Subject: RE: G-ZIP test
From: Erik Katsavounidis (Erik.Katsavounidis@lngs.infn.it)
Date: Wed Jun 21 2000 - 22:29:26 EDT
From: VAXGS::KATSVNDS "Erik Katsavounidis" 22-JUN-2000 04:28:57.78
To: VAXGS::ZACCHEO
CC: MACRO-GS@LNGS.INFN.IT,KATSVNDS
Subj: RE: G-ZIP test
Having the GZIP run on the RAW file rather than the record-fixed one
has to do with the fact that record-fixing of a RAW file DROPS out the
last part of the file that does not add up to an integer multiple of
the record length, i.e., the input file in wrong. GZIP can indeed work
with any file, independent of fixed or variable records. Now, why GZIP
before VMSTOTAR: I do not have a good answer besides maybe the fact that
four years ago we went through the same problem of putting the MUON DSTs
on tape (which are unformatted variable length fortran files) and we concluded
that the scheme I described and is implemented in the special TAPEMAKER
*worked* for them. So, trying the smae for the GC DSTs is my best shot.
Let's get these GC's on tape and have the project moving.
--Erik
This archive was generated by hypermail 2a24 : Wed Jun 21 2000 - 22:29:32 EDT