You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@jmeter.apache.org by "Selvam T, Palani" <Pa...@altisource.com> on 2011/07/06 17:01:24 UTC
[GENERAL]Perf Testing on virtualized servers
Hi,
We are planning to do load testing on virtual machines as servers, which
is scaled down version of production environment. I would like to know
the pros & cons for virtualized servers for performance testing.
What are the things, should I take care...?
Thanks & Regards,
Palani Selvam
*******************************************************************************************************
This E-mail message and its attachments, if any are intended solely for the use of the addressee hereof. In addition, this message and the attachments, if any may contain information that is confidential, privileged and exempt from disclosure under applicable law. If you are not the intended recipient of this message, you are prohibited from reading, disclosing, reproducing, distributing, disseminating or otherwise using this transmission.
Delivery of this message to any person other than the intended recipient is not intended to waive any right or privilege. If you have received this message in error, please promptly notify the sender by reply E-mail and immediately delete this message from your system. Instructions transmitted over this system are not binding on us until they are confirmed by us. Message transmission is not guaranteed to be secure or free of software virus. While Altisource Portfolio Solutions and its subsidiaries collectively "Altisource" takes every reasonable precaution to minimize such risks, Altisource cannot accept liability for any damage sustained by you or any third party as a result of software viruses.
*******************************************************************************************************
RE: [GENERAL]Perf Testing on virtualized servers
Posted by "Selvam T, Palani" <Pa...@altisource.com>.
Deepak & Oliver,
I agree with you. As of now, there is no other choice. I need to live
with it. But it would be changed definitely in future.
Thanks & Regards,
Palani Selvam
-----Original Message-----
From: Oliver Lloyd [mailto:oliver_lloyd@hotmail.com]
Sent: Saturday, July 09, 2011 3:46 PM
To: jmeter-user@jakarta.apache.org
Subject: Re: [GENERAL]Perf Testing on virtualized servers
Deepak, you're absolutely right, practical realities for small/medium
projects are such that this type of thing happens and always will. The
trick
though, is to be aware of the limitations - if you understand why
scaling
and extrapolation are flawed then you can try to mitigate.
Too many people believe scaling is a valid solution with no problems and
that extrapolation works, this is bad for the industry in general.
--
View this message in context:
http://jmeter.512774.n5.nabble.com/GENERAL-Perf-Testing-on-virtualized-s
ervers-tp4557026p4567817.html
Sent from the JMeter - User mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
*******************************************************************************************************
This E-mail message and its attachments, if any are intended solely for the use of the addressee hereof. In addition, this message and the attachments, if any may contain information that is confidential, privileged and exempt from disclosure under applicable law. If you are not the intended recipient of this message, you are prohibited from reading, disclosing, reproducing, distributing, disseminating or otherwise using this transmission.
Delivery of this message to any person other than the intended recipient is not intended to waive any right or privilege. If you have received this message in error, please promptly notify the sender by reply E-mail and immediately delete this message from your system. Instructions transmitted over this system are not binding on us until they are confirmed by us. Message transmission is not guaranteed to be secure or free of software virus. While Altisource Portfolio Solutions and its subsidiaries collectively "Altisource" takes every reasonable precaution to minimize such risks, Altisource cannot accept liability for any damage sustained by you or any third party as a result of software viruses.
*******************************************************************************************************
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: [GENERAL]Perf Testing on virtualized servers
Posted by Oliver Lloyd <ol...@hotmail.com>.
Deepak, you're absolutely right, practical realities for small/medium
projects are such that this type of thing happens and always will. The trick
though, is to be aware of the limitations - if you understand why scaling
and extrapolation are flawed then you can try to mitigate.
Too many people believe scaling is a valid solution with no problems and
that extrapolation works, this is bad for the industry in general.
--
View this message in context: http://jmeter.512774.n5.nabble.com/GENERAL-Perf-Testing-on-virtualized-servers-tp4557026p4567817.html
Sent from the JMeter - User mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: [GENERAL]Perf Testing on virtualized servers
Posted by Kirk <ki...@gmail.com>.
