Received: from malur.postgresql.org ([217.196.149.56]) by arkaria.postgresql.org with esmtp (Exim 4.84_2) (envelope-from ) id 1eFHte-0000T3-4v for pgsql-admin@arkaria.postgresql.org; Thu, 16 Nov 2017 10:59:50 +0000 Received: from localhost ([127.0.0.1] helo=postgresql.org) by malur.postgresql.org with smtp (Exim 4.84_2) (envelope-from ) id 1eFHtd-0004tf-LB for pgsql-admin@arkaria.postgresql.org; Thu, 16 Nov 2017 10:59:49 +0000 Received: from magus.postgresql.org ([2a02:c0:301:0:ffff::29]) by malur.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA384:256) (Exim 4.84_2) (envelope-from ) id 1eFHtc-0004tK-UQ for pgsql-admin@postgresql.org; Thu, 16 Nov 2017 10:59:49 +0000 Received: from mail-wm0-x231.google.com ([2a00:1450:400c:c09::231]) by magus.postgresql.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_CBC_SHA1:256) (Exim 4.89) (envelope-from ) id 1eFHtY-00006p-BD for pgsql-admin@postgresql.org; Thu, 16 Nov 2017 10:59:48 +0000 Received: by mail-wm0-x231.google.com with SMTP id r68so8634655wmr.1 for ; Thu, 16 Nov 2017 02:59:44 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20161025; h=from:to:subject:date:message-id:mime-version:thread-index :content-language; bh=z/X5ikPoktLGUMWkn0+Mb2V4L0gNQMGODb5lyIusx/U=; b=UFlBO/udD6YMXlyPlx/cln0/IrsjW6QODsAVA6nL9ahoPWvvfuKEjN8i4YqChLQpYe aLxkNwTQBTq0tSxlpcoCXwxAbrvLXe8P95eJgG7Z/uN1wYsDo1MAJuvJRbgcthDf7zuS w12yRsf4v/pT/6EHvQFDRU1lobfIn/Y5pS1PW4ZtNe5qjnbkjk4H99HglobXXaJwNT3x 1QetMnVyR1qat5U1xYxyp6X5zeAm6qRYio8vzCjeuvODuHkQSRJUbz24r6Czv7vjpLd9 jNtNrijXPz/uTXjn25M1E305ER9k3TY0fdn80ewpvxw6qZu/zCSYQLlcFVrzGxcM/gvK tEKg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:to:subject:date:message-id:mime-version :thread-index:content-language; bh=z/X5ikPoktLGUMWkn0+Mb2V4L0gNQMGODb5lyIusx/U=; b=VpERAntvzALpq76CVxHiqeXvO1gT5Yf2QuBr/wfMty8/tbdn+7denKQgciKSVIWQV1 mFQbquTrgITfWvjTBAg0XjCiAsGMmGn5Y5pgfbLbSkzA3rRaBGnp4XlK2pawk5gjQJAR eZVhRpAo42SZsMZexzTKOgPZhs6+XEmEbjFKuEybmM9TK5IbzrFaD45kGia0zwsDVwKi Qj8NLhmOxaodR09cqp9r73+y+QsWyb8Vvmu0SstRQYb2k26K9xA9pK+inWGg6J16frRC jhO0qs91FpemWC1aKIYB7hqoX0KiCezWX/6ga8dgSvF02bw/E9vTSa3OUpXwUtZHs5XT ImTg== X-Gm-Message-State: AJaThX5G5oxbC22j/5ifMrdHxRPTz8vj1PXdDBAXq7FTihEY/u87TQab aBDbIyz06wmLVlYSLFWEOq+AMw== X-Google-Smtp-Source: AGs4zMboTBKWBXJwkZyV9S9PSEbklUeX28xXrGppxW1jxAzPeD4jnMS/IGgbjo/peF2rAATqhtz7wA== X-Received: by 10.80.140.170 with SMTP id q39mr2276355edq.186.1510829983085; Thu, 16 Nov 2017 02:59:43 -0800 (PST) Received: from PC016 (mail.ibeq.com. [212.100.42.241]) by smtp.gmail.com with ESMTPSA id w7sm870572edw.65.2017.11.16.02.59.41 for (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Thu, 16 Nov 2017 02:59:42 -0800 (PST) From: To: Subject: [SOLVED] AW: Performance difference between servers Date: Thu, 16 Nov 2017 11:59:40 +0100 Message-ID: <005301d35eca$0102ecd0$0308c670$@gmail.com> MIME-Version: 1.0 Content-Type: multipart/alternative; boundary="----=_NextPart_000_0054_01D35ED2.62C8DB70" X-Mailer: Microsoft Outlook 16.0 Thread-Index: AdNexmQujSlespXTReGhuhG6ox5nqA== Content-Language: de List-Archive: List-Help: List-ID: List-Owner: List-Post: List-Subscribe: List-Unsubscribe: X-Mailing-List: pgsql-admin Precedence: bulk Sender: pgsql-admin-owner@postgresql.org This is a multipart message in MIME format. ------=_NextPart_000_0054_01D35ED2.62C8DB70 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable In short: setting the energy options on the host to =E2=80=9Ehigh = performance=E2=80=9C solved the issue.