id you use for these tests: power-8 or x86? How long was
single run?
Per-run variation seems quite high. It also seems that it depends on which
version runs first. But that could be a coincidence.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
usly not a 9.6
> topic at this point, but it'd probably be good to that early in 9.7.
>
+1
Also, what do you think about validation script which checks consistency
between them?
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
On Mon, Mar 28, 2016 at 2:05 PM, Alexander Korotkov <
a.korot...@postgrespro.ru> wrote:
> After some debugging I found that vac_truncate_clog consumes xid just to
> produce warning. I wrote simple patch which replaces
> GetCurrentTransactionId() with ShmemVariabl
On Sun, May 22, 2016 at 12:39 PM, Amit Kapila <amit.kapil...@gmail.com>
wrote:
> On Mon, Mar 28, 2016 at 4:35 PM, Alexander Korotkov <
> a.korot...@postgrespro.ru> wrote:
>
>> Hackers,
>>
>> one our customer meet near xid wraparound situation. xid cou
scalability issues including high ProcArrayLock contention.
References.
1.
http://www.postgresql.org/message-id/CAPpHfdsytkTFMy3N-zfSo+kAuUx=u-7jg6q2byb6fpuw2cd...@mail.gmail.com
2.
http://www.postgresql.org/message-id/20151211130413.go14...@awork2.anarazel.de
3.
http://www.po
could package up a
> data directory after the issue hit?
>
FWIW, I'm also trying to reproduce it on big x86 machine on 9888b34f.
I'll write about results when get any.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
>
I've checked GIN code. Have no idea of how such wal record could be
generated...
The only idea I have is to add check that we're inserting valid WAL record
immediately before XLogInsert().
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
m
the syntax side it very looks like defining something boxes regions for two
keys which could be replacement for subpartitioning. But it isn't so.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
1116256 1167681
130 1066475 1120891
140 1040379 1085904
150 974064 1022160
160 938396 976487
170 953636 978120
180 920772 953843
We can see small but certain regression after snapshot too old feature.
--
Alexander Korotkov
Postgres Professional: http://www.postgre
there is small regression. I think this
regression could be caused by alignment of other data structures.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes
ould take a hit. But I think that would be
> acceptable.
Oh, I found that I underestimated complexity of async commit... :)
Do I understand right that now async commit right as follows?
1) Async transaction confirms commit before flushing WAL.
2) Other transactions sees effect of async transaction only when its WAL
flushed.
3) In the session which just committed async transaction, effect of this
transaction is visible immediately (before WAL flushed). Is it true?
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
an XID array.
FYI, I'm still fan of idea to rewrite XID with CSN in tuple in the same way
we're writing hint bits now.
I'm going to make prototype of this approach which would be enough for
performance measurements.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
ake async commits to work.
>
I didn't get async commits problem at first glance. AFAICS, the difference
between sync commit and async is only that async commit doesn't wait WAL
log to flush. But async commit still receives LSN.
Could you describe it in more details?
--
A
s just an idea without any
proof so soon.
2) We already have functionality which trades insight into database with
way more huge overhead. auto_explain.log_analyze = true can slowdown
queries *in times*. Do you think we should remove it?
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
at we do not have such accurate
> tests now?
>
Do you think that large shared buffers is a kind a stress test for wait
events monitoring? If so, why?
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
On Wed, Aug 10, 2016 at 8:26 PM, Greg Stark <st...@mit.edu> wrote:
> On Wed, Aug 10, 2016 at 5:54 PM, Alexander Korotkov
> <a.korot...@postgrespro.ru> wrote:
> > Oh, I found that I underestimated complexity of async commit... :)
>
> Indeed. I think Tom's attitude
indexes and so on). But imperative was that it should have its own
executor nodes,
and it doesn't have to compatible with existing index access methods.
Therefore, I think my design presented at PGCon and your current proposal
are
about orthogonal features which could coexist.