On Jul 9, 2011, at 4:02 AM, Deepak Shetty wrote:
>> This is arguably worse than trying to use a scaled down environment for
> load
>> testing and then extrapolating the results up. That generally doesn't work
>> either.
> And yet the world goes on :). Practical realities of most small and medium
> sized projects usually mean that the above is the norm.
It is often *much* easier for small and medium sized organizations to setup representative test environments as they generally have much less to setup. When ever I've walked into a small or medium sized organization I generally have no problems replicating with a few hours of work. The reality is that *most* developers either don't understand or value performance testing. They think they can solve any problem that might show up with minimal effort. Some times it works, other times it sinks the project. And in small to medium sized that generally means it sinks the company behind the project.
That is my experience.
Regards,
Kirk
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: [GENERAL]Perf Testing on virtualized servers
Posted by Deepak Shetty <sh...@gmail.com>.
>This is arguably worse than trying to use a scaled down environment for
load
>testing and then extrapolating the results up. That generally doesn't work
>either.
And yet the world goes on :). Practical realities of most small and medium
sized projects usually mean that the above is the norm.
regards
deepak
On Fri, Jul 8, 2011 at 1:15 PM, Oliver Lloyd <ol...@hotmail.com>wrote:
> Kirk is completely right. No doubt.
>
> Virtualisation can be useful when building a test rig to use as injectors /
> generators. But it is a disaster if you try and virtualise the test system
> (the AUT) if your production is not also virtualised.
>
> This is arguably worse than trying to use a scaled down environment for
> load
> testing and then extrapolating the results up. That generally doesn't work
> either.
>
> --
> View this message in context:
> http://jmeter.512774.n5.nabble.com/GENERAL-Perf-Testing-on-virtualized-servers-tp4557026p4566365.html
> Sent from the JMeter - User mailing list archive at Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>
>
Re: [GENERAL]Perf Testing on virtualized servers
Posted by Oliver Lloyd <ol...@hotmail.com>.
Kirk is completely right. No doubt.
Virtualisation can be useful when building a test rig to use as injectors /
generators. But it is a disaster if you try and virtualise the test system
(the AUT) if your production is not also virtualised.
This is arguably worse than trying to use a scaled down environment for load
testing and then extrapolating the results up. That generally doesn't work
either.
--
View this message in context: http://jmeter.512774.n5.nabble.com/GENERAL-Perf-Testing-on-virtualized-servers-tp4557026p4566365.html
Sent from the JMeter - User mailing list archive at Nabble.com.
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: [GENERAL]Perf Testing on virtualized servers
Posted by Kirk <ki...@gmail.com>.
well, they all pretty much confirm what I've been saying. There are plenty of good reasons to virtualize, performance (under the vast majority of use cases) isn't one of them. Software rarely scales linearly with respect to it's consumption of hardware. Virtualization will parcel out hardware to your application how it sees fit and that is generally in a different pattern with different scaling characteristics than the native environment would have. Net result, test != prod. Conclusion, test environment is likely to expose phantom performance bottlenecks that once "fixed" will likely offer very little if any difference in the performance of the production system. virtualization
advice, advance at your own peril with this one ;-)
Regards,
Kirk
On Jul 8, 2011, at 7:04 AM, Selvam T, Palani wrote:
> Kirk,
> Thanks for the points.
>
> I have gone through following links and suggested not to use virtualized
> servers..
>
> Virtualization: Performance testing -
> http://www.macworld.com/article/145878/2010/01/virtulapptesting.html
>
> Measuring the Performance Impact of Virtualizing a Web Application
> Server
> http://www.webperformance.com/library/reports/LoadTestingVirtualizationP
> erformance/index.html
>
> Response to 'Load Testing a Virtual Web Application'
> http://blogs.vmware.com/vmtn/2007/04/response_to_loa.html
>
> Virtualization performance testing tips
> http://blogs.intel.com/idf/2008/08/virtualization_performance_tes.php
>
> Thanks & Regards,
> Palani Selvam
>
>
> -----Original Message-----
> From: Kirk [mailto:kirk.pepperdine@gmail.com]
> Sent: Thursday, July 07, 2011 11:59 PM
> To: JMeter Users List
> Subject: Re: [GENERAL]Perf Testing on virtualized servers
>
>
> 1) You cannot virtualize yourself into more hardware
> 2) prod != test environment == potential to shift bottlenecks.
