That worked great - THANKS!
CREATE INDEX sct_descriptions_k2
ON sct_descriptions
USING btree
(term_index varchar_pattern_ops);
I noticed I had to keep the original index for the non-like operator
but that is not a big deal
CREATE INDEX sct_descriptions_k1
ON sct_descriptions
USING btr
>>> "Greg Caulton" <[EMAIL PROTECTED]> wrote:
> but I get a sequential scan when I do where term_index like
>
> select * from sct_descriptions where term_index like 'CHILLS AND
FEVER
> (FINDING)'
> Is there anything else I can do? Settings below, this is PostgreSQL
8.3
> "lc_collate";"Engl
Hi,
I have a table sct_descriptions which I have vacuumed, analyzed and
reindexed. The index is on term_index
INFO: analyzing "public.sct_descriptions"
INFO: "sct_descriptions": scanned 3000 of 22861 pages, containing
91877 live rows and 0 dead rows; 3000 rows in sample, 700133 estimated
total