--
Alexander
a buffer
pin and vacuum would wait for pin release. But other heap operations could
still call heap_page_prune() which correspondingly can relocate tuple.
Probably, I'm missing something...
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
On Fri, Aug 12, 2016 at 3:15 PM, Alexander Korotkov <
a.korot...@postgrespro.ru> wrote:
> I'm now exploring code working with heap tuples. The following code
> in heap_update() catch my eyes.
>
> if (DoesMultiXactIdConflict((MultiXactId) xwait, infomask,
>> *lockmode))
&
inserts
> faster and to enable new advanced page layouts in indices.
>
+1 for PageReplaceItem()
Even if item is not the same size, we can move the tail of page once
instead of twice.
I think you should implement PageReplaceItem() version and add it to the
commitfest.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
r index scan too.
> Probably it
> > worth to discuss the way to normalize index expressions and clauses and
> work
> > out more convenient way to match them.
>
> pg_operator.oprcommutative ?
Do you mean pg_operator.oprcom?
In these examples we're also lacking of pg_operator.oprasso
On Fri, Feb 3, 2017 at 8:01 PM, Alexander Korotkov <
a.korot...@postgrespro.ru> wrote:
> Unfortunately, I have no big enough Power machine at hand to reproduce
> that results. Actually, I have no Power machine at hand at all. So,
> lwlock-power-2.patch was written "bli
eone would help me with testing and benchmarking.
1. https://www.postgresql.org/message-id/CAPpHfdsogj38HTDhNMLE56uJy9N8-
=gya2nnuwbpujgp2n1...@mail.gmail.com
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
lwlock-power-1.patch
Descripti
because
that machine is used for tries to make Postgres work properly on AIX.
So, benchmarking help is very relevant. I would very appreciate that.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
lwlock-power-3.patch
Description: Binary d
On Tue, Feb 7, 2017 at 3:16 PM, Bernd Helmle <maili...@oopsware.de> wrote:
> Am Montag, den 06.02.2017, 22:44 +0300 schrieb Alexander Korotkov:
>
>2. Also could you run each test longer: 3-5 mins, and run them with
>variety of clients count?
>
>
> So here a
r project ideas
> submitted so we can include them.
>
Thank you for reminder. I added couple more project ideas and added me,
Oleg and Teodor to possible mentors of some projects.
That seems to be all that I have for now.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
On Mon, Feb 6, 2017 at 8:28 PM, Bernd Helmle <maili...@oopsware.de> wrote:
> Am Montag, den 06.02.2017, 16:45 +0300 schrieb Alexander Korotkov:
> > I tried lwlock-power-2.patch on multicore Power machine we have in
> > PostgresPro.
> > I realized that using labels in as
ogic back unless
we have proof that this change make it better.
Patch is attached. I tried to make some comments, but probably they are
not enough.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
put-buffer-usagecount–logic–back.patch
nly pgbench results on 72 physically cores Intel server are attached.
It quite similar to results I posted before, but it's curious that
performance degradation of master on high concurrency became larger.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Pos
WAL writes in a new
> cluster are a more expensive than in an old one, because of segment
> recycling.
>
+1
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
s me think that probably there is something
wrong in testing methodology.
Are you doing re-initdb and pgbench -i before each run? I would ask you to
do the both.
Also, if regression would still exist, let's change the order of versions.
Thus, if you run master before patched version, I would ask
On Tue, Feb 21, 2017 at 1:47 PM, Bernd Helmle <maili...@oopsware.de> wrote:
> Am Samstag, den 11.02.2017, 15:42 +0300 schrieb Alexander Korotkov:
> > I think it would make sense to run more kinds of tests. Could you
> > try set
> > of tests provided by Tomas Vondra?
Also patch has set of assorted code and comments improvements.