=20 Side note: The energy option has impact only on server #2 and #3, not on = #1 (see below).=20 =20 In detail: BIOS RAID 1, different processors, operatings systems and a = mix of fixed and dynamic virtuals disks did not explain a factor of 2 = =E2=80=A6 3 in performance. Changing the virtual disk from dynamic to = fixed, different settings in postgresql.conf, changing memory allocation = for the virtual guests did not show any significant improvements = =E2=80=93 the oldest server with the slowest processor, minimum RAM and = slow disks was still performing better than the others.=20 All hosts were running on the energy option =E2=80=9Ebalanced=E2=80=9C = with no modification. However, the processor in server #1 was always = running at 3.4 GHz and did not reduce clock speed in idle mode. Server = #2 and #3 were running at 0.8 GHz in idle mode.=20 The real issue was that server #2 and #3 did not speed up when the = database operation started =E2=80=93 not to nominal speed and clearly = not entering turbo mode.=20 =20 Solution for server #2 and #3:=20 1. activate high performance mode =3D> this causes the processor to run = at nominal speed at all times 2. reduce the setting for processor minimum performance back to 5%=20 3. activate high performance mode again =3D> this re-enables the dynamic = clock speed but ensures that the processor speeds up when needed =20 Why is this not needed for server #1?=20 I believe that the processor running in server #1 is simply too old to = reduce the clock speed. Perhaps a missing or outdated chipset driver is = also somehow involved (that=E2=80=99s just a guess).=20 =20 Thanks to everybody providing input =20 Klaus =20 Von: Rui DeSousa [mailto:rui.desousa@icloud.com]=20 Gesendet: Donnerstag, 16. November 2017 00:43 An: kpi6288@gmail.com Cc: pgsql-admin@postgresql.org Betreff: Re: [ADMIN] Performance difference between servers =20 There is a log of variables there. Is the i7 single core performance = faster than the E3? Bus/memory speeds, etc. I would also consider any = vCPU imbalances. You could end up with guest scheduling issue; i.e if = guest requires 1 core but is allocated 4 cores; then for 4 cores still = need to be available and ready to schedule the guest even if it current = need is only 1 core at the moment. =20 =20 On Nov 15, 2017, at 12:43 PM, kpi6288@gmail.com = wrote: =20 Trying as best as a can:=20 =20 #1 HP 8200 Elite microtower, i7-2600, 16GB RAM, 1 SSD 480 GB, 1 SSD 900 = GB, Hyper-V 2012 R2 (not 2008 R2 as written in my original question)=20 #2 Lenovo TS 150, E3-1245 V5, 40GB RAM, 2 SSD 850GB as RAID 1 (BIOS = level), 2 HDDs 1TB as RAID 1 (BIOS level) #3 model unknown, E3-1270 V3, 32GB RAM, 2 SSD 480GB as RAID 1 (BIOS = level) =20 Server #1 is more a PC than a server.=20 =20 Server #1 runs in total 4 virtual guests. 8GB and 6 = =E2=80=9Ecores=E2=80=9C dedicated to the server running PostgreSQL and = 3GB to other machines which are almost 100% idle. Server #2 has currently on the one guets running PostgreSQL with 16GB = RAM and 6 =E2=80=9Ecores=E2=80=9C dedicated.=20 Server #3 runs 2 virtual guests. 12GB and 4 cores dedicated tot he guest = runnning PostgreSQL. 4 GB and 3 cores decicated to another guest with is = practically idle.=20 =20 Servers #2 and #3 have still plenty of free disk space while server #1 = is using approx. 