| From: | Robert Haas <robertmhaas(at)gmail(dot)com> |
|---|---|
| To: | Josh Berkus <josh(at)agliodbs(dot)com> |
| Cc: | pgsql-hackers(at)postgresql(dot)org |
| Subject: | Re: COUNT(*) and index-only scans |
| Date: | 2011-10-12 02:50:11 |
| Message-ID: | CA+TgmobQrP+dH1rmss8nGm5bRDPANM+nT2p1m6ovsVTpayE1_g@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Tue, Oct 11, 2011 at 3:18 PM, Josh Berkus <josh(at)agliodbs(dot)com> wrote:
>
>> The trouble is that if we VACUUM and then ANALYZE, we'll often get
>> back a value very close to 100%, but then the real value may diminish
>> quite a bit before the next auto-analyze fires. I think if we can
>> figure out what to do about that problem we'll be well on our way...
>
> It's not so much an issue of when the last auto-analyze was as an issue
> of the number of rows in write transactions against that table in the
> last X minutes. This is where it really hurts us that
> pg_stat_user_tables is not time-based.
The number of write transactions in the last X minutes seems pretty
much irrelevant.
What matters is the number of previously-all-visible pages written
since the last vacuum.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Robert Haas | 2011-10-12 04:12:08 | Re: Index only scan paving the way for "auto" clustered tables? |
| Previous Message | Jun Ishiduka | 2011-10-12 02:43:59 | Re: Online base backup from the hot-standby |