| From: | Hannu Krosing <hannu(at)2ndQuadrant(dot)com> |
|---|---|
| To: | Andres Freund <andres(at)2ndquadrant(dot)com>, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
| Cc: | Dimitri Fontaine <dimitri(at)2ndQuadrant(dot)fr>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Save Hash Indexes |
| Date: | 2013-11-01 18:13:49 |
| Message-ID: | 5273EF5D.5060503@2ndQuadrant.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 11/01/2013 03:49 PM, Andres Freund wrote:
> On 2013-11-01 09:49:57 -0400, Tom Lane wrote:
>> Lastly: what real-world problem are we solving by kicking that code
>> to the curb?
> It makes hashed lookups much easier to use. Currently if you want
> indexed access over wide columns and equality is all you need you need
> to write rather awkward queries to make that happen in each query,
> something like:
> WHERE hash(column) = hash('value') AND column = 'value'.
>
> So some magic that hides having to do that would be nice, even it
> doesn't have O(log(n)) properties.
Is'nt there a way to do this using a special operator class ?
> The interesting part is that the
> index gets much denser and smaller, and that the individual comparisons
> are much cheaper.
>
> Greetings,
>
> Andres Freund
>
--
Hannu Krosing
PostgreSQL Consultant
Performance, Scalability and High Availability
2ndQuadrant Nordic OÜ
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Andres Freund | 2013-11-01 18:31:07 | buffile.c resource owner breakage on segment extension |
| Previous Message | Tom Lane | 2013-11-01 16:51:34 | pgsql: Remove internal uses of CTimeZone/HasCTZSet. |