| From: | Jason Petersen <jason(at)citusdata(dot)com> |
|---|---|
| To: | Atri Sharma <atri(dot)jiit(at)gmail(dot)com> |
| Cc: | Bruce Momjian <bruce(at)momjian(dot)us>, Peter Geoghegan <pg(at)heroku(dot)com>, Greg Stark <stark(at)mit(dot)edu>, Ants Aasma <ants(at)cybertec(dot)at>, Pg Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: Clock sweep not caching enough B-Tree leaf pages? |
| Date: | 2014-04-18 20:11:04 |
| Message-ID: | 5D48DA93-F622-4DA3-B9A3-2D5FB0AAD329@citusdata.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Apr 18, 2014, at 1:51 PM, Atri Sharma <atri(dot)jiit(at)gmail(dot)com> wrote:
> Counting clock sweeps is an intersting idea. I think one concern was
> tracking hot buffers in cases where there is no memory pressure, and
> hence the clock sweep isn't running --- I am not sure how this would
> help in that case.
>
Yes, we obviously want a virtual clock. Focusing on the use of gettimeofday seems silly to me: it was something quick for the prototype.
The problem with the clocksweeps is they don’t actually track the progression of “time” within the PostgreSQL system.
What’s wrong with using a transaction number or some similar sequence? It would accurately track “age” in the sense we care about: how long ago in “units of real work being done by the DB” something was added.
—Jason
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Atri Sharma | 2014-04-18 20:14:42 | Re: Clock sweep not caching enough B-Tree leaf pages? |
| Previous Message | Atri Sharma | 2014-04-18 19:51:29 | Re: Clock sweep not caching enough B-Tree leaf pages? |