[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

WOO HOO! NFS works... (fwd)

This archive contains answers to questions sent to Unidata support through mid-2025. Note that the archive is no longer being updated. We provide the archive for reference; many of the answers presented here remain technically correct, even if somewhat outdated. For the most up-to-date information on the use of NSF Unidata software and data services, please consult the Software Documentation first.

  • Subject: WOO HOO! NFS works... (fwd)
  • Date: Tue, 29 Jan 2002 12:55:52 -0700
>To: address@hidden
>From: Steve Carson <address@hidden>
>Subject: Re: 20020128: test case for netCDF/NFS write
>Organization: RAP
>Keywords: NFS performance test

In reference to possible problem with netCDF performance on an
NFS-mounted disk; turned out not to be a netCDF problem ...

------- Forwarded Message

Date:    Tue, 29 Jan 2002 11:44:13 -0700
From:    Steve Carson <address@hidden>
To:      address@hidden
Subject: WOO HOO! NFS works... (fwd)

Russ,

Looks like our sysadmin Jeff fixed things by setting the "write
buffer size" back to 32768.

We only recently mounted this /scooby_d2 disk on etg-dev. At first it
was horrendously slow even when using 'cp' and 'ls'. Jeff experimented
with the write buffer size, and inadvertantly left it set at 1024.
This did not fix the problems. Then he applied a patch, which fixed
the 'cp' problem, but our netCDF writes were still slow. Today one of
our other engineers remembered that Jeff had set the write buffer size
to 1024, and asked Jeff to set it back to 32768. This appears to have
fixed the problem quite nicely.

Thank you for helping us verify that the problem was NOT in our
application code!

Steve C.

Forwarded message:
> From address@hidden  Tue Jan 29 11:35:47 2002
> From: Steve Carson <address@hidden>
> Message-Id: <address@hidden>
> Subject: WOO HOO! NFS works...
> To: address@hidden, address@hidden, address@hidden, address@hidden,
>         address@hidden, address@hidden, address@hidden, address@hidden,
>         address@hidden
> Date: Tue, 29 Jan 2002 11:35:46 -0700 (MST)
> Cc: address@hidden, address@hidden
> X-Mailer: ELM [version 2.5 PL0pre8]
> MIME-Version: 1.0
> Content-Type: text/plain; charset=us-ascii
> Content-Transfer-Encoding: 7bit
> 
> ETG Group & Terri & Jeff;
> 
> I have just repeated the tests I ran yesterday with MUCH better
> results:
> 
> Reading from :'/scooby_d2/ppdf/sgc/MM5/40N_105W/20020105/Lambert'
> Writing to   :'/scooby_d2/ppdf/sgc/mm5_to_nc/40N_105W/20020105/Lambert'
> 3.81u 0.93s 0:09.53 49.7%
> 
> Reading from :'/scooby_d2/ppdf/sgc/MM5/40N_105W/20020105/Lambert'
> Writing to   :'/scooby_d2/ppdf/sgc/mm5_to_nc/40N_105W/20020105/Lambert'
> 3.55u 0.90s 0:08.05 55.2%
> 
> Reading from :'/scooby_d2/ppdf/sgc/MM5/40N_105W/20020105/Lambert'
> Writing to   :'/scooby_d2/ppdf/sgc/mm5_to_nc/40N_105W/20020105/Lambert'
> 3.72u 0.93s 0:07.86 59.1%
> 
> Reading from :'/scooby_d2/ppdf/sgc/MM5/40N_105W/20020105/Lambert'
> Writing to   :'/scooby_d2/ppdf/sgc/mm5_to_nc/40N_105W/20020105/Lambert'
> 3.73u 0.80s 0:07.42 61.0%
> 
> Reading from :'/scooby_d2/ppdf/sgc/MM5/40N_105W/20020105/Lambert'
> Writing to   :'/scooby_d2/ppdf/sgc/mm5_to_nc/40N_105W/20020105/Lambert'
> 3.81u 0.67s 0:07.68 58.3%
> 
> 
> Steve C.

------- End of Forwarded Message