Links
1.
https://www.postgresql.org/message-id/CA+TgmoZapyHRm7NVyuyZ+yAV=u1a070bogre7pkgyraegr4...@mail.gmail.com
2.
https://www.postgresql.org/message-id/cam3swzql4yd2sndhemcgl0q2b2otdkuvv_l6zg_fcgoluwm...@mail.gmail.com
--
Alexand
On Mon, Feb 13, 2017 at 7:07 PM, Alvaro Herrera <alvhe...@2ndquadrant.com>
wrote:
> Alexander Korotkov wrote:
>
> > Yes, influence seems to be low. But nevertheless it's important to
> insure
> > that there is no regression here.
> > Despite pg_prewarm'ing and r
On Tue, Feb 14, 2017 at 3:57 PM, Alvaro Herrera <alvhe...@2ndquadrant.com>
wrote:
> Alexander Korotkov wrote:
> > On Mon, Feb 13, 2017 at 7:07 PM, Alvaro Herrera <
> alvhe...@2ndquadrant.com>
> > wrote:
>
> > > Re the coding of the padding computat
On Mon, Feb 13, 2017 at 10:17 PM, Tomas Vondra <tomas.von...@2ndquadrant.com
> wrote:
> On 02/13/2017 03:16 PM, Bernd Helmle wrote:
>
>> Am Samstag, den 11.02.2017, 15:42 +0300 schrieb Alexander Korotkov:
>>
>>> Thus, I see reasons why in your tests absolute
/www.postgresql.org/message-id/20160822021747.u5bqx2xwwjzac...@alap3.anarazel.de
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
pgxact-align-3.patch
Description: Binary data
--
Sent via pgsql-hackers mailing list (pgsql-hackers@postgre
ee this variation on the graphs.
Another issue with graphs is that we can't see details of read and write
TPS variation on the same scale, because write TPS values are too low. I
think you should draw write benchmark on the separate graph.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
On Sat, Feb 11, 2017 at 4:17 PM, Tomas Vondra <tomas.von...@2ndquadrant.com>
wrote:
> On 02/11/2017 01:21 PM, Alexander Korotkov wrote:
>
>> Hi, Tomas!
>>
>> On Sat, Feb 11, 2017 at 2:28 AM, Tomas Vondra
>> <tomas.von...@2ndquadrant.com <mail
tarting a new
> thread to make sure we collect all the relevant votes, but I really,
> really think it's past time for this to go away. The last actual
> change to tsearch2 which wasn't part of a wider cleanup was
> 3ca7eddbb7c4803729d385a0c9535d8a972ee03f in January 2009, so it's been
> 7 years since t
if needed). This way wouldn't be
that brief, but it is much more correct.
Also this like catch my eyes.
> info->amcanorderbyop = (void (*)()) amroutine->amcanorderbyop;
It's not necessary to use cast here. For instance, we don't use cast
for amcostestimate.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
On Wed, Feb 15, 2017 at 11:49 AM, Alexander Korotkov <
a.korot...@postgrespro.ru> wrote:
> On Mon, Feb 13, 2017 at 8:02 PM, Ashutosh Sharma <ashu.coe...@gmail.com>
> wrote:
>
>> I too have performed benchmarking of this patch on a large machine
>> (with 12
FROM pg_class WHERE
relkind IN ('i', 'r') ORDER BY oid) x;
In my experience pg_prewarm both takes less time and leaves less variation
afterwards.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
On Wed, Feb 8, 2017 at 5:00 PM, Bernd Helmle <maili...@oopsware.de> wrote:
> Am Dienstag, den 07.02.2017, 16:48 +0300 schrieb Alexander Korotkov:
> > But win isn't
> > as high as I observed earlier. And I wonder why absolute numbers are
> > lower
> > than in our
On Fri, Feb 17, 2017 at 12:44 PM, Ashutosh Sharma <ashu.coe...@gmail.com>
wrote:
> >> On Thu, Feb 16, 2017 at 5:07 AM, Alexander Korotkov
> >> <a.korot...@postgrespro.ru> wrote:
> >> > On Wed, Feb 15, 2017 at 8:49 PM, Alvaro Herrera
>
On Thu, Feb 16, 2017 at 4:58 PM, Robert Haas <robertmh...@gmail.com> wrote:
> On Thu, Feb 16, 2017 at 5:07 AM, Alexander Korotkov
> <a.korot...@postgrespro.ru> wrote:
> > On Wed, Feb 15, 2017 at 8:49 PM, Alvaro Herrera <
> alvhe...@2ndquadrant.com>
> &g
performance regression
in other cases. At least we should test read-write and smaller machines.
