[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Fsuk-manchester] Open ARM GPU drivers
From: |
Bob Ham |
Subject: |
Re: [Fsuk-manchester] Open ARM GPU drivers |
Date: |
Thu, 07 Feb 2013 09:44:23 +0000 |
User-agent: |
RoundCube Webmail/0.3.1 |
On Tue, 05 Feb 2013 17:42:02 +0000, Dave Love <address@hidden> wrote:
> Bob Ham <address@hidden> writes:
>
>>> If so, what
>>> sort of performance might Mali 400, for instance, provide?
>>
>> Good question. I can't imagine you'd get anywhere near the kind of
>> performance of a desktop graphics chip due to the power constraints.
>
> Yes, but the comparison wouldn't be with other GPUs, but with SIMD
> (SSE-type) units.
I'm confused by this I must say. I thought GPUs *were* SIMD units. In
fact, looking at Wikipedia, it explictly lists GPUs as an example of SIMD
architecture:
http://en.wikipedia.org/wiki/Flynn%27s_taxonomy
>> I'm not surprised; the focus of most efforts are intended to support
>> things like simulations, computational biology and other scientific
>> computing, often in supercomputers.
>
> [That's what I do, for some value of "supercomputers".]
Cool. You'll have to excuse me then for trying to "teach your grandmother
how to suck eggs", as it were :-)
>> In that kind of context it doesn't make much sense to target low
>> power, performance-constrained chips like the Mali.
>
> It might, if you have them anyway in a well-ARMed system intended for
> specific sorts of jobs, even if you wouldn't use them for HPC.
It may do, yes, and looking at the power consumption numbers for the
Epiphany processors, I can imagine that there may be much more interest in
future.
Really though the point I was making, and please correct me if I'm wrong,
was that in HPC the general focus is on high-powered processors like
Xeon/Opteron CPUs and ATI/NVIDIA GPUs. For example, Titan has Opteron CPUs
and Tesla GPUs, Tianhe-1A has Xeon CPUs and Tesla GPUs upgraded from Xeon
CPUs and Radeon GPUs, and our own HECToR has Opteron CPUs.
>> (Unless, of course, you're able to engineer your own floating-point
>> applications directly for the hardware.
>
> The appeal of OpenCL is in avoiding some level of hardware specifics
> generally, though you typically do have to write hardware-specific
> assembler to get performance in the kernels :-(.
Hmm.. that is disappointing. It does kind of defeat the purpose of OpenCL
:-/
--
Bob Ham <address@hidden>
for (;;) { ++pancakes; }