Slashdot Mirror
Cray CTO: Linux clusters don't play in HPC
jagger writes "Linux clustering was touted as the next big thing by many vendors last week at ClusterWorld Conference & Expo 2004. But supercomputer vendor Cray Inc. scoffed at the notion of putting Linux clusters in the high-performance computing (HPC) category. "Despite assertions made by Linux vendors, a Linux cluster is not a high performance computer," said Dr. Paul Terry, CTO of Cray Canada."
While Paul Terry makes some good points, in his statements, including the partial quote from the post, "Despite assertions made by Linux vendors, a Linux cluster is not a high performance computer, said Dr. Paul Terry, CTO of Cray Canada. "At best, clusters are a loose collection of unmanaged, individual, microprocessor-based computers."
Remember to take this with a grain of salt. The inflammatory nature of the comment is nothing more than a marketing ploy to increase visibility of, and sell, the new Cray XD1
Kinetic stupidity has a new brand leader: Allen Zadr.
A quote I've seen before:
"If you were plowing a field, which would you rather use? Two strong oxen or 1024 chickens?"
Maybe he meant penguins?
Nuke Gay Whales for Jesus.
Oracle disclaim MySQL and PostgreSQL as "toy databases", Microsoft claims that "Apache cannot be used for real web serving", and Sun announces that "Intel and Linux simply cannot be used for enterprise computing".
So all those supercomputing labs that use Linux clustering (that invented Linux clustering, even) have been wasting their time?
Ceci n'est pas une signature
I guess that the simple problem is just that the algorithm applied is usually not suitable for massively parallel computing.
It all depends on the problem you are trying to solve. I have been doing some work of late that would not complete in my life time on the 108 node cluster that we have. But when programmed for and run on two Cray X1s I should complete inside of a week.
Granted there are many codes (and more every day) that will run on clusters, the big iron will never die.
Just because we love Lunux doesn't mean that clusters are HPCs.
There are real issues that differentiate mainframe/supercomputers from large, powerful, clusters.
Of course this all depends on your definition of an HPC. But I believe that it's reasonable to say that if parts of your computer are connected with low bandwidth connections (10/100,gigabit) they just can't handle the same kinds of transactions that a computer with parts that are connected by 10 gigabit or 1000 gigabit connections or whatever it is nowadays.
As far as I know if you're deploying a large database it's still advisable to have a big huge IBM mainframe or a Unisys box or a Sun 10k instead of 4,8 or 16 clustered 8 proc machines.
My point is there are valid arguments for not including clusters of commodity hardware in the HPC category.
In my mind they aren't High Performance Computers... they are High Performance Clusters of Commodity Computers.
~foooo
Clusters can rival a supercomupter when they are assigned is a task that's suitable for distributed computing. That is, work units can be divided up and worked on in any sequence... the result of segment 45 doesn't depend on knowing the result of 44 and such. Effectively, you can have the sum of all of the processors minus just a little overhead for the clustering.
What Cray's rightfully pointing out is that for most business applications, however, distributed computing is not a viable option. When processing on a transaction basis, the transactions often need to posted in the exact order they were recieved, which means they must be taken serially. In those situations, the programs can't multithread work out to the other processors so well, and the cluster will end up running at roughly the speed of just one processor while the others waste clock cycles waiting for something to do.
The cluster isn't the solution to everything. Nor is the supercomputer. You've gotta think about the job, then figure out which tool is right for the task.
I happen to work in a facility that has large had both large supercomputers (cray t3e, j90, sgi) and linux and *nix based clusters (beowulf/linux, compaq/Tru64). The Cray CTO is correct that you can't just call every linux cluster out there HPC. Just about anyone with networking and linux knowledge can build a linux cluster.
What really makes a difference between an HPC cluster and your normal every day cluster is the hardware interconnects used. There is a comment in the artical that refers to not using I/O for memory and message passing. I am not quite sure what he means by that, but I am guessing that he is saying that the network is not used for shared memory/message passing (MPI/openMP/SHMEM).
If a cluster can limit the impact of latency between nodes either through smarter software or faster interconnects then I can't see any reason not to concider a linux cluster as HPC.
Clusters without smarter software tend to be a real difficult coding platforms. Some developments with things like globally shared memory might make the difference, but there will still be the problem of latency between nodes.
Oh, here's the TOP500 list, btw.
So depending on the task at hand, the cluster might perform very well, or perhaps a little less well.
Surely what you meant to say is that, depending on the task at hand, a cluster might perform very well, or perhaps perform attrociously. :-)
Clusters tend to work well when the various nodes don't need to communicate very often but you need lots of cycles for the subtasks, while dedicated supercomputers tend to perform very well in tasks requiring vast amounts of internode communications bandwidth along with large numbers of cycles. If you need vast bandwidth and relatively low numbers of cycles, your pricepoint is likely a mainframe. And if you don't need either, you get a cheap desktop machine.
Certain problems parallelize well on a cluster ... others don't. Some don't parallelize at all, and a cluster won't do you a darn bit of good. The different machines are designed for different uses ... and one should be careful not to push a "one size fits all" solution. The Cray guy clearly got it wrong on that point, and likely knows it, but he was marketting, not teaching a course in choosing hardware for the task at hand.
Cray could easily be at or close to the top of the top500 list, their X1 architecture will extend that far. However, for a lot of really important supercomputing codes, it's no contest: The cray will trounce the clusters (linux or otherwise). Those #19 crays are only 256 processors. To get similar performance a stack of xeons requires thousands of processors. Some tasks just can be split appart that easily.
A cray processor has eight floating-point units running at 800Mhz. The big Mac cluster (for example) uses G5 processors which have 2 FPUs at 2000Mhz. Thus the cray has a ~40% advantage. However, the G5 processor has ~4GB/s memory bandwidth. The Cray has ~50GB/s memory bandwidth. If you have a problem that needs to do a HUGE amount of math on a tiny amount of data, the G5 will rock. If you have a problem that needs to do a HUGE amount math on a GINORMOUS amount of data, buy the cray. (for a GINORMOUS amount of money too)
Similaraly infiniband (ala the big mac) is really hot in the cluster interconnect space because it gives 2.5GB/s per node. The Cray gives you 51GB/s.
You need to move a little data, buy a cluster. You need to move a lot of data, buy the Cray.
There's no one solution for all problems.