70% oft he disks.=20 =20 Server #1 has much older SSDs than #2. Not sure about #3.=20 =20 Von: Scott Whitney [ = mailto:scott@journyx.com]=20 Gesendet: Mittwoch, 15. November 2017 18:19 An: kpi6288@gmail.com Betreff: Re: [ADMIN] Performance difference between servers =20 Can you post the hardware host specs for #1, #2 and #3? If those ARE the = hardware specs (for example, you list SSDs), can you list the resources = assigned to the guests? =20 _____ =20 From: kpi6288@gmail.com < = kpi6288@gmail.com> Sent: Wednesday, November 15, 2017 11:15 AM To: Scott Whitney Subject: AW: [ADMIN] Performance difference between servers =20 Yes, exactly: =20 Three different (hardware) machines, each having a hyper-V host and the = Windows Servers as Hyper-V guests.=20 =20 =20 Von: Scott Whitney [ = mailto:scott@journyx.com]=20 Gesendet: Mittwoch, 15. November 2017 18:11 An: kpi6288@gmail.com; = pgsql-admin@postgresql.org Betreff: Re: [ADMIN] Performance difference between servers =20 Are you saying that each of these servers is a virtual HyperV machine = running under a hardware host? =20 =20 =20 _____ =20 From: = pgsql-admin-owner@postgresql.org < = = pgsql-admin-owner@postgresql.org> on behalf of = kpi6288@gmail.com < = kpi6288@gmail.com> Sent: Wednesday, November 15, 2017 11:06 AM To: pgsql-admin@postgresql.org Subject: [ADMIN] Performance difference between servers=20 =20 We have 3 different servers, one (the oldest) ist significantly faster = at certain operations and I can=E2=80=99t find out the reason:=20 =20 (table created in monospace font) =20 Server Processor RAM CrystalDiskMark pgbench CREATE DATABASE = OS #1 i7-2600 8GB 495 / 473 26.000 4.5 =E2=80=A6 6 s = W/ Server 2008 R2 on Hyper V Server 2008=20 #2 E3-1245 16GB 797 / 514 50.000 13 =E2=80=A6 19 = s W/ Server 2016 Std. on Hyper V Server 2016 #3 E3-1270 12GB 933 / 334 40.000 13 =E2=80=A6 20 = s W/ Server 2008 Essentials on Hyper V Server 2016 =20 CrystalDiskMark: =E2=80=9ESeq Q32T1=E2=80=9C Pgbench:=20 pgbench -i -s 200 -U postgres pgbench pgbench -c 40 -j 10 -T 150 -U postgres -S -P 5 pgbench CREATE DATABASE:=20 create database test with template myTemplate =20 All running PostgreSQL 9.6 x64 with similar configuration. All running = as virtual machines, no other virtual machines creating any load. Using = the same template database on all machines.=20 =20 Although the indicative numbers of pgbench and CrystalDiskMark show = better values for server #2 and #3, server #1 creates a new database = considerably faster.=20 All servers have SSDs. Server #1 has two single SSDs (a system disk and = a data disk). Servers #2 and #3 have only one logical disk for system = and data, and this is two SSDs running as RAID 1. This may be slower but = if it is, I would excpect this also to show up in the pgbench results.=20 =20 The reason behind this exercise is that a test application was also = reported to be slower on server #2 and 3.=20 =20 Any idea how to track down the performance difference?=20 =20 Thank you=20 Klaus =20 =20 =20 Journyx, Inc. 7600 Burnet Road #300=20 Austin, TX 78757=20 www.journyx.com =20 p 512.834.8888=20 f 512-834-8858=20 =20 Do you receive our promotional emails? Click = here or visit = = http://journyx.com/communication-preferences to unsubscribe. =20 ------=_NextPart_000_0054_01D35ED2.62C8DB70 Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: quoted-printable

