[PRL] Parallel Performance: Optimize Managed Code For Multi-Core Machines -- MSDN Magazine, October 2007
Riccardo Pucella
riccardo at ccs.neu.edu
Tue Sep 25 06:01:40 EDT 2007
On Sep 24, 2007, at 8:22 PM, Mitchell Wand wrote:
> shared http://msdn.microsoft.com/msdnmag/issues/07/10/Futures/
> default.aspx
>
> Multi-processor machines are now becoming standard while the speed
> increases of single processors have slowed down. The key to
> performance improvements is therefore to run a program on multiple
> processors in parallel. Unfortunately, it is still very hard to
> write algorithms that actually take advantage of those multiple
> processors. In fact, most applications use just a single core and
> see no speed improvements when run on a multi-core machine. We need
> to write our programs in a new way.
> ...
I wonder how they get the performance up so much on the first
Parallel.For example over the multi-threaded implementation; I mean,
internally, they must use something similar, unless they have a
difference interface to theads running on multiple cores as opposed
to software threads.
Note that one of the authors is Daan Leijen, who was a student of
Erik Meijer back in his Haskell days.
Cheers,
R
More information about the PRL
mailing list