[EMAIL PROTECTED] wrote:
> > I tried to implement LRU-2 awhile ago, and got discouraged when I
> > couldn't see any performance improvement. But I was using pgbench as
> > the test case, and failed to think about its lack of seqscans.
>
> Yes , lru-2 will behave like LRU under 'normal' load. it w
On Tue, 24 Jun 2003 10:27:09 -0400
Tom Lane <[EMAIL PROTECTED]> wrote:
> I tried to implement LRU-2 awhile ago, and got discouraged when I
> couldn't see any performance improvement. But I was using pgbench as
> the test case, and failed to think about its lack of seqscans.
How about cache hit
On Tue, 24 Jun 2003, Tom Lane wrote:
> Yutaka tanida <[EMAIL PROTECTED]> writes:
> > [EMAIL PROTECTED] wrote:
> >> does pgbench test with relatively large sequential scans?
>
> > Probably no.
>
> pgbench tries to avoid any seqscans at all, I believe, so it wouldn't be
> very useful for testing
On Tue, 24 Jun 2003 [EMAIL PROTECTED] wrote:
> I was researching on cache replacement strategy as well. 2Q has one
> disadvantage see this exellent paper:
> http://www.almaden.ibm.com/cs/people/dmodha/#ARC see the paper
> "ARC: A Self-Tuning, Low Overhead Replacement Cache" for theory and "One
> U
On Wed, 25 Jun 2003 00:43:56 +0900
Yutaka tanida <[EMAIL PROTECTED]> wrote:
> > BTW, when you were running your test case, what shared_buffers did you
> > use?
>
> I use 16,64,256 and 4096.
I missed. My shown result(+4% cache hit rate) is shared_buffers=64.
--
On Tue, 24 Jun 2003 10:27:09 -0400
Tom Lane <[EMAIL PROTECTED]> wrote:
> I tried to implement LRU-2 awhile ago, and got discouraged when I
> couldn't see any performance improvement. But I was using pgbench as
> the test case, and failed to think about its lack of seqscans.
How about cache hit
> Yutaka tanida <[EMAIL PROTECTED]> writes:
> > [EMAIL PROTECTED] wrote:
> >> does pgbench test with relatively large sequential scans?
>
> > Probably no.
>
> pgbench tries to avoid any seqscans at all, I believe, so it wouldn't be
> very useful for testing a method that's mainly intended to pre
Yutaka tanida <[EMAIL PROTECTED]> writes:
> [EMAIL PROTECTED] wrote:
>> does pgbench test with relatively large sequential scans?
> Probably no.
pgbench tries to avoid any seqscans at all, I believe, so it wouldn't be
very useful for testing a method that's mainly intended to prevent
seqscans fr
xoror,
On Tue, 24 Jun 2003 12:13:51 +0200 (MEST)
[EMAIL PROTECTED] wrote:
> I was researching on cache replacement strategy as well. 2Q has one
> disadvantage see this exellent paper:
> http://www.almaden.ibm.com/cs/people/dmodha/#ARC see the paper
> "ARC: A Self-Tuning, Low Overhead Replacement