> 3) not testing in the environment of deployment significantly increases
> the risk of the deployment failing
>
> I see no pro's
> Regards,
> Kirk
> On Jul 6, 2011, at 5:01 PM, Selvam T, Palani wrote:
>
>> Hi,
>>
>>
>>
>> We are planning to do load testing on virtual machines as servers,
> which
>> is scaled down version of production environment. I would like to know
>> the pros & cons for virtualized servers for performance testing.
>>
>>
>>
>> What are the things, should I take care...?
>>
>>
>>
>>
>>
>> Thanks & Regards,
>>
>> Palani Selvam
>>
>>
>>
>>
>>
>>
>>
> ************************************************************************
> *******************************
>>
>> This E-mail message and its attachments, if any are intended solely
> for the use of the addressee hereof. In addition, this message and the
> attachments, if any may contain information that is confidential,
> privileged and exempt from disclosure under applicable law. If you are
> not the intended recipient of this message, you are prohibited from
> reading, disclosing, reproducing, distributing, disseminating or
> otherwise using this transmission.
>>
>>
>> Delivery of this message to any person other than the intended
> recipient is not intended to waive any right or privilege. If you have
> received this message in error, please promptly notify the sender by
> reply E-mail and immediately delete this message from your system.
> Instructions transmitted over this system are not binding on us until
> they are confirmed by us. Message transmission is not guaranteed to be
> secure or free of software virus. While Altisource Portfolio Solutions
> and its subsidiaries collectively "Altisource" takes every reasonable
> precaution to minimize such risks, Altisource cannot accept liability
> for any damage sustained by you or any third party as a result of
> software viruses.
>>
>>
> ************************************************************************
> *******************************
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>
>
>
>
> *******************************************************************************************************
>
> This E-mail message and its attachments, if any are intended solely for the use of the addressee hereof. In addition, this message and the attachments, if any may contain information that is confidential, privileged and exempt from disclosure under applicable law. If you are not the intended recipient of this message, you are prohibited from reading, disclosing, reproducing, distributing, disseminating or otherwise using this transmission.
>
>
> Delivery of this message to any person other than the intended recipient is not intended to waive any right or privilege. If you have received this message in error, please promptly notify the sender by reply E-mail and immediately delete this message from your system. Instructions transmitted over this system are not binding on us until they are confirmed by us. Message transmission is not guaranteed to be secure or free of software virus. While Altisource Portfolio Solutions and its subsidiaries collectively "Altisource" takes every reasonable precaution to minimize such risks, Altisource cannot accept liability for any damage sustained by you or any third party as a result of software viruses.
>
> *******************************************************************************************************
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
RE: [GENERAL]Perf Testing on virtualized servers
Posted by "Selvam T, Palani" <Pa...@altisource.com>.
Kirk,
Thanks for the points.
I have gone through following links and suggested not to use virtualized
servers..
Virtualization: Performance testing -
http://www.macworld.com/article/145878/2010/01/virtulapptesting.html
Measuring the Performance Impact of Virtualizing a Web Application
Server
http://www.webperformance.com/library/reports/LoadTestingVirtualizationP
erformance/index.html
Response to 'Load Testing a Virtual Web Application'
http://blogs.vmware.com/vmtn/2007/04/response_to_loa.html
Virtualization performance testing tips
http://blogs.intel.com/idf/2008/08/virtualization_performance_tes.php
Thanks & Regards,
Palani Selvam
-----Original Message-----
From: Kirk [mailto:kirk.pepperdine@gmail.com]
Sent: Thursday, July 07, 2011 11:59 PM
To: JMeter Users List
Subject: Re: [GENERAL]Perf Testing on virtualized servers
1) You cannot virtualize yourself into more hardware
2) prod != test environment == potential to shift bottlenecks.
