[PBS-USERS] Re: Any experience with SCore?
Eray Ozkural (exa)
erayo at cs.bilkent.edu.tr
Wed Oct 31 03:17:40 PST 2001
-----BEGIN PGP SIGNED MESSAGE-----
On Tuesday 30 October 2001 18:32, Yong Duan wrote:
> Tends to over rate the lower-end clusters. It wouldn't make much difference
> if you use 10BT (or even shoe strings for that matter). Communication is
> basically not counted in the benchmark. Useful for certain type of
> applications, useless for most ...
Could you elaborate? It certainly looks better than LAM where we have latency
in terms of miliseconds rather than microseconds. ;)
Are you telling me that I wouldn't have any improvement in my code if I
switched from LAM/MPI to SCoop mpich implementation?
Eray Ozkural (exa) <erayo at cs.bilkent.edu.tr>
Comp. Sci. Dept., Bilkent University, Ankara
GPG public key fingerprint: 360C 852F 88B0 A745 F31B EA0F 7C07 AE16 874D 539C
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.6 (GNU/Linux)
Comment: For info see http://www.gnupg.org
-----END PGP SIGNATURE-----
More information about the Beowulf