ca.portal.admin

Duplicate index keys

Discussion created by ca.portal.admin on Aug 6, 2009
Latest reply on Aug 6, 2009 by ca.portal.admin
When defining an index key that is composed of multiple fields and
duplicate keys are allowed the duplicates characteristics of the
individual fields has no impact. IDMS considers the value of the entire
concatenated key when determining if a duplicate key exists. If FIELDA
has many duplicate values and FIELD2 has very few duplicates then your
index will have very few duplicate keys. =20

=20

Dick Weiland

=20

CA

Senior Software Engineer

CA-IDMS Level-II Support

=20

Phone 630-505-6561

Fax: 630 505-6097

Email: Richard.Weiland@ca.com <mailTo:Richard.Weiland@ca.com>=20

=20

=20

=20

=20

=20

=20

=20

=20

=20

=20

=20

=20

=20
"
IDMS 3rd-party providers forum
IDMSVENDOR-L@LISTSERV.IUASSN.COM
SMTP
IDMSVENDOR-L@LISTSERV.IUASSN.COM
IDMSVENDOR-L@LISTSERV.IUASSN.COM
SMTP








Normal

Normal
Re: Duplicate index keys
"Wellllll.
Duplicate keys is the least of your problems if you have such a
situation as do we . . .
We might have hundreds of thousands, even millions, of very large keys
(30 - 60 bytes) with the first half are duplicates and the last half
might not be, but this causes indexing to have many intermediate records
with identical keys, so that the search algorithm (IMVHO) for a record
somewhere within that range is not efficient, if you see what I mean . .
.

Plus doesn't this lend itself to hotspots in the index, where
split-and-spawns are likely to occur?

Outcomes