[Mitgcm-support] RE: joinmds pickup problem

[mitgcm-support at dev.mitgcm.org]

Tom,

 Stephanie and Helen ( stephd at plume.mit.edu, helen at plume.mit.edu )
have various joinmds[r] driver scripts that I wrote for merging large
amounts
of stuff semi-automatically. You could check with them ( cc'd )
to see if they are useful for you.

Chris

-----Original Message-----
From: Tom Haine [mailto:twnh at stommel.eps.jhu.edu]
Sent: Monday, July 01, 2002 4:07 PM
To: support at mitgcm.org
Cc: Thomas Haine
Subject: joinmds pickup problem


Hi,

I'm trying to pick up a 16 processor calculation by joining the individual
pickup files with joinmds (ver 0.3.0):

joinmds pickup 0000904320

works OK, except the file I get (pickup.0000904320.data) is the same size
as the individual files used to create it
(pickup.0000904320.00?.00?.data). And, I can't restart the calculation
with pickup.0000904320.data:

 attempt to read non-existent record (direct access).
 File name = pickup.0000904320 unformatted, direct access record = 56450
 In source file mdsio_readfield.f, at line number 1453

When I do:

joinmds T 0000904320

I get T.0000904320.data that's 16x bigger than each tile, as I expect.

Any ideas where I'm going wrong?

Also, does anyone have scripts to automate the joinmds process?  Reducing
the tens of thousands of output files by 32x would help. And running each
joinmds by hand isn't practical!

Thanks, Tom.
--------------------------------------------
 Thomas W. N. Haine

 Associate Professor of Physical Oceanography,
 Department of Earth & Planetary Sciences,
 329 Olin Hall, The Johns Hopkins University,
 Baltimore, MD 21218, USA.
 Tel : 410 516 7048,   Fax : 410 516 7933
 Thomas.Haine at jhu.edu
 http://www.jhu.edu/~eps/faculty/haine
--------------------------------------------