> Your CPUs support HyperThreading (look at /proc/cpuinfo - your CPUs
> will have the ht flag set), a feature of nearly all P4s that turns
> your single CPU into 2 CPUs (albeit ones that share their caches). An
> attempt by Intel to try offset the very long pipeline of P4 and high
> cost of pipeline stalls/flushes.
>> For single-task, cache intensive workloads (eg computational), HT
> costs performance. For most other workloads, HT is a win.
I see.
The only place where this would make a difference is MPI-CH, as
it can be told how many processors there are per host. Has anyone
here worked sufficiently with MPI-CH to make a recommendation?
I _think_ that most of the stuff running here is more cpu bound
than i/o bound, but this probably change.
Maintained by the ILUG website team. The aim of Linux.ie is to
support and help commercial and private users of Linux in Ireland. You can
display ILUG news in your own webpages, read backend
information to find out how. Networking services kindly provided by HEAnet, server kindly donated by
Dell. Linux is a trademark of Linus Torvalds,
used with permission. No penguins were harmed in the production or maintenance
of this highly praised website. Looking for the
Indian Linux Users' Group? Try here. If you've read all this and aren't a lawyer: you should be!
