From: tushar
Sent: Monday, September 27, 2021 11:50
To: Andrew Dunstan ; l...@laurent-hasson.com; Julien
Rouhaud
Cc: Tom Lane ; Ranier Vilela ; Justin
Pryzby ; pgsql-performa...@postgresql.org
Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2 and 13.4
On 9/27/21 6:55 PM
From: tushar
Sent: Monday, September 27, 2021 11:50
To: Andrew Dunstan ; l...@laurent-hasson.com; Julien
Rouhaud
Cc: Tom Lane ; Ranier Vilela ; Justin
Pryzby ; pgsql-performa...@postgresql.org
Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2 and 13.4
On 9/27/21 6:55 PM
On 9/27/21 6:55 PM, Andrew Dunstan wrote:
Hello Andrew,
I just download the 13.4 Windows x86-64 installer
fromhttps://www.enterprisedb.com/downloads/postgres-postgresql-downloads but
it's the exact same file bit for bit from the previous version I had. Am I
looking at the wrong place?
Than
On 9/25/21 9:33 PM, l...@laurent-hasson.com wrote:
>> > EDB has now published new installers for versions later than release
>> > 11, containing Postgres built with an earlier version of gettext that
>> > does not exhibit the problem. Please verify that these fix the issue.
>>
> > EDB has now published new installers for versions later than release
> > 11, containing Postgres built with an earlier version of gettext that
> > does not exhibit the problem. Please verify that these fix the issue.
> > If you already have Postgres installed from our installer
ember 13, 2021 11:36
> >>> To: l...@laurent-hasson.com; Julien Rouhaud
> >>> Cc: Tom Lane ; Ranier Vilela
> ;
> >> > Justin Pryzby ; pgsql-
> >>> performa...@postgresql.org
> >>> Subject: Re: Big Performance drop o
Rouhaud
>>> Cc: Tom Lane ; Ranier Vilela ;
>>> Justin Pryzby ; pgsql-
>>> performa...@postgresql.org
>>> Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2
>>> and 13.4
>>>
>>
a ;
>> Justin Pryzby ; pgsql-
>> performa...@postgresql.org
>> Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2
>> and 13.4
>>
>>
>> On 9/13/21 10:32 AM, l...@laurent-hasson.com wrote:
>> >
>
Re: Big Performance drop of Exceptions in UDFs between V11.2
> and 13.4
>
>
> On 9/13/21 10:32 AM, l...@laurent-hasson.com wrote:
> >
> > Hello all,
> >
> > Any further update or guidance on this issue at this time?
> >
On 9/13/21 10:32 AM, l...@laurent-hasson.com wrote:
>
> Hello all,
>
> Any further update or guidance on this issue at this time?
>
Wait for a new installer. Our team is working on it. As I have
previously advised you, please be patient.
cheers
andrew
--
Andrew Dunstan
EDB: https://www.ente
Re: Big Performance drop of Exceptions in UDFs between V11.2
> and 13.4
>
>
> On 9/2/21 11:34 AM, Julien Rouhaud wrote:
> > On Thu, Sep 2, 2021 at 11:22 PM Andrew Dunstan
> wrote:
> >> Here are a couple of pictures of profiles made with
On 9/2/21 11:34 AM, Julien Rouhaud wrote:
> On Thu, Sep 2, 2021 at 11:22 PM Andrew Dunstan wrote:
>> Here are a couple of pictures of profiles made with a tool called
>> sleepy. The bad profile is from release 13.4 built with the latest
>> gettext, built with vcpkg. The good profile is the same
Em qui., 2 de set. de 2021 às 12:22, Andrew Dunstan
escreveu:
>
> On 8/31/21 1:55 PM, Andrew Dunstan wrote:
> > On 8/31/21 11:37 AM, Julien Rouhaud wrote:
> >> On Tue, Aug 31, 2021 at 10:51 PM l...@laurent-hasson.com
> >> wrote:
> >>> OK... I thought that track had been abandoned as per Julien's
On Thu, Sep 2, 2021 at 11:22 PM Andrew Dunstan wrote:
>
> Here are a couple of pictures of profiles made with a tool called
> sleepy. The bad profile is from release 13.4 built with the latest
> gettext, built with vcpkg. The good profile is the same build but using
> the intl-8.dll copied from th
On Wed, Sep 1, 2021 at 1:56 AM Andrew Dunstan wrote:
>
> The culprit turns out to be the precise version of libiconv/libintl
> used. There is a slight difference between the versions used in the
> 11.13 installer and the 13.4 installer. We need to dig into performance
> more (e.g. why does the tes
On 8/31/21 11:37 AM, Julien Rouhaud wrote:
> On Tue, Aug 31, 2021 at 10:51 PM l...@laurent-hasson.com
> wrote:
>> OK... I thought that track had been abandoned as per Julien's last message.
