[Beowulf] Compare and contrast MPI implementations
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Greg Lindahl lindahl at pathscale.comTue Dec 13 23:00:43 PST 2005
- Previous message: [Beowulf] Compare and contrast MPI implementations
- Next message: [Beowulf] Compare and contrast MPI implementations
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Fri, Dec 09, 2005 at 01:16:21PM -0800, David Mathog wrote: > What advantages and disadvantages do each of these have? I see no one else wants to volunteer to answer this one... my impression has been that the reason some folks suggest using LAM is mainly because LAM has a better tcp driver than MPICH-1 has. Other than that there isn't much difference. Just looking at an strace of what MPICH-1 does gives me the willies. But I don't think that means anything is wrong with the overall design, it's just some implementation details gone wrong in that one device. I've never looked at MPICH-2's tcp driver, perhaps it's better. > Are the MPI 2 standards fully backwards compatible with code written > to the MPI 1.2 standard? Yep. -- greg
- Previous message: [Beowulf] Compare and contrast MPI implementations
- Next message: [Beowulf] Compare and contrast MPI implementations
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
