| From: | Andrew Borodin <borodin(at)octonica(dot)com> |
|---|---|
| To: | Teodor Sigaev <teodor(at)sigaev(dot)ru> |
| Cc: | pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Review: GIN non-intrusive vacuum of posting tree |
| Date: | 2017-03-23 05:03:01 |
| Message-ID: | CAJEAwVGZP0iyt7=oDxbCh1OExa=7xmQ1PfkezXMxvz6cCTq8vQ@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
2017-03-22 22:48 GMT+05:00 Teodor Sigaev <teodor(at)sigaev(dot)ru>:
> hasEmptyChild? and hasNonEmptyChild (BTW, isAnyNonempy has missed 't')
Yes, I think this naming is good. It's clear what's in common in these
flags and what's different.
> And if the whole posting tree is empty,then we could mark root page as leaf
> and remove all other pages in tree without any locking. Although, it could
> be a task for separate patch.
From the performance point of view, this is a very good idea. Both,
performance of VACUUM and performance of Scans. But doing so we risk
to leave some garbage pages in case of a crash. And I do not see how
to avoid these without unlinking pages one by one. I agree, that
leaving this trick for a separate patch is quite reasonable.
Best regards, Andrey Borodin.
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Andrew Gierth | 2017-03-23 05:09:46 | Re: Hash support for grouping sets |
| Previous Message | Amit Kapila | 2017-03-23 04:58:47 | Re: Patch: Write Amplification Reduction Method (WARM) |