In short: setting the energy = options on the host to =E2=80=9Ehigh performance=E2=80=9C solved the = issue.

Side note: The energy option has = impact only on server #2 and #3, not on #1 (see below). =

 

In detail: = BIOS RAID 1, different processors, operatings systems and a mix of fixed = and dynamic virtuals disks did not explain a factor of 2 =E2=80=A6 3 in = performance. Changing the virtual disk from dynamic to fixed, different = settings in postgresql.conf, changing memory allocation for the virtual = guests did not show any significant improvements =E2=80=93 the oldest = server with the slowest processor, minimum RAM and slow disks was still = performing better than the others.

All hosts = were running on the energy option =E2=80=9Ebalanced=E2=80=9C with no = modification. However, the processor in server #1 was always running at = 3.4 GHz and did not reduce clock speed in idle mode. Server #2 and #3 = were running at 0.8 GHz in idle mode.

The real = issue was that server #2 and #3 did not speed up when the database = operation started =E2=80=93 not to nominal speed and clearly not = entering turbo mode.

 

Solution = for server #2 and #3:

1. activate high performance mode = =3D> this causes the processor to run at nominal speed at all = times

2. reduce the setting for processor = minimum performance back to 5%

3. activate = high performance mode again =3D> this re-enables the dynamic clock = speed but ensures that the processor speeds up when = needed

 

Why is this = not needed for server #1?

I believe = that the processor running in server #1 is simply too old to reduce the = clock speed. Perhaps a missing or outdated chipset driver is also = somehow involved (that=E2=80=99s just a guess).

 

Thanks to = everybody providing input

 

Klaus

 

Von: Rui DeSousa = [mailto:rui.desousa@icloud.com]
Gesendet: Donnerstag, 16. = November 2017 00:43
An: kpi6288@gmail.com
Cc: = pgsql-admin@postgresql.org
Betreff: Re: [ADMIN] Performance = difference between servers

 

There is a = log of variables there.  Is the i7 single core performance faster = than the E3? Bus/memory speeds, etc. I would also consider any vCPU = imbalances.  You could end up with guest scheduling issue; i.e if = guest requires 1 core but is allocated 4 cores; then for 4 cores still = need to be available and ready to schedule the guest even if it current = need is only 1 core at the moment.

 

 

On Nov 15, 2017, at 12:43 PM, kpi6288@gmail.com = wrote:

 

Trying as best as a can: 

 

#1  HP 8200 = Elite microtower, i7-2600, 16GB RAM, 1 SSD 480 GB, 1 SSD 900 GB, Hyper-V = 2012 R2 (not 2008 R2 as written in my original question) 

#2  = Lenovo TS 150, E3-1245 V5, 40GB RAM, 2 SSD 850GB as RAID 1 (BIOS level), = 2 HDDs 1TB as RAID 1 (BIOS level)

#3  model = unknown, E3-1270 V3, 32GB RAM, 2 SSD 480GB as RAID 1 (BIOS = level)

 

Server #1 is more a PC than a server. 

 

Server #1 runs in total 4 virtual guests. 8GB and 6 = =E2=80=9Ecores=E2=80=9C dedicated to the server running PostgreSQL and = 3GB to other machines which are almost 100% = idle.

Server #2 has = currently on the one guets running PostgreSQL with 16GB RAM and 6 = =E2=80=9Ecores=E2=80=9C dedicated. 

Server #3 runs 2 virtual guests. 12GB and 4 cores = dedicated tot he guest runnning PostgreSQL. 4 GB and 3 cores decicated = to another guest with is practically idle. 

