On 07/01/2014 12:17 AM, Matt Wallis wrote:
On 01/07/14 13:45, Jonathan Aquilina wrote:
This question probably sounds like a stupid one, but what difference
in an
HPC environment and to parallel written code does compiler version make?
Depends on the day of the week, the processor, the code, and your
cluster really.
Sometimes code from GCC is faster, sometimes code from the Intel
compiler is faster.
Some people write code that only compiles with the Intel compiler, or
link against libraries distributed with the compiler.
Then of course you might have a BlueGene with PPC cores rather than
x86, and while the difference in performance between GCC for PPC and
the IBM XL compilers is slight these days, it wasn't always the case.
One of my researchers has found that his code that is integer-heavy runs
up to 5x faster on our BG/P when compiled with GCC over the same code
compiled with the IBM XL compilers. For floating point heavy codes, the
IBM XL compiles produce much faster code.
Intel also distributes highly tuned versions of some maths libraries
with their compilers, and libraries designed for offloading work to
the the Xeon Phi, without recompiling the application.
And then when it comes time to debug why the application doesn't work,
commercial compiler suites often come with very good debugging tools.
Matt.
_______________________________________________
Beowulf mailing list, [email protected] sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