3) not testing in the environment of deployment significantly increases
the risk of the deployment failing
I see no pro's
Regards,
Kirk
On Jul 6, 2011, at 5:01 PM, Selvam T, Palani wrote:
> Hi,
>
>
>
> We are planning to do load testing on virtual machines as servers,
which
> is scaled down version of production environment. I would like to know
> the pros & cons for virtualized servers for performance testing.
>
>
>
> What are the things, should I take care...?
>
>
>
>
>
> Thanks & Regards,
>
> Palani Selvam
>
>
>
>
>
>
>
************************************************************************
*******************************
>
> This E-mail message and its attachments, if any are intended solely
for the use of the addressee hereof. In addition, this message and the
attachments, if any may contain information that is confidential,
privileged and exempt from disclosure under applicable law. If you are
not the intended recipient of this message, you are prohibited from
reading, disclosing, reproducing, distributing, disseminating or
otherwise using this transmission.
>
>
> Delivery of this message to any person other than the intended
recipient is not intended to waive any right or privilege. If you have
received this message in error, please promptly notify the sender by
reply E-mail and immediately delete this message from your system.
Instructions transmitted over this system are not binding on us until
they are confirmed by us. Message transmission is not guaranteed to be
secure or free of software virus. While Altisource Portfolio Solutions
and its subsidiaries collectively "Altisource" takes every reasonable
precaution to minimize such risks, Altisource cannot accept liability
for any damage sustained by you or any third party as a result of
software viruses.
>
>
************************************************************************
*******************************
>
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
*******************************************************************************************************
This E-mail message and its attachments, if any are intended solely for the use of the addressee hereof. In addition, this message and the attachments, if any may contain information that is confidential, privileged and exempt from disclosure under applicable law. If you are not the intended recipient of this message, you are prohibited from reading, disclosing, reproducing, distributing, disseminating or otherwise using this transmission.
Delivery of this message to any person other than the intended recipient is not intended to waive any right or privilege. If you have received this message in error, please promptly notify the sender by reply E-mail and immediately delete this message from your system. Instructions transmitted over this system are not binding on us until they are confirmed by us. Message transmission is not guaranteed to be secure or free of software virus. While Altisource Portfolio Solutions and its subsidiaries collectively "Altisource" takes every reasonable precaution to minimize such risks, Altisource cannot accept liability for any damage sustained by you or any third party as a result of software viruses.
*******************************************************************************************************
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
Re: [GENERAL]Perf Testing on virtualized servers
Posted by Kirk <ki...@gmail.com>.
1) You cannot virtualize yourself into more hardware
2) prod != test environment == potential to shift bottlenecks.
3) not testing in the environment of deployment significantly increases the risk of the deployment failing
I see no pro's
Regards,
Kirk
On Jul 6, 2011, at 5:01 PM, Selvam T, Palani wrote:
> Hi,
>
>
>
> We are planning to do load testing on virtual machines as servers, which
> is scaled down version of production environment. I would like to know
> the pros & cons for virtualized servers for performance testing.
>
>
>
> What are the things, should I take care...?
>
>
>
>
>
> Thanks & Regards,
>
> Palani Selvam
>
>
>
>
>
>
> *******************************************************************************************************
>
> This E-mail message and its attachments, if any are intended solely for the use of the addressee hereof. In addition, this message and the attachments, if any may contain information that is confidential, privileged and exempt from disclosure under applicable law. If you are not the intended recipient of this message, you are prohibited from reading, disclosing, reproducing, distributing, disseminating or otherwise using this transmission.
>
>
> Delivery of this message to any person other than the intended recipient is not intended to waive any right or privilege. If you have received this message in error, please promptly notify the sender by reply E-mail and immediately delete this message from your system. Instructions transmitted over this system are not binding on us until they are confirmed by us. Message transmission is not guaranteed to be secure or free of software virus. While Altisource Portfolio Solutions and its subsidiaries collectively "Altisource" takes every reasonable precaution to minimize such risks, Altisource cannot accept liability for any damage sustained by you or any third party as a result of software viruses.
>
> *******************************************************************************************************
>
---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org