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

960129: Pb with netcdf on Cray T3D

>From: Serge Wuthrich <address@hidden>
>Organization: ?
>Keywords: 199601291638.AA22274 netCDF CRAY T3D

Serge,

>I'm trying to install netcdf on Cray T3D. I used the following
>environment variables:
>
>setenv CC /mpp/bin/cc
>setenv CFLAGS -Tcray-t3d
>
>The ./configure script runs successfully. I had to change the
>test xdr/Makefile by adding a -X1 (i.e. to tell that I want only one node
>of the T3D). However, the test failed with the folowing error message:
>
>        ./xdrtest > xdrtest_out.new
>Assertion failed: *fp < floats[ii] + EPSILON && *fp > floats[ii] - EPSILON, fi
> le xdrtest.c, line 344
>
>I suspect that during the compilation of the library, some Unicos variables
>have been used instead of the T3D specific variables (since the T3D is not
>a Unicos but a 64-bit IEEE machine).
>
>Do you have some experience with the port of netcdf on the T3D or do 
>you know somebody who did it sucessfully ?

We have had some contact with users interested/pursuing this port.  You
see the transactions that we have had with them through a query function
accessible on our netCDF WWW Homepage:

http://www.unidata.ucar.edu/packages/netcdf/index.html

In the section entitled Search Mail Archives, you would enter:

cray and t3d

as the search parameters.  When you do this, you will get a list of
all email transactions that we have had that deal with this subject.
The email message that had the "best" match is included below.

>Many thanks in advance,
>
>serge Wuthrich
>address@hidden

>From address@hidden Thu Dec 14 02:50:35 1995
Message-Id: <address@hidden>
Full-Name: General Support
To: "Jeff Young (919)541-3929" <address@hidden>
Subject: 951214: netCDF on the Cray T3D 
In-Reply-To: Your message of "Thu, 14 Dec 95 11:33:55 EST."              
<address@hidden> 
Date: Thu, 14 Dec 1995 09:50:35 -0700
From: Unidata Support <address@hidden>

>From: "Jeff Young (919)541-3929" <address@hidden>
>Organization: EPA
>Keywords: 199512141635.AA05553 netCDF CrayT3D

Jeff,

>Has netCDF been developed for the Cray T3D?

Probably. Someone in the community has been working on this.  Here are
some comments on the subject available through the email search
function of the netCDF WWW Homepage:

http://www.unidata.ucar.edu/packages/netcdf/index.html

 ...
> I am wondering if anybody made (or is planing on making) a parallel
> implementation of NetCDF on machines like Intel Paragon, TMC CM-5,
> Cray T3D, etc. which would take advantage of available parallel IO on
> the target machine.

Yes, Rick Light from Los Alamos (address@hidden) was very close to
completing a port of netCDF to a CM-5 and had even managed to exploit some
parallelism, if I remember correctly from the last time I talked to him
about this.

Also, we have had contact from programmers from Cray Computer Corporation
about porting netCDF to the Cray 3 in a way that exploited vectorization for
floating-point reads and writes.  I know Chris Anderson
(address@hidden) from NERSC has some interest in porting netCDF to a
T3D.  Tim Sheehan (address@hidden) from ORNL has tried to port
netCDF to an Intel Paragon though I don't know if he succeeded.  Harry
Moffat (address@hidden) from Sandia Labs was also interested in
getting netCDF ported to an nCube and a Paragon.  Mike Wehner from LLNL
(address@hidden) was working on porting netCDF to a Paragon last
fall, but I think he ran into problems with Intels XDR library, and I don't
know whether he solved them.

> I have posted this question to the Newsgroup sci.data.formats and
> received responses indicating a strong interest in the field. Also, I
> attended the DOE/CHAMMP (global climate change) meeting last week at
> Albuquerque and there was also a strong interest in having a parallel
> implementation of NetCDF (the NCAR's CCM2 will be using NetCDF soon).

We aren't currently planning on developing our own port to any parallel
machines because we don't have convenient access to any parallel machines,
and it's not in Unidata's current mission to develop software for such
platforms since none of our sites are now using parallel machines.  However,
we may soon find a need to exploit multithreading on SPARCstations and other
workstations with multiple CPUs, so we may eventually have to addreess the
issues.


Please let us know if you can't find the information you are after using
this facility.

Tom Yoksas