PostgreSQL LIKE query performance variations

The previously accepted answer was incorrect.

Full Text Search with its full text indexes are not for the LIKE operator at all, they have their own operators and don't work for arbitrary strings. They operate on words based on dictionaries and stemming.

Install the additional module pg_trgm which provides operator classes for GIN or GiST indexes to support (almost) all LIKE patterns, not just left-anchored ones:

Example index:

CREATE INDEX tbl_col_gist_trgm_idx ON tbl USING gist (col gist_trgm_ops);

Example query:

SELECT * FROM tbl WHERE col LIKE '%foo%';  -- wild card at beginning

For just left-anchored patterns you get the optimum with a suitable operator class for a B-tree index: text_pattern_ops or varchar_pattern_ops (part of standard Postgres, no additional module needed).

Example index:

CREATE INDEX tbl_col_text_pattern_ops_idx ON tbl(col text_pattern_ops);

Example query:

SELECT * FROM tbl WHERE col LIKE 'foo%';  -- no wild card at beginning

More details, explanation, examples and links in these related answers on dba.SE:

• Pattern matching with LIKE, SIMILAR TO or regular expressions in PostgreSQL
• How is LIKE implemented?

You could install Wildspeed, a different type of index in PostgreSQL. Wildspeed does work with %word% wildcards, no problem. The downside is the size of the index, this can be large, very large.

Possibly the fast ones are anchored patterns with case-sensitive like that can use indexes. i.e. there is no wild card at the beginning of the match string so the executor can use an index range scan. (the relevant comment in the docs is here) Lower and ilike will also lose your ability to use the index unless you specifically create an index for that purpose (see functional indexes).

If you want to search for string in the middle of the field, you should look into full text or trigram indexes. First of them is in Postgres core, the other is available in the contrib modules.

Your like queries probably cannot use the indexes you created because:

1) your LIKE criteria begins with a wildcard.

2) you've used a function with your LIKE criteria.