Any other ideas?
1.
https://www.postgresql.org/message-id/20160411214029.ce3fw6zxim5k6...@alap3.anarazel.de
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres
On Fri, Aug 19, 2016 at 4:46 PM, Tom Lane <t...@sss.pgh.pa.us> wrote:
> Alexander Korotkov <a.korot...@postgrespro.ru> writes:
> > originally this idea was proposed by Andres Freund while experimenting
> with
> > lockfree Pin/UnpinBuffer [1].
> > The
On Fri, Aug 19, 2016 at 3:19 PM, Amit Kapila <amit.kapil...@gmail.com>
wrote:
> On Fri, Aug 19, 2016 at 11:42 AM, Alexander Korotkov
> <a.korot...@postgrespro.ru> wrote:
> > Hackers,
> >
> > originally this idea was proposed by Andres Freund while experimenting
On Mon, Aug 22, 2016 at 12:09 PM, Alexander Korotkov <
a.korot...@postgrespro.ru> wrote:
> I took a look at your patch. Couple of notes from me.
>
> const char *
>> GetEventIdentifier(uint16 eventId)
>> {
>> const char *res;
>> switch (eventId)
&
EventIdentifier seems too general name for me, isn't it? Could we name it
WaitEventIdentifier? Or WaitEventId for shortcut?
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
Hi all!
It seems that PostgreSQL has passed to GSoC mentoring organizations this
year!
https://summerofcode.withgoogle.com/organizations/4558465230962688/
Congratulations!
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
; Attached is a graph with the results. Full results are available at
>> https://hlinnaka.iki.fi/temp/csn-4-results/. In short, the patch
>> improved throughput, measured in TPS, with >= 32 or so clients. The
>> biggest difference was with 44 clients, which saw about 5% improvement.
>&g
On Fri, Apr 8, 2016 at 10:09 PM, Peter Geoghegan <p...@heroku.com> wrote:
> On Wed, Mar 30, 2016 at 8:02 AM, Alexander Korotkov
> <aekorot...@gmail.com> wrote:
> > Hmm... I'm not completely agree with that. In typical usage partial sort
> > should definitely use
flexibility yet without adding a branch
> point in the code?
>
It's impossible without branch point in code. The question is where this
branch should be located.
In particular, be can put this branch point into planner by defining
distinct executor nodes for each pluggable storage. In this
vacuum of
single index.
But pointing each worker at separate index seems to be fair enough for
majority of cases.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
On Wed, Oct 19, 2016 at 12:53 PM, Simon Riggs <si...@2ndquadrant.com> wrote:
> On 18 October 2016 at 23:46, Alexander Korotkov
> <a.korot...@postgrespro.ru> wrote:
>
> > Then vacuum removes (0;1) from heap, reference to (0;1) from tbl_pk_idx.
> > But ho
But it seems like rather better design for me.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
ted this yet.
>
Imagine another situation: PK column was not updated, but indirect indexed
column was updated.
Thus, for single heap tuple we would have single PK tuple and two indirect
index tuples (correct me if I'm wrong).
How are we going to delete old indirect index tuple?
--
On Wed, Oct 19, 2016 at 12:21 AM, Alexander Korotkov <
a.korot...@postgrespro.ru> wrote:
> On Tue, Oct 18, 2016 at 9:28 PM, Alvaro Herrera <alvhe...@2ndquadrant.com>
> wrote:
>
>> Vacuuming presents an additional challenge: in order to remove index
>> items from
are just different kind of maps. And they could implement the same
interface. Imagine you can select between heap-organized table
and index-organized table just by choosing its primary access method. If
you select heap for primary access method, indexes would
refer TID. If you select btree o
kers will get free admission to the conference (all
days), tickets to Moscow and accommodation booked and paid by the
organizers.