>> Anyways, I'll be patient!
>>
> I just happened to have both standard installer and locally compiled
>
On Tue, Aug 31, 2021 at 10:51 PM l...@laurent-hasson.com
wrote:
>
> OK... I thought that track had been abandoned as per Julien's last message.
> Anyways, I'll be patient!
>
I just happened to have both standard installer and locally compiled
versions available, so I could confirm that I reprodu
Big Performance drop of Exceptions in UDFs between V11.2
> and 13.4
>
>
> On 8/30/21 10:18 PM, l...@laurent-hasson.com wrote:
> >> I see on https://www.postgresql.org/download/ that there is a
> different
> >> installer from 2
On 8/30/21 10:18 PM, l...@laurent-hasson.com wrote:
>> I see on https://www.postgresql.org/download/ that there is a different
>> installer from 2ndQuadrant. I am going to try that one and see what I
>> come up with. Are there any other "standard" distros of Postgres that I
>>
-Message d'origine-
De : l...@laurent-hasson.com
Envoyé : mardi 31 août 2021 04:18
À : l...@laurent-hasson.com; Julien Rouhaud
Cc : Tom Lane ; Ranier Vilela ; Andrew
Dunstan ; Justin Pryzby ;
pgsql-performa...@postgresql.org
Objet : RE: Big Performance drop of Exceptions in
> I see on https://www.postgresql.org/download/ that there is a different
> installer from 2ndQuadrant. I am going to try that one and see what I
> come up with. Are there any other "standard" distros of Postgres that I
> could try out?
>
> I found out I could download Vis
> I tried this scenario using edb's 13.3 x64 install:
>
> postgres=# select version();
>version
>
> PostgreSQL 13.3, compiled by Visual C++ build 1914, 64-bit
> (1 row)
>
>
Big Performance drop of Exceptions in UDFs between V11.2
> and 13.4 (workarounds)
>
> On Mon, Aug 30, 2021 at 04:43:23AM +0200, Pavel Stehule wrote:
> > po 30. 8. 2021 v 2:44 odesílatel l...@laurent-hasson.com napsal:
> > > At this point, I a
On Mon, Aug 30, 2021 at 04:43:23AM +0200, Pavel Stehule wrote:
> po 30. 8. 2021 v 2:44 odesílatel l...@laurent-hasson.com napsal:
> > At this point, I am not sure how to proceed except to rethink that
> > toFloat() function and many other places where we use exceptions. We get
> > such dirty data t
Hi
po 30. 8. 2021 v 2:44 odesílatel l...@laurent-hasson.com <
l...@laurent-hasson.com> napsal:
>
>
>
> At this point, I am not sure how to proceed except to rethink that
> toFloat() function and many other places where we use exceptions. We get
> such dirty data that I need a "safe" way to conver
On Mon, Aug 30, 2021 at 8:44 AM l...@laurent-hasson.com
wrote:
>
> Yeah, grasping at straws... and no material changes 😊 This is mystifying.
>
> show lc_messages;
> -- English_United States.1252
>
> create table sampletest (a varchar, b varchar);
> insert into sampletest (a, b)
> select substr(md5
esql.org
> Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2 and
> 13.4
>
> Em dom., 29 de ago. de 2021 às 13:03, mailto:l...@laurent-hasson.com
> <mailto:l...@laurent-hasson.com> escreveu:
> >Sure, there's no question that message trans
> -Original Message-
> From: Tom Lane
> Sent: Sunday, August 29, 2021 12:19
> To: l...@laurent-hasson.com
> Cc: Ranier Vilela ; Andrew Dunstan
> ; Justin Pryzby ; pgsql-
> performa...@postgresql.org
> Subject: Re: Big Performan
From: Ranier Vilela
Sent: Sunday, August 29, 2021 14:20
To: l...@laurent-hasson.com
Cc: Tom Lane ; Andrew Dunstan ; Justin
Pryzby ; pgsql-performa...@postgresql.org
Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2 and 13.4
Em dom., 29 de ago. de 2021 às 13:03, mailto:l
Em dom., 29 de ago. de 2021 às 13:03, l...@laurent-hasson.com <
l...@laurent-hasson.com> escreveu:
> >Sure, there's no question that message translation will have *some* cost.
