RE: performance degredation after upgrade from 9.6 to 12

"Andrew Zakharov" <[email protected]> Mon, 16 Dec 2019 15:27:01 +0300
Newsgroups gmane.comp.db.postgresql.performance
Message-ID <[email protected]>
This is a multipart message in MIME format.

------=_NextPart_000_0171_01D5B425.4355BE20
Content-Type: text/plain;
	charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi there =E2=80=93

I have no idea why this happening. But I suspect the parallel requires =
more internal machine resources like CPU etc because you can faster =
retrieve the disk data from the one hand but you ought to spend more =
resources for maintaining several threads and theirs coordination (px =
coordinator process in Oracle terms) from another one. Thus there could =
be more serious hardware requirements even just to keep performance the =
same. I believe that the real benefit of the parallel will be shown when =
you have pair of large and wide tables (30M or more each) with hash join =
(typical task for mart construction) but such class of databases is =
supposed to be big and required enough resources initially.

=20

=20

From: Mariel Cherkassky <[email protected]>=20
Sent: Monday, December 16, 2019 2:48 PM
To: Jeff Janes <[email protected]>
Cc: Andrew Zakharov <[email protected]>; =
[email protected]
Subject: Re: performance degredation after upgrade from 9.6 to 12

=20

Hey Jeff,Andrew,

I continued testing the 12version vs the 96 version and it seems that =
there is almost non diff and in some cases pg96 is faster than 12. I =
compared the content of pg_stat_statements after each test that I have =
done and it seems that the db time is almost the same and sometimes 96 =
is faster by 5%.

=20

Any idea why there isnt any improvement even when I enabled the parallel =
params in 12 ?

I can add a few examples if needed..


------=_NextPart_000_0171_01D5B425.4355BE20
Content-Type: text/html;
	charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<html xmlns:v=3D"urn:schemas-microsoft-com:vml" =
xmlns:o=3D"urn:schemas-microsoft-com:office:office" =
xmlns:w=3D"urn:schemas-microsoft-com:office:word" =
xmlns:m=3D"http://schemas.microsoft.com/office/2004/12/omml" =
xmlns=3D"http://www.w3.org/TR/REC-html40"><head><meta =
http-equiv=3DContent-Type content=3D"text/html; charset=3Dutf-8"><meta =
name=3DGenerator content=3D"Microsoft Word 15 (filtered =
medium)"><style><!--
/* Font Definitions */
@font-face
	{font-family:"Cambria Math";
	panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
	{font-family:Calibri;
	panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
	{margin:0cm;
	margin-bottom:.0001pt;
	font-size:12.0pt;
	font-family:"Times New Roman",serif;}
a:link, span.MsoHyperlink
	{mso-style-priority:99;
	color:#0563C1;
	text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
	{mso-style-priority:99;
	color:#954F72;
	text-decoration:underline;}
p.msonormal0, li.msonormal0, div.msonormal0
	{mso-style-name:msonormal;
	mso-margin-top-alt:auto;
	margin-right:0cm;
	mso-margin-bottom-alt:auto;
	margin-left:0cm;
	font-size:12.0pt;
	font-family:"Times New Roman",serif;}
span.EmailStyle18
	{mso-style-type:personal-reply;
	font-family:"Calibri",sans-serif;
	color:#1F497D;}
.MsoChpDefault
	{mso-style-type:export-only;
	font-family:"Calibri",sans-serif;
	mso-fareast-language:EN-US;}
@page WordSection1
	{size:612.0pt 792.0pt;
	margin:2.0cm 42.5pt 2.0cm 3.0cm;}
div.WordSection1
	{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext=3D"edit" spidmax=3D"1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext=3D"edit">
<o:idmap v:ext=3D"edit" data=3D"1" />
</o:shapelayout></xml><![endif]--></head><body lang=3DRU =
link=3D"#0563C1" vlink=3D"#954F72"><div class=3DWordSection1><p =
class=3DMsoNormal><span lang=3DEN-US =
style=3D'font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D;=
mso-fareast-language:EN-US'>Hi there =E2=80=93<o:p></o:p></span></p><p =
class=3DMsoNormal><span lang=3DEN-US =
style=3D'font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D;=
mso-fareast-language:EN-US'>I have no idea why this happening. But I =
suspect the parallel requires more internal machine resources like CPU =
etc because you can faster retrieve the disk data from the one hand but =
you ought to spend more resources for maintaining several threads and =
theirs coordination (px coordinator process in Oracle terms) from =
another one. Thus there could be more serious hardware requirements even =
just to keep performance the same. I believe that the real benefit of =
the parallel will be shown when you have pair of large and wide tables =
(30M or more each) with hash join (typical task for mart construction) =
but such class of databases is supposed to be big and required enough =
resources initially.<o:p></o:p></span></p><p class=3DMsoNormal><span =
lang=3DEN-US =
style=3D'font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D;=
mso-fareast-language:EN-US'><o:p>&nbsp;</o:p></span></p><p =
class=3DMsoNormal><span lang=3DEN-US =
style=3D'font-size:11.0pt;font-family:"Calibri",sans-serif;color:#1F497D;=
mso-fareast-language:EN-US'><o:p>&nbsp;</o:p></span></p><div =
style=3D'border:none;border-left:solid blue 1.5pt;padding:0cm 0cm 0cm =
4.0pt'><div><div style=3D'border:none;border-top:solid #E1E1E1 =
1.0pt;padding:3.0pt 0cm 0cm 0cm'><p class=3DMsoNormal><b><span =
style=3D'font-size:11.0pt;font-family:"Calibri",sans-serif'>From:</span><=
/b><span style=3D'font-size:11.0pt;font-family:"Calibri",sans-serif'> =
Mariel Cherkassky &lt;[email protected]&gt; <br><b>Sent:</b> =
Monday, December 16, 2019 2:48 PM<br><b>To:</b> Jeff Janes =
&lt;[email protected]&gt;<br><b>Cc:</b> Andrew Zakharov =
&lt;[email protected]&gt;; =
[email protected]<br><b>Subject:</b> Re: =
performance degredation after upgrade from 9.6 to =
12<o:p></o:p></span></p></div></div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p><div><div><p =
class=3DMsoNormal>Hey Jeff,Andrew,<o:p></o:p></p></div><div><p =
class=3DMsoNormal>I continued testing the 12version vs the 96 version =
and it seems that there is almost non diff and in some cases pg96 is =
faster than 12. I compared the content of pg_stat_statements after each =
test that I have done and it seems that the db time is almost the same =
and sometimes 96 is faster by 5%.<o:p></o:p></p></div><div><p =
class=3DMsoNormal><o:p>&nbsp;</o:p></p></div><div><p =
class=3DMsoNormal>Any idea why there isnt any improvement even when I =
enabled the parallel params in 12 ?<o:p></o:p></p></div><div><p =
class=3DMsoNormal>I can add a few examples if =
needed..<o:p></o:p></p></div></div></div></div></body></html>
------=_NextPart_000_0171_01D5B425.4355BE20--