On Tue, May 22, 2012 at 11:36 PM, Ants Aasma a...@cybertec.at wrote:
... The free list
itself is a bit trickier, but if it's still necessary/useful then
SC-firstFreeBuffer and buf-freeNext are in effect a linked-list
stack, there should plenty of tested lock free algorithms floating
around
, May 22, 2012 7:55 PM
To: Robert Haas
Cc: Amit Kapila; PostgreSQL-development
Subject: Re: [HACKERS] Readme of Buffer Management seems to have wrong
sentence
Robert Haas robertmh...@gmail.com writes:
On Tue, May 22, 2012 at 10:01 AM, Tom Lane t...@sss.pgh.pa.us wrote:
Well, keep in mind
On 05/23/2012 11:36 AM, Amit Kapila wrote:
Do you feel I can attempt to address this problem with some prototypes and
discuss here after few days when I have some results ready.
I don't think there is a clear picture yet of what benchmark to use for
testing changes here. Items like Consider
: Greg Smith [mailto:g...@2ndquadrant.com]
Sent: Wednesday, May 23, 2012 10:35 PM
To: pgsql-hackers@postgresql.org; amit.kap...@huawei.com
Subject: Re: [HACKERS] Readme of Buffer Management seems to have wrong
sentence
On 05/23/2012 11:36 AM, Amit Kapila wrote:
Do you feel I can attempt to address
On Wed, May 23, 2012 at 10:33 AM, Amit Kapila amit.kap...@huawei.com wrote:
I don't think there is a clear picture yet of what benchmark to use for
testing changes here.
I will first try to generate such a scenario(benchmark). I have still not
thought completely.
However the idea in my mind is
On Wed, May 23, 2012 at 8:36 AM, Amit Kapila amit.kap...@huawei.com wrote:
And besides
if the decrements are decoupled from the allocation requests it's no
longer obvious that the algorithm is even an approximation of LRU.
I was trying to highlight that we can do the clocksweep in bgwriter and
, 2012 11:39 PM
To: Amit Kapila
Cc: Greg Smith; pgsql-hackers@postgresql.org
Subject: Re: [HACKERS] Readme of Buffer Management seems to have wrong
sentence
On Wed, May 23, 2012 at 10:33 AM, Amit Kapila amit.kap...@huawei.com
wrote:
I don't think there is a clear picture yet of what benchmark to use
Jeff Janes jeff.ja...@gmail.com writes:
One thing I wanted to play with is having newly read buffers get a
usage count of 0 rather than 1. The problem is that there is no way
to test it in enough different situations to convince people it would
be a general improvement.
Hmm ... ISTM that
to generate the scenarios?
-Original Message-
From: Jeff Janes [mailto:jeff.ja...@gmail.com]
Sent: Wednesday, May 23, 2012 11:48 PM
To: Amit Kapila
Cc: Tom Lane; Robert Haas; PostgreSQL-development
Subject: Re: [HACKERS] Readme of Buffer Management seems to have wrong
sentence
On Wed, May 23
On Wed, May 23, 2012 at 11:40 AM, Tom Lane t...@sss.pgh.pa.us wrote:
Jeff Janes jeff.ja...@gmail.com writes:
One thing I wanted to play with is having newly read buffers get a
usage count of 0 rather than 1. The problem is that there is no way
to test it in enough different situations to
On Wed, May 23, 2012 at 2:09 PM, Jeff Janes jeff.ja...@gmail.com wrote:
On Wed, May 23, 2012 at 10:33 AM, Amit Kapila amit.kap...@huawei.com wrote:
I don't think there is a clear picture yet of what benchmark to use for
testing changes here.
I will first try to generate such a
Jeff Janes jeff.ja...@gmail.com writes:
On Wed, May 23, 2012 at 11:40 AM, Tom Lane t...@sss.pgh.pa.us wrote:
Hmm ... ISTM that that was discussed back when we instituted buffer
usage counts, and rejected on the grounds that a newly-read buffer could
then have negligible life expectancy. The
On Tue, May 8, 2012 at 9:37 PM, Amit Kapila amit.kap...@huawei.com wrote:
I have checked the code and logic according to which usage counter is
increased when the buffer is pinned.
Fixed, thanks for the report.
Another Doubt : Why in function BufferAlloc, it needs to hold the
BufFreelistLock
Robert Haas robertmh...@gmail.com writes:
Mind you, I think this whole area of the code needs some reengineering
for better performance, but I'm not sure this is the right place to
start. What I think is really bad is that we're forcing every
BufferAlloc() to iterate over buffers checking
On Tue, May 22, 2012 at 10:01 AM, Tom Lane t...@sss.pgh.pa.us wrote:
Robert Haas robertmh...@gmail.com writes:
Mind you, I think this whole area of the code needs some reengineering
for better performance, but I'm not sure this is the right place to
start. What I think is really bad is that
Robert Haas robertmh...@gmail.com writes:
On Tue, May 22, 2012 at 10:01 AM, Tom Lane t...@sss.pgh.pa.us wrote:
Well, keep in mind that that action is not merely there to obtain a
victim buffer; it is also maintaining the global LRU state (by
decrementing the usage counts of buffers it passes
On Tue, May 22, 2012 at 10:25 AM, Tom Lane t...@sss.pgh.pa.us wrote:
The idea would be to have a background process (like bgwriter)
maintain the global LRU state and push candidate buffers onto the
freelist.
Amit was trying to convince me of the same idea at PGCon, but I don't
buy it.
Robert Haas robertmh...@gmail.com writes:
On Tue, May 22, 2012 at 10:25 AM, Tom Lane t...@sss.pgh.pa.us wrote:
My own thoughts about this had pointed in the direction of getting rid
of the central freelist entirely, instead letting each backend run its
own independent clock sweep as needed.
On Tue, May 22, 2012 at 12:11 PM, Tom Lane t...@sss.pgh.pa.us wrote:
With respect to the control logic for the background writer, one idea
I had was to get rid of the idea that the background writer's job is
to write in advance of the strategy point. Instead, every time the
clock sweep passes
Robert Haas robertmh...@gmail.com writes:
If we're going to throw our current algorithm over wholesale, I'd
rather use some approach that has been demonstrated to work well in
other systems. Buffer eviction is a problem that's been around since
the 1970s, and our algorithm is just about that
On Tue, May 22, 2012 at 12:39 PM, Tom Lane t...@sss.pgh.pa.us wrote:
Robert Haas robertmh...@gmail.com writes:
If we're going to throw our current algorithm over wholesale, I'd
rather use some approach that has been demonstrated to work well in
other systems. Buffer eviction is a problem
On Tue, May 22, 2012 at 8:28 PM, Robert Haas robertmh...@gmail.com wrote:
Totally agreed. We're not the first people to think of this, either:
CLOCK and GLOCK have been extensively studied and found to be almost
as good as LRU in selecting good victim pages, but with less
contention. That's
22 matches
Mail list logo