hey
thanks once again for the pointers. you know what bothers me most with
these IBM products? Their lack of uniformity when it comes to going
from one version to another. For ex, I am currently working on an
upgrade project from one version of a document management system to
another version of the SAME product. I do agree that there are a lot of
improvements in the later versions but these people did not carry these
enhancements through to the database level.
Now this DMS has several additions in their current version to the
previous versions. One significant change is a column in the underlying
database which was previously a FIELD stored in the database which
acted as the PRIMARY key for the table [with millions and millions of
rows] and hence INDEXED very efficiently. All was well. Right?
Now, when they move to the next version, they change this column to a
URL based PID which is 96 characters long. This URL [concatenation of
several columns in the database, hence COMPOSITE INDEX] SHOULD be
extremely efficient in retrieval because it has the server name and the
EXACT location of the ITEM embedded in it. But they made a HUGEEEEEEEE
blunder. They forgot[???????????] to include the concatenated columns
in the INDEX. And this PID is also not stored in the database. Now the
freakin retrieval is slow as HELL and I have a PMR open for this is a
MAJOR code change and I now have to figure out how in the world am I
gonna release this to all my clients..
|
