Maxwell, A 64 FPGA Supercomputer (pdf) • Hacker News Board

Re: Maxwell, A 64 FPGA Supercomputer (pdf)

#21

Post by modeless » Thu, Sep 08, 2011, 10:58 PM UTC

jedbrown wrote: post #19

modeless wrote: post #17

Earlier quoted context omitted.

They address software development cost at the end of the paper, and basically conclude that FPGAs are too hard to program to replace CPUs yet, even for highly specialized HPC applications.

My point is that they should demonstrate what the potential hardware advantage is before evaluating whether the software development costs are worthwhile. There are a few HPC tasks that are very simple and possibly amenable to specialized hardware. D.E. Shaw's Anton is an extreme example of pursuing dedicated hardware. But the metric needs to involve purchasing and power costs for the hardware. FPGAs can come out loo…

I disagree that software development cost shouldn't be considered before hardware cost. The software development costs are so much higher for FPGAs (both money and time) that they probably outweigh the hardware costs for many applications. They are both important and equally worth investigation.

Re: Maxwell, A 64 FPGA Supercomputer (pdf)

#22

Post by kristianp » Fri, Sep 09, 2011, 4:31 AM UTC

neopallium wrote: post #20

alain94040 wrote: post #9

Not to rain on anyone's parade, but next to my desk is a machine about 10 times more powerful than what this paper describes. The main issues with the setup in the paper: latency between nodes. Gigabit Ethernet is just not good for latency. To nitpick some more, Virtex-4 is getting old.

Gigabit Ethernet is mainly used as a control network. Each FPGA has 4 RocketIO links (2-3Gb/s) and they are directly connected in a 2-D torus of point-to-point links. The RocketIO network is used for nearest-neighbor communication patterns. For reduction operations such as global sums, they call back to the host CPUs for MPI reduction operations to be preformed over the GE network. Also one of the main goals was to b…

s/cheaper then the/cheaper than the/g s/then just/than just/g

Re: Maxwell, A 64 FPGA Supercomputer (pdf)

#23

Post by jedbrown » Fri, Sep 09, 2011, 8:30 AM UTC

modeless wrote: post #21

jedbrown wrote: post #19

Earlier quoted context omitted.

My point is that they should demonstrate what the potential hardware advantage is before evaluating whether the software development costs are worthwhile. There are a few HPC tasks that are very simple and possibly amenable to specialized hardware. D.E. Shaw's Anton is an extreme example of pursuing dedicated hardware. But the metric needs to involve purchasing and power costs for the hardware. FPGAs can come out loo…

I disagree that software development cost shouldn't be considered before hardware cost. The software development costs are so much higher for FPGAs (both money and time) that they probably outweigh the hardware costs for many applications. They are both important and equally worth investigation.

If the hardware cost benefit is not large, there isn't any point discussing software development cost. If the benefit is huge, then the software discussion is relevant. In other words, the decision table looks something like

                  HW efficient     HW inefficient
  SW cheap        Yes              No
  SW expensive    Maybe            No

Re: Maxwell, A 64 FPGA Supercomputer (pdf)

#24

Post by dhjones » Fri, Sep 09, 2011, 5:05 PM UTC

modeless wrote: post #18

sliverstorm wrote: post #15

Earlier quoted context omitted.

And why should it? The CPU and the GPU are entirely different beasts.

That was true 5 years ago, but not today; GPUs and CPUs are converging. GPUs are quite suitable for all of the applications described in the article, and provide orders of magnitude speedup over CPUs while remaining relatively cheap and easy to program vs. FPGAs.

I couldn't agree with this more.

Re: Maxwell, A 64 FPGA Supercomputer (pdf)

#25

Post by sliverstorm » Sat, Sep 10, 2011, 3:31 AM UTC

modeless wrote: post #18

sliverstorm wrote: post #15

Earlier quoted context omitted.

And why should it? The CPU and the GPU are entirely different beasts.

That was true 5 years ago, but not today; GPUs and CPUs are converging. GPUs are quite suitable for all of the applications described in the article, and provide orders of magnitude speedup over CPUs while remaining relatively cheap and easy to program vs. FPGAs.

Converging in what sense?