Just how often do you think you have data duplicated in the IFS that you
would want deduplicated? For example if I do a CRTDUPOBJ of the IIM file
into a test library that is, in effect, duplicating something from
What would deduplication do in this case?
Or do you want deduplication to only work on stream file directories?
IBM has something like this with Domino and DAOS. In which case Marketing
Dude sends out an email with PowerPointFromHell.ppt as an attachment in an
email to 100 people in the company it stores one copy as a separate file
(.nlo) in a different directory and every email just links to that file.
Of course, as I type this up I'm in the process of bouncing one such
Domino server to diagnose file corruption of these .nlo files and/or their
links. (Oops it just died with a "Panic" error.)
What's a little corruption when it saves a serious load of disk space? :-(
And if I try to OS copy one mail file from one server to another it
doesn't know where the links are but you learn to deal with stuff with