| From: | Daniel Gustafsson <daniel(at)yesql(dot)se> |
|---|---|
| To: | Tomas Vondra <tomas(at)vondra(dot)me> |
| Cc: | wenhui qiu <qiuwenhuifx(at)gmail(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: PGLZ Compression Optimization |
| Date: | 2026-07-16 12:16:14 |
| Message-ID: | B911DCE7-0146-488A-994C-88E8FB1F1058@yesql.se |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
> On 13 Jul 2026, at 20:17, Tomas Vondra <tomas(at)vondra(dot)me> wrote:
> On 7/13/26 13:15, wenhui qiu wrote:
>> The patch also optimizes the legacy path with a rolling hash, indexed
>> history rings, bounded candidate searches, word/SIMD match extension,
>> and an AVX2 implementation selected by PostgreSQL's runtime CPU check.
>> Other platforms use PostgreSQL's existing SIMD abstraction.
>>
>
> This, on the other hand would be a *huge* improvement for all existing
> systems with a lot of data compressed with pglz (and there's fair amount
> of those). If we could just make them faster by optimizing the code,
> that seems very desirable.
+1, if optimizing the code can unlock performance improvements while being
backwards compatible it would be a very big win.
--
Daniel Gustafsson
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Robert Haas | 2026-07-16 12:19:50 | Re: walsummarizer can get stuck when switching timelines |
| Previous Message | Peter Eisentraut | 2026-07-16 12:03:02 | Re: Type assertions without GCC builtins |