We will be glad to see you in Moscow!
------
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
ly update) tuple routing for foreign partitions
- hash partitioning
- not scanning the parent
Time is growing short, but it would
> be great to polish this a little more before we ship it.
>
Yes. Getting at least some of this features committed to v10 would be great
and improve partitioning usabi
. But I don't like this
approach.
If we have large enough index scan, then we could either run out of cache
or consume too much memory
for that cache.
So, I think storing visibility information in index tuple is great for
regular index scans too. At least,
it allow us to evade #2 problem. That is great by itself.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
-0mW6M03St7A%40mail.gmail.com
[2]
https://www.postgresql.org/message-id/flat/CALxAEPuGpAjBSN-PTuxHfuLLqDS47BEbO_ZYxUYQR3ud1nwbww%40mail.gmail.com
[3] https://developers.google.com/open-source/gsoc/timeline
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
nly connections
2017-03-27 17:32:16.920 MSK [30788] LOG: started streaming WAL from
primary at 0/900 on timeline 1
done
server started
# select pg_is_in_recovery();
pg_is_in_recovery
───
t
(1 row)
Thus, I think patch is working as expected in this case.
Also,
On Thu, Aug 4, 2016 at 2:28 PM, Ildar Musin <i.mu...@postgrespro.ru> wrote:
> I was looking through the buffer manager's code and have noticed that
> function PinBuffer has an unused 'strategy' argument. It's seems that after
> refactoring made by Alexander Korotkov an
On Sat, Mar 18, 2017 at 12:41 AM, Pritam Baral <pri...@pritambaral.com>
wrote:
> On Friday 10 March 2017 07:59 PM, Alexander Korotkov wrote:
>
>> Hi, Pritam! > > I've assigned to review this patch. > > On Thu, Feb 23,
>> 2017 at 2:17 AM, Pritam
On Fri, Mar 10, 2017 at 6:06 PM, Pavel Stehule <pavel.steh...@gmail.com>
wrote:
> 2017-03-10 16:00 GMT+01:00 Alexander Korotkov <a.korot...@postgrespro.ru>:
>
>> On Fri, Mar 10, 2017 at 5:16 PM, Stephen Frost <sfr...@snowman.net>
>> wrote:
>
On Tue, Mar 28, 2017 at 10:12 AM, Pavel Stehule <pavel.steh...@gmail.com>
wrote:
> 2017-03-27 13:59 GMT+02:00 Alexander Korotkov <a.korot...@postgrespro.ru>:
>
>> On Fri, Mar 10, 2017 at 6:06 PM, Pavel Stehule <pavel.steh...@gmail.com>
>> wrote:
>>
On Tue, Mar 28, 2017 at 5:27 PM, David Steele <da...@pgmasters.net> wrote:
> Hi Alexander,
>
> On 3/20/17 10:19 AM, Heikki Linnakangas wrote:
>
>> On 03/20/2017 11:33 AM, Alexander Korotkov wrote:
>>
>>> Please, find rebased patch in the a
On Thu, Mar 16, 2017 at 8:35 PM, David Steele <da...@pgmasters.net> wrote:
> On 2/21/17 9:54 AM, Bernd Helmle wrote:
> > Am Dienstag, den 14.02.2017, 15:53 +0300 schrieb Alexander Korotkov:
> >> +1
> >> And you could try to use pg_wait_sampling
>
On Sat, Mar 25, 2017 at 8:44 PM, Tom Lane <t...@sss.pgh.pa.us> wrote:
> Alexander Korotkov <a.korot...@postgrespro.ru> writes:
> > [ lwlock-power-3.patch ]
>
> I experimented with this patch a bit. I can't offer help on testing it
> on large PPC machines, b
On Sat, Mar 25, 2017 at 11:32 PM, Tom Lane <t...@sss.pgh.pa.us> wrote:
> Alexander Korotkov <a.korot...@postgrespro.ru> writes:
> > I moved PPC implementation of pg_atomic_fetch_mask_add_u32() into
> > port/atomics/arch-ppc.h. I also had to declare pg_atomic_uint32
re end of online backup"),
> errhint("Online backup started with pg_start_backup() must be ended with
> pg_stop_backup(), and all WAL up to that point must be available at
> recovery.")));
> else
> ereport(FATAL,
> (errmsg("WAL ends before consistent recovery point")));
Should we consider fixing them?
