| From: | Guyren Howe <guyren(at)gmail(dot)com> |
|---|---|
| To: | "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org> |
| Subject: | Re: jsonb case insensitive search |
| Date: | 2017-06-02 00:00:42 |
| Message-ID: | 066ECF7C-8C9D-45F5-A7BA-3FC31870C071@gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
On Jun 1, 2017, at 13:44 , Karl Czajkowski <karlcz(at)isi(dot)edu> wrote:
> If I remember correctly, isn't a compound index always just using
> btree? In general, I have found better luck using several smaller
> btree indices than one large compound one. Unless your entire query
> can be answered from an index-only lookup, the extra columns just
> bloat the btree index.
The issue is *much* more subtle than this. For repetitive queries, a compound index can be a great speedup.
Best discussion I’ve seen around index design is on the website use-the-index-luke.com.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Wetzel, Juergen (Juergen) | 2017-06-02 05:51:49 | Re: Slow query plan used |
| Previous Message | David G. Johnston | 2017-06-01 22:42:21 | Re: repmgr cascade replication node delay |