[Beowulf] Athlon64 / Opteron test
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.comFri May 14 13:43:51 PDT 2004
- Previous message: [Beowulf] Athlon64 / Opteron test
- Next message: [Beowulf] Opteron and 32-bit libraries
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On Fri, May 14, 2004 at 11:49:34AM -0400, Robert G. Brown wrote: > rgb at s02|B:1003>./Ospin > -bash: ./Ospin: /lib/ld-linux.so.2: bad ELF interpreter: No such file or directory > rgb at s02|B:1004> Pilot error. You have to install a couple of additional rpms to run 32-bit stuff on an Opteron. The annoying thing about it is that they have the same names as x86_64 packages, grrrr. In this case you need glibc-*.i686.rpm. > ...it also refers to anecdotal accounts that numerical performance > significantly degrades if one runs i386 code compared to recompiled > x86_64 code. The hard thing about anecdotes is that they're hard to disprove. However, if you look at SPECfp scores for Opteron using the Intel compiler (32-bit mode) and the PathScale compiler (64-bit mode), I don't think that you'd come to that conclusion. It is an apples to oranges comparison: our compiler has a better optimization framework. But the benefit from 64-bit mode is clearly moderate. I talk about it in a recent ClusterWorld article. -- greg
- Previous message: [Beowulf] Athlon64 / Opteron test
- Next message: [Beowulf] Opteron and 32-bit libraries
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