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
On Sun, Mar 26, 2017 at 12:29 AM, Alexander Korotkov <
a.korot...@postgrespro.ru> wrote:
> On Sat, Mar 25, 2017 at 11:32 PM, Tom Lane <t...@sss.pgh.pa.us> wrote:
>
>> Alexander Korotkov <a.korot...@postgrespro.ru> writes:
>> > I moved PPC impleme
approaches
you're going to try, and specify them in proposal. Schedule should include
prototyping each of these approaches, then testing and benchmarking to
selecting between them, then work on bringing selected approach to
committable shape.
--
Alexander Korotkov
Postgres Professional: ht
>
> if (BUF_STATE_GET_USAGECOUNT(buf_state) != BM_MAX_USAGE_COUNT)
>
> being prone to paranoia, I prefer the first, but I've seen both styles in
> the code so I don't know if it's worth futzing with.
>
Ok, let's be paranoic and do this same way as before. Revised patch is
attac
48354581a49c30f5757c203415aa8412d85b0f70 at April, 10. Does it mean the
> fix should be backpatched too?
I think so. This patch reverts unintentional change and can be considered
as bug fix.
BTW, sorry for unicode filename in previous letter.
Patch with normal ASCII name is attached.
--
Alexander
On Tue, Apr 11, 2017 at 5:46 PM, Jeff Davis <pg...@j-davis.com> wrote:
> On Tue, Apr 11, 2017 at 2:17 AM, Alexander Korotkov
> <a.korot...@postgrespro.ru> wrote:
> > FYI, I've implemented this algorithm for pgsphere. See following branch.
> > https://gith
On Mon, Apr 3, 2017 at 10:05 PM, Andres Freund <and...@anarazel.de> wrote:
> On April 3, 2017 12:03:56 PM PDT, Alexander Korotkov <
> a.korot...@postgrespro.ru> wrote:
> >On Mon, Apr 3, 2017 at 9:34 PM, Andres Freund <and...@anarazel.de>
> >wrote:
> >
&g
On Tue, Apr 4, 2017 at 12:09 AM, Andres Freund <and...@anarazel.de> wrote:
> On 2017-04-04 00:04:09 +0300, Alexander Korotkov wrote:
> > > >Thank you!
> > > >I already sent version of patch after David's reminder.
> > > >Please find rebased patch
On Mon, Apr 3, 2017 at 9:58 PM, Andres Freund <and...@anarazel.de> wrote:
> On 2017-03-25 19:35:35 +0300, Alexander Korotkov wrote:
> > On Wed, Mar 22, 2017 at 12:23 AM, David Steele <da...@pgmasters.net>
> wrote:
> >
> > > Hi Alexander
> > >
&g
y own, but it uses replace_nestloop_params() (again, static),
> and the problem goes further and further.
>
As I understand, making build_path_tlist a non-static function would solve
the problem.
Tom, do you think it's possible?
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
On Wed, Apr 12, 2017 at 12:40 AM, Alexander Korotkov <
a.korot...@postgrespro.ru> wrote:
> On Tue, Apr 11, 2017 at 8:24 PM, Alexander Kuzmenkov <
> a.kuzmen...@postgrespro.ru> wrote:
>
>> I would like to propose a patch that speeds up the queries of the form
>&
scan
returns all rechecks, alternative to Bitmap Count is still Aggregate +
Bitmap Heap Scan. Thus, I think Bitmap Count would have the same
performance or even slightly faster. That's worth testing.
