacity of the mainframe. They work better than clusters for parallel algorithms requiring frequent syncronization, but still suffer from a lack of overall data storage capacity and throughput.
5. Mainframes, OTOH, possess relatively modest processors, but the combined effect of having several of them, and the massive IO capability makes them very good for data processing. However, their processors aren't fast at anything, and often run at 1/2 or 1/3 the speed of their desktop counterparts.
6. Supercomputers combine the IO throughput of a mainframe with the fast processors typically associated with RISC architectures (if you can still consider anything RISC or CISC nowadays). They have faster processors, more memory, and much greater IO throughput than any other category.

It used to be that the prime reason for faster computers came from the scientific and business communities. But now that the internet has turned computers into glorified televisions, the challenges have gone from that of crunching numbers to serving content:

1. Clusters are great for serving read-only content, because there's very little active synchronization required between nodes, and the aggregate IO capacity scales well.
2. Mainframes reign when it comes to IO throughput - companies that formerly had use for a supercomputer are finding that their role is shifting to more of an information-provider role; faster processors are no longer as important as fast IO subsystems.
3. Scientists aren't being trained to use the computer as a tool; most think of a computer more or less as a means of verifying their hypothesis, rather than a means of discovering possible explanations. Their primary work is done with a calculator and pencil, and only later, when they need something to back up their ideas, do they turn to a computer simulation. The computer is a verification tool, not a means of discovery.

As our economy has shifted away from a technological base to an entertainment one, the need for supercomputers has begun to evaporate. We outsource innovation overseas so that we can lounge around on the couch watching tv and drinking beer (or surfing the net and drinking beer). The primary purpose of technological innovation has shifted from that of discovering the universe to merely bringing us better entertainment.

Re:it makes sense

[grawk]

As someone who works for a supercomping center, I can say that some things work VERY well on cheap unix based clusters. I am the primary admin on a 5 TFLOP cluster. We've also got a Cray X1, and while it's only 2.6 TFLOPs, it will eat my IBM's lunch when it comes to some specificly tuned tasks. Much in the same way that we can outperform mac clusters that have significantly higher floating point performance because of the speeds of the interconnects. Supercomputing is about a LOT more than just raw CPU power.