> >But on my machine it is an incremental tens-of-percent kind of cost,
> >and that is the result you're getting as well.
"l...@laurent-hasson.com" writes:
> Is it possible that the client I am using or the way I am creating the test
> database might affect this scenario? I use DBeaver and use the default
> settings to create the database:
> - default encoding: UTF8
> - collate: English_United States.1252
> - ctype
>Sure, there's no question that message translation will have *some* cost.
>But on my machine it is an incremental tens-of-percent kind of cost,
>and that is the result you're getting as well. So it's not very clear
>where these factor-of-several-hundred differences are coming from.
>A hypothesis
Em dom., 29 de ago. de 2021 às 10:35, Tom Lane escreveu:
> Ranier Vilela writes:
> > I retested this case with HEAD, and it seems to me that NLS does affect
> it.
>
> Sure, there's no question that message translation will have *some* cost.
> But on my machine it is an incremental tens-of-percen
Ranier Vilela writes:
> I retested this case with HEAD, and it seems to me that NLS does affect it.
Sure, there's no question that message translation will have *some* cost.
But on my machine it is an incremental tens-of-percent kind of cost,
and that is the result you're getting as well. So it'
> Cc: Andrew Dunstan ; Justin Pryzby
>> ; Ranier Vilela ; pgsql-
>> performa...@postgresql.org
> > Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2
>> and 13.4
>>
>> "l...@laurent-hasson.com" writes:
>
> -Original Message-
> From: Tom Lane
> Sent: Saturday, August 28, 2021 15:51
> To: l...@laurent-hasson.com
> Cc: Andrew Dunstan ; Justin Pryzby
> ; Ranier Vilela ; pgsql-
> performa...@postgresql.org
> Subject: Re: Big Performan
"l...@laurent-hasson.com" writes:
> SET lc_messages = 'C';
> show lc_messages; --> OK 'C'
> explain (analyze,buffers,COSTS,TIMING)
> select MAX(toFloat(b, null)) as "b" from sampletest
> ...
> Execution Time: 175.600 ms
> explain (analyze,buffers,COSTS,TIMING)
> select MAX(toFloat(a, null)) as
> -Original Message-
> From: Tom Lane
> Sent: Friday, August 27, 2021 13:43
> To: l...@laurent-hasson.com
> Cc: Justin Pryzby ; Ranier Vilela
> ; pgsql-performa...@postgresql.org
> Subject: Re: Big Performance drop of Exceptions
> -Original Message-
> From: Tom Lane
> Sent: Friday, August 27, 2021 13:43
> To: l...@laurent-hasson.com
> Cc: Justin Pryzby ; Ranier Vilela
> ; pgsql-performa...@postgresql.org
> Subject: Re: Big Performance drop of Exceptions
"l...@laurent-hasson.com" writes:
> That being said, do you have any suggestion how I could circumvent the
> issue altogether?
Based on Andrew's report, it seems like you might be able to work around
it for the time being by disabling message translations, i.e.
SET lc_messages = 'C';
Even
> -Original Message-
> From: Andrew Dunstan
> Sent: Thursday, August 26, 2021 12:39
> To: l...@laurent-hasson.com; Justin Pryzby ;
> Ranier Vilela
> Cc: Tom Lane ; pgsql-performa...@postgresql.org
> Subject: Re: Big Performance drop
On 8/26/21 10:47 AM, l...@laurent-hasson.com wrote:
> Hello all,
>
> Any update on this issue?
Please don't top-post.
We are working on the issue. Please be patient.