Also, what is planning overhead of this patch? That's worth testing too.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
ve that “<@( is
> contained by)” can be used in this scope.
What exactly you're going to do in this item? Check that GIN index over
array support given operator, do the performance benchmark or something
else?
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
do not hit a broken line.
I think that situation when backend is directly reading file during COPY is
not typical. More typical case is \copy psql command. In that case "COPY
... FROM stdin;" is actually executed while psql is streaming the data.
How can we apply parallel COPY in th
On Thu, Apr 6, 2017 at 5:37 PM, Alexander Korotkov <
a.korot...@postgrespro.ru> wrote:
> On Thu, Apr 6, 2017 at 2:16 AM, Andres Freund <and...@anarazel.de> wrote:
>
>> On 2017-04-03 11:56:13 -0700, Andres Freund wrote:
>> > Have you done x86 benchmarking?
>&
+ }
> > > + pg_unreachable();
> > > +}
> > > +#endif
> >
> > Have you done x86 benchmarking?
>
> I think unless such benchmarking is done in the next 24h we need to move
> this patch to the next CF...
>
Thank you for your comments.
I didn't x86 benchmarking. I even didn't manage to reproduce previous
results on Power8.
Presumably, it's because previous benchmarks were done on bare metal, while
now I have to some kind of virtual machine on IBM E880 where I can't
reproduce any win of Power8 LWLock optimization. But probably there is
another reason.
Thus, I'm moving this patch to the next CF.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
;
I'd love to see a comment from Andres Freund who is leading executor
performance improvements.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
//github.com/akorotkov/pgsphere/tree/crossmatch_cnode
You also can find some experimental evaluation here:
http://www.adass2016.inaf.it/images/presentations/10_Korotkov.pdf
Feel free to experiment with this code and/or ask any question.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
to the
> alignment patch. Do we have evidence to commit both?
>
> Well, We have seen some regression in read-write test with pgxact
> alignment patch - [1]. I may need to run the test with both the
> patches to see the combined effect on performance.
>
> [1]
CT a FROM GENERATE_SERIES(1,
1000) a;
You can achieve the same result by execution of single SQL statement.
INSERT INTO con_hash_index_table (keycol) SELECT (a - 1) % 1000 + 1 FROM
GENERATE_SERIES(1, 1002000) a;
Unless you have some special idea of doing this in 1002 separate
transactions.
-
tes:
> >>> On Thu, Feb 16, 2017 at 8:05 AM, Alexander Korotkov
> >>> <a.korot...@postgrespro.ru> wrote:
> >>>> My idea is that we need more general redesign of specifying ordering
> which
> >>>> index can produce. Ideally, we should repla
letely separate feature. And it should be posted and discussed
separately.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
mark it as "ready for committer"?
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
between different sortings
quickly. For instance, after seeing tables sorted by name, user can
require them sorted by size descending, then sorted by size ascending,
etc...
Therefore, I find user-defined SQL clause to be cumbersome. Even psql
variable itself seems to be cumbersome for me.
I would propose t
Node *leftop, Oid expr_op, Oid
opfamily,
^
Could you please recheck that you published right version of patch?
Also, I noticed that patch haven't regression tests. Some mention of this
optimization in docs is also nice to have.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
196 27628 28511 3.19603301
> 256 28467 28529 0.2177960445
>
Results look good for me. Idea of committing both of patches looks
attractive.
We have pretty much acceleration for read-only case and small acceleration
for read-write case.
I'll run benchmark on 72-cores machine as well.
--
Alexander Korotkov
Postgres Professional: http://www.postgrespro.com
The Russian Postgres Company
801 - 900 of 1043 matches
Mail list logo