NEC SX-9 to be World's Fastest Vector Computer

An anonymous reader writes "NEC has announced the NEC SX-9 claiming it to be the fastest vector computer, with single core speeds of up to 102.4 GFLOPS and up to 1.6TFLOPS on a single node incorporating multiple CPUs. The machines can be used in complex large-scale computation, such as climates, aeronautics and space, environmental simulations, fluid dynamics, through the processing of array-handling with a single vector instruction. Yes, it runs a UNIX System V-compatible OS."

Comments

"up to"

[Duncan3]

The only text that can ever follow the words "up to" in computing is "0.1 *". As in "speeds of up to 0.1 * 102.4 GFLOPS". Every time a marketing droid published a press release, a kitten dies.

Grid Computing vs. Supercomputers?

[SamP2]

Hate to burst your bubble, but while grid computing can certainly achieve strong speeds, it is not quite AS fast as you might think.

The entire SETI@HOME project (biggest grid computing project on the net) pumps out 274 teraflops. By comparison, Blue Gene L (first in series) pumps out 360 teraflops, and newer versions will achieve petaflop range, much faster than similar anticipation for grid computing projects.

Sure, you might say, that just like supercomputers evolve, so does grid computing. The problem is that a supercomputer is built for a particular purpose, while grid computing is saturated by all the stuff you can do with it (SETI, protein folding, cancer research, or whatever). Now I'm not saying any of these projects is not totally awesome, nor trying to put down the spirit of the community, but as more and more projects compete with each other for user's CPU, the individual share per project will drop. If you combine all grid computing projects put together with all supercomputers put together, the supercomputers win by a huge margin, and even if every single PC on the world would be hooked to a grid project when not used for its primary purpose, it would still unlikely to beat the sum total for dedicated supercomputer power as far as raw computational capability goes.

And it gets even worse if you account for centralized management of all the computing tasks, coordination, error and fake result checking, and simply the lag of transmitting all the grid packets across the net.

Grid computing projects are a very interesting and useful concept, but they won't ever replace supercomputers. Nor should they. They each are good for their own purpose.

Re:Quite possibly.

[SamP2]

CAN run Linux and RUNS Linux are not quite the same thing.

To put things in perspective, 99% of PCs in the world CAN run Linux. :-)

Re:Quite possibly.

[deniable]

The front end OS for these things is pretty meaningless. Being a Unix like will keep the programmers and admins happy. The front-end is only a shell for the code running on the back-end processing units. These do all of the work and rely on specific hardware, instructions, and libraries to do things in *actual* parallel. These things basically exist to run big number crunching tasks for mathematicians and mathematicians in disguise like physicists. :) These people will generally be running their own code with tweaks for the hardware. They see Intel's SIMD instructions and think it's 'cute' and wonder what it will be when it grows up.

Re:vector machines in the top500 list refuse to di

[cerberusss]

Will vector machines live forever?

Well, I actually doubt it. You could say 'those vector processors are used for matrix calculations and are wildly different from general purpose CPUs' and you'd be right.

However, I could see a point in time where hybrids like the Cell (one scalar processor and eight vector processors) will become so cheap that the number of vector machines will decline even more.

The idea will never die of course, I mean, hardware is so flexible nowadays that a good student could make a vector processor at home, if he had a development board with a fast Xilinx FPGA on it. But I think the decline will continue if hybrids will be used more often.

Re:Quite possibly.

[Chrisq]

This was certainly the case when I used vector processors. It is possible that the vector processor does not run an OS at all. It has been many years since I have worked on such a beast but when I did we ran a loader system with a standard OS which would cross compile code for the processor and load it almost onto the hardware (there was actually a small program we called a monitor to deal with I/O, etc, but no multi-tasking, security or anything). It would then run and the results were read back into the front-end computer.

Re:Logical question:

[ajs318]

Something people constantly misunderstand ..... is that they assume that all problems can be broken apart into manageable portions that can be split

Exactly! Having sex 39 times does not mean you will be able to get a baby in one week. Some operations are by nature sequential -- and while there is scope for some parallelisation, doing so in a highly-distributed fashion can end up increasing latency, because you end up spending more time splitting the data up and putting the results back together than actually doing any maths.

A vector processor basically has several separate logic matrices that work in parallel, performing the same operation on different (sets of) operands simultaneously. When you want to solve simultaneous equations in many variables, some of which are themselves multi-dimensional vectors, that can be extremely useful.