Prev: Fast Coarse Spectrum Analysis of Sound Samples in Java
Next: Which digital input class D audio amplifier ?
From: Manny on 9 Aug 2010 18:01
On Aug 9, 9:58 pm, Fred Marshall <fmarshall_xremove_the...(a)xacm.org>
wrote:
> The Portland Group has been around for a long time working on the
> software technology for these things.  I just read from their website:
> "A CUDA programmer is required to partition the program into coarse
> grain blocks that can be executed in parallel."
> Some related folks (in the late '80s) talked about optimizing compilers
> that would target heterogeneous parallel machines based on a "partition
> spec".  Sounds like the quote above, eh?
> The idea was to write the program, specify the partitioning and then
> manually iterate the partitioning to remove bottlenecks.
> I don't sense that things have improved all that much in the last 20
> years.  Well, compute power has certainly increased but our underlying
> technical ability to make good/general use of parallel machines is still
> at issue.

The third generation of ESL tools is very credible. That said, it is
not for the faint-hearted and at least some sort of insight into
what's going on underneath is required by the user. From an HDL
standpoint, I find that not having to fiddle with low-level timing to
be a huge step forward. Still serious users will find themselves at
least writing something to purpose on top of these tools, which is
hard. And if you'r a bit pedantic about your methodology i.e. a bit
hung up on form, that could mean a serious solo development effort.

-Momo

First  |  Prev  | 
Pages: 1 2
Prev: Fast Coarse Spectrum Analysis of Sound Samples in Java
Next: Which digital input class D audio amplifier ?