cheers
andrew
--
Andrew Dunstan
EDB: https://www.enterprisedb.com
> > Cc: l...@laurent-hasson.com; Tom Lane ; pgsql-
> > performa...@postgresql.org
> > Subject: Re: Big Performance drop of Exceptions in UDFs between
> V11.2
> > and 13.4
> >
> > On Sun, Aug 22, 2021 at 08:44
> -Original Message-
> From: Justin Pryzby
> Sent: Sunday, August 22, 2021 20:43
> To: Ranier Vilela
> Cc: l...@laurent-hasson.com; Tom Lane ; pgsql-
> performa...@postgresql.org
> Subject: Re: Big Performance drop of Exceptions
On Sun, Aug 22, 2021 at 08:44:34PM -0300, Ranier Vilela wrote:
> > If there is any way I can help further... I am definitely not able to do a
> > dev environment and local build, but if we have a windows developer
> > reproducing the issue between 11 and 12, then that should help. If someone
> > ma
> Cc: Justin Pryzby ; Ranier Vilela
>> ; pgsql-performa...@postgresql.org
> > Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2
>> and 13.4
>>
>> "l...@laurent-hasson.com" writes:
>> > I do have a Linux in
Em dom., 22 de ago. de 2021 às 14:50, Tom Lane escreveu:
> Justin Pryzby writes:
> > This looks useful, thanks. It seems like maybe win64 builds are very
> slow
> > running this:
>
> > exec_stmt_block() /
> > BeginInternalSubTransaction() /
> > AbortSubTransaction() /
> > reschedule_timeouts()
gt; ; pgsql-performa...@postgresql.org
> > Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2
>> and 13.4
>>
>>
>> On 8/22/21 4:11 PM, Tom Lane wrote:
>> > "l...@laurent-hasson.com" writes:
> -Original Message-
> From: Andrew Dunstan
> Sent: Sunday, August 22, 2021 17:27
> To: Tom Lane ; l...@laurent-hasson.com
> Cc: Justin Pryzby ; Ranier Vilela
> ; pgsql-performa...@postgresql.org
> Subject: Re: Big Performance drop
On 8/22/21 4:11 PM, Tom Lane wrote:
> "l...@laurent-hasson.com" writes:
>> I do have a Linux install of 13.3, and things work beautifully, so this is
>> definitely a Windows thing here that started in V12.
> It's good to have a box around it, but that's still a pretty large
> box :-(.
>
> I'm h
> -Original Message-
> From: Tom Lane
> Sent: Sunday, August 22, 2021 16:11
> To: l...@laurent-hasson.com
> Cc: Justin Pryzby ; Ranier Vilela
> ; pgsql-performa...@postgresql.org
> Subject: Re: Big Performance drop of Exceptions
"l...@laurent-hasson.com" writes:
> I do have a Linux install of 13.3, and things work beautifully, so this is
> definitely a Windows thing here that started in V12.
It's good to have a box around it, but that's still a pretty large
box :-(.
I'm hoping that one of our Windows-using developers w
> > ; pgsql-performa...@postgresql.org
> > Subject: Re: Big Performance drop of Exceptions in UDFs between
> V11.2
> > and 13.4
> >
> > "l...@laurent-hasson.com" writes:
> > > So, now, in summary:
&
> -Original Message-
> From: Tom Lane
> Sent: Sunday, August 22, 2021 15:24
> To: l...@laurent-hasson.com
> Cc: Justin Pryzby ; Ranier Vilela
> ; pgsql-performa...@postgresql.org
> Subject: Re: Big Performance drop of Exceptions
"l...@laurent-hasson.com" writes:
> So, now, in summary:
> - I have tried V13.4, V12.3, 11.13, 11.2, 11.1 on several Windows VMs and my
> personal laptop (no VM).
> - All V11.x seem to behave uniformly.
> - Starting with 12.3, I am experiencing the major slowdown in the "with
> exceptions" scen
.@laurent-hasson.com;
> > pgsql-performa...@postgresql.org
> > Subject: Re: Big Performance drop of Exceptions in UDFs between
> V11.2
> > and 13.4
> >
> > Justin Pryzby writes:
> > >
> -Original Message-
> From: Tom Lane
> Sent: Sunday, August 22, 2021 13:51
> To: Justin Pryzby
> Cc: Ranier Vilela ; l...@laurent-hasson.com;
> pgsql-performa...@postgresql.org
> Subject: Re: Big Performance drop of Exceptions
> -Original Message-
> From: Justin Pryzby
> Sent: Sunday, August 22, 2021 11:48
> To: Ranier Vilela
> Cc: l...@laurent-hasson.com; Tom Lane ; pgsql-
> performa...@postgresql.org
> Subject: Re: Big Performance drop of Exceptions
Justin Pryzby writes:
> This looks useful, thanks. It seems like maybe win64 builds are very slow
> running this:
> exec_stmt_block() /
> BeginInternalSubTransaction() /
> AbortSubTransaction() /
> reschedule_timeouts() /
> schedule_alarm() /
> setitimer() /
> pg_timer_thread() /
> WaitForSingl
On Sun, Aug 22, 2021 at 10:50:47AM -0300, Ranier Vilela wrote:
> > Tried to check this with Very Sleepy at Windows 10 (bare metal).
> > Not sure it can help if someone can guide how to test this better?
