Xref: utzoo comp.protocols.nfs:2417 comp.protocols.tcp-ip.ibmpc:6084
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!spool.mu.edu!uunet!stanford.edu!leland.Stanford.EDU!mikef
From: mikef@leland.Stanford.EDU (Michael Fallavollita)
Newsgroups: comp.protocols.nfs,comp.protocols.tcp-ip.ibmpc
Subject: SOSS Performance - Is it really this bad?
Keywords: SOSS PC-NFS NFS performance
Message-ID: <1991Jun5.055230.24965@leland.Stanford.EDU>
Date: 5 Jun 91 05:52:30 GMT
Article-I.D.: leland.1991Jun5.055230.24965
Sender: mikef@leland.Stanford.EDU (Michael Fallavollita)
Distribution: usa
Organization: AIR, Stanford University
Lines: 24

The small business I'm associated with was hoping that SOSS was the solution
to it's networking problems, so we set up a test with one PC running SOSS,
and the other PC running PC-NFS 3.01.  As a test, we attempted to transfer
a 0.5MB file from the Server to the Client.  This transfer took 14 seconds!!!
Attempting to copy the file from one directory on the Server to another
directory on the Server caused a stack error and crash on the Server.  The
server was running PC-Kwik as well as Qemm.

Has anyone seen similar performance, or can anyone report much better
performance?  I can't believe that it is this bad, but I can't think of
anything that could be causing problems.  The same ethernet cables are
currently supporting a LAN running PCI but I don't think that should cause
any interference.  How does PC-NFS 3.5 differ as far as performance is
concerned?

Any suggestions/pointers/information would be greatly appreciated.

				Mike Fallavollita
				mikef@joshua.stanford.edu
-- 
                o
               o 
                \\
         -----  ||     Mike Fallavollita