 

Servers #2 and #3 have still plenty of free disk space = while server #1 is using approx. 70% oft he disks. 

 

Server #1 has much older SSDs than #2. Not sure about = #3. 

 

Von: Scott Whitney [mailto:scott@journyx.com] 
Gesendet: Mittwoch, 15. November 2017 = 18:19
An: kpi6288@gmail.com
Betreff: Re: [ADMIN] Performance = difference between servers

 

Can you = post the hardware host specs for #1, #2 and #3? If those ARE the = hardware specs (for example, you list SSDs), can you list the resources = assigned to the guests?<= /span>

 

From: kpi6288@gmail.com <kpi6288@gmail.com>
Sent: Wednesday, November 15, = 2017 11:15 AM
To: Scott = Whitney
Subject: AW: [ADMIN] Performance = difference between servers

 

=

Yes, = exactly:

 

Three different = (hardware) machines, each having a hyper-V host and the Windows Servers = as Hyper-V guests. 

 

 

Von: Scott Whitney [mailto:scott@journyx.com] 
Gesendet: Mittwoch, 15. November 2017 = 18:11
An: kpi6288@gmail.com; pgsql-admin@postgresql.org
Betref= f: Re: [ADMIN] = Performance difference between = servers

 

Are you = saying that each of these servers is a virtual HyperV machine running = under a hardware host?<= /span>

 <= /span>

 <= /span>

 

From: pgsql-admin-owner@postgresql.org <pgsql-admin-owner@postgresql.org> = on behalf of kpi6288@gmail.com <kpi6288@gmail.com>
Sent: Wednesday, November 15, = 2017 11:06 AM
To: pgsql-admin@postgresql.org
Subjec= t: [ADMIN] = Performance difference between servers 

 

=

We = have 3 different servers, one (the oldest) ist significantly faster at = certain operations and I can=E2=80=99t find out the reason: 

 

(table created in = monospace font)

 

Server  Processor =  RAM   CrystalDiskMark   pgbench   = CREATE DATABASE   OS

  #1  = ;  i7-2600   8GB     495 / = 473   =      26.000      4.5 = =E2=80=A6 6 s      W/ Server 2008 R2 on Hyper V = Server 2008 

   #2 = ;   E3-1245   16GB    797 / = 514    =     50.000      13 = =E2=80=A6 19 s      W/ Server 2016 Std. on = Hyper V Server 2016

  =  #3    E3-1270   12GB  =   933 / 334        = 40.000      13 =E2=80=A6 20 = s      W/ Server 2008 Essentials on Hyper V = Server 2016

 

CrystalDiskMark: = =E2=80=9ESeq Q32T1=E2=80=9C

Pgbench: 

   pgbench -i -s 200 -U = postgres pgbench

   pgbench = -c 40 -j 10 -T 150 -U postgres -S -P 5 = pgbench

CREATE DATABASE: 

   create database test with = template myTemplate

 

All running = PostgreSQL 9.6 x64 with similar configuration. All running as virtual = machines, no other virtual machines creating any load. Using the same = template database on all machines. 

 

Although the = indicative numbers of pgbench and CrystalDiskMark show better values for = server #2 and #3, server #1 creates a new database considerably = faster. 

All servers = have SSDs. Server #1 has two single SSDs (a system disk and a data = disk). Servers #2 and #3 have only one logical disk for system and data, = and this is two SSDs running as RAID 1. This may be slower but if it is, = I would excpect this also to show up in the pgbench results. 

 

The reason behind = this exercise is that a test application was also reported to be slower = on server #2 and 3. 

 

Any idea how to track = down the performance difference? 

 

Thank you 

Klaus

 

=

 

 

<= div>

Journyx, = Inc.

7600 Burnet Road = #300 
Austin, TX = 78757 
www.journyx.com

 

<= div>

p = 512.834.8888 

f = 512-834-8858 

 

=

Do you = receive our promotional emails? Click here or visit http://journyx.com/communication-preferences to = unsubscribe.

 

------=_NextPart_000_0054_01D35ED2.62C8DB70--