> explain (analyze,buffers,COSTS,TIMING) select MAX(toFloat(a, null)) as "a"
> from sampletest
> -Original Message-
> From: l...@laurent-hasson.com
> Sent: Saturday, August 21, 2021 19:02
> To: Justin Pryzby
> Cc: Tom Lane ; pgsql-performa...@postgresql.org
> Subject: RE: Big Performance drop of Exceptions in UDFs between V1
> -Original Message-
> From: Justin Pryzby
> Sent: Saturday, August 21, 2021 18:17
> To: l...@laurent-hasson.com
> Cc: Tom Lane ; pgsql-performa...@postgresql.org
> Subject: Re: Big Performance drop of Exceptions in UDFs between V1
Could you send SELECT * FROM pg_config()
and try to find the CPU model ?
I think it's possible the hypervisor is trapping and emulating unhandled CPU
instructions.
Actually, it would be interesting to see if the performance differs between
11.2 and 11.13. It's possible that EDB compiled 11.13 on
> -Original Message-
> From: Justin Pryzby
> Sent: Saturday, August 21, 2021 15:20
> To: Tom Lane
> Cc: l...@laurent-hasson.com; pgsql-performa...@postgresql.org
> Subject: Re: Big Performance drop of Exceptions in UDFs between V1
> -Original Message-
> From: Tom Lane
> Sent: Saturday, August 21, 2021 14:05
> To: Mladen Gogala
> Cc: pgsql-performance@lists.postgresql.org
> Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2
> and 13.4
&g
On Sat, Aug 21, 2021 at 02:19:50PM -0500, Justin Pryzby wrote:
> As I recall, you're running postgres under a windows VM - I'm not sure if
> that's relevant.
I tried under a couple hyperv VMs but could not reproduce the issue (only an
~8x difference "with exceptions").
Which hypervisor are you us
I happen to have a stock 13.3 and 11.12 on Ubuntu here so I thought I'd
contribute numbers in case it's helpful:
v13.3:
Time: 4368.413 ms (00:04.368)
Time: 837.046 ms
v11.12:
Time: 5178.595 ms (00:05.179)
Time: 1027.857 ms (00:01.028)
So I'm also seeing a slight improvement in 13, not a degradat
On Sat, Aug 21, 2021 at 02:17:26PM -0400, Tom Lane wrote:
> "l...@laurent-hasson.com" writes:
> > So you mean that on average, the 4x overhead of exceptions is around what
> > you'd expect?
>
> Doesn't surprise me any, no. Exception recovery has to clean up after
> a wide variety of possible er
"l...@laurent-hasson.com" writes:
> So you mean that on average, the 4x overhead of exceptions is around what
> you'd expect?
Doesn't surprise me any, no. Exception recovery has to clean up after
a wide variety of possible errors, with only minimal assumptions about
what the system state had be
Mladen Gogala writes:
> I know that 14 is a beta version but the performance is significantly
> worse than v13 (I assume it's 13.4). Head revision is better than v14
> but still worse than v13. Can you expand a bit on the difference?
[ shrug... ] I don't see any meaningful differences between
-Original Message-
From: Tom Lane
Sent: Saturday, August 21, 2021 11:05
To: l...@laurent-hasson.com
Cc: pgsql-performa...@postgresql.org
Subject: Re: Big Performance drop of Exceptions in UDFs between V11.2 and 13.4
"l...@laurent-hasson.com" writes:
> OK... I apologize
I know that 14 is a beta version but the performance is significantly
worse than v13 (I assume it's 13.4). Head revision is better than v14
but still worse than v13. Can you expand a bit on the difference? Where
does the difference come from? Are there any differences in the
execution plan? I
"l...@laurent-hasson.com" writes:
> OK... I apologize for the long email before. Right after I sent it, I thought
> of a much simpler use-case to illustrate the issue which doesn't depend on
> any special data I have access o and complex pivoting. It's as raw as I can
> make it.
> I create a ta
t?
* Between 11.2 and 13.4, the no-exceptions scenario "b" performs 30x slower
(26.6s vs 774.1s).
Thank you!
Laurent Hasson.
From: l...@laurent-hasson.com
Sent: Saturday, August 21, 2021 03:57
To: pgsql-performa...@postgresql.org
Subject: Big Performance drop of Exceptions in
Hello all,
I think I have identified a major performance issue between V11.2 and 13.4 with
respect to exception handling in UDFs. I have the following simplified query
that pivots data and makes use of a UDF to convert data to a specific type, in
this case, float:
select "iccqa_iccassmt_fk"
75 matches
Mail list logo
