Quantcast

IOmeter Bench-marking Analysis

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

IOmeter Bench-marking Analysis

Rizul khanna
Hello All, 

Greetings.

I have one concern. How to decide on the duration of the benchmark test while using IOmeter ? 

I am working on a New HyperV 2012 R2 Host configured with 6 Disks(4 SATA + 2 SSD) with Auto-tiering enabled between the two disk types and the volume derived out of this pool is configured with RAID 1. Thus, no striping is involved. I have been assigned to check for the Max. stress this physical host can handle. 

Can you please brief on what is the best possible roadmap for this, the host has 192 GB of RAM and 2*6 CPUs ? Our applications are Read-intensive, but no idea about the random/ sequential nature of the I/O and also, I have no idea about the I/O Size. What can be the duration of any such test ? 

Please let me know if more details would be required ? Please advise.




Warm Regards,

Rizul Khanna



------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: IOmeter Bench-marking Analysis

Robert Randall
You write "Max stress".  That is a very difficult thing to understand.  What applications will you be running on your HyperV guest machines?  If I new that I might be able to suggest how to stress the machine.  Will the guests be file servers, email servers, web servers, database servers, remote desktop servers, etc?


On Wed, Apr 9, 2014 at 5:46 AM, Rizul khanna <[hidden email]> wrote:
Hello All, 

Greetings.

I have one concern. How to decide on the duration of the benchmark test while using IOmeter ? 

I am working on a New HyperV 2012 R2 Host configured with 6 Disks(4 SATA + 2 SSD) with Auto-tiering enabled between the two disk types and the volume derived out of this pool is configured with RAID 1. Thus, no striping is involved. I have been assigned to check for the Max. stress this physical host can handle. 

Can you please brief on what is the best possible roadmap for this, the host has 192 GB of RAM and 2*6 CPUs ? Our applications are Read-intensive, but no idea about the random/ sequential nature of the I/O and also, I have no idea about the I/O Size. What can be the duration of any such test ? 

Please let me know if more details would be required ? Please advise.




Warm Regards,

Rizul Khanna

r[hidden email]  |  <a href="tel:%2B91%209501074400" value="+919501074400" target="_blank">+91 9501074400, <a href="tel:%2B91%208595370298" value="+918595370298" target="_blank">+91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  



------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user




--
Robert Randall | [hidden email]

------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: IOmeter Bench-marking Analysis

Rizul khanna
In reply to this post by Rizul khanna

Hello Robert, 

Thanks a lot for sharing interest.

We will be hosting financial applications like Quickbooks, Sage, ATX, Peachtree, Drake and many others on these Windows based VMs. These machines will be used for Remote Desktop Connections via VPN/ Secured Public IP access. 

As per our observation based on I/O consumption patterns, these Windows 2008 R2 Datacenter based Guest machines are expected to do more than 60% of Read most of the times.

Please let me know if you would need some more inputs from my side.



Warm Regards,

Rizul Khanna

System Administrator

[hidden email]  |  +91 9501074400, +91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  




On Wed, Apr 9, 2014 at 10:59 PM, Robert Randall <[hidden email]> wrote:
You write "Max stress".  That is a very difficult thing to understand.  What applications will you be running on your HyperV guest machines?  If I new that I might be able to suggest how to stress the machine.  Will the guests be file servers, email servers, web servers, database servers, remote desktop servers, etc?


On Wed, Apr 9, 2014 at 5:46 AM, Rizul khanna <[hidden email]> wrote:
Hello All, 

Greetings.

I have one concern. How to decide on the duration of the benchmark test while using IOmeter ? 

I am working on a New HyperV 2012 R2 Host configured with 6 Disks(4 SATA + 2 SSD) with Auto-tiering enabled between the two disk types and the volume derived out of this pool is configured with RAID 1. Thus, no striping is involved. I have been assigned to check for the Max. stress this physical host can handle. 

Can you please brief on what is the best possible roadmap for this, the host has 192 GB of RAM and 2*6 CPUs ? Our applications are Read-intensive, but no idea about the random/ sequential nature of the I/O and also, I have no idea about the I/O Size. What can be the duration of any such test ? 

Please let me know if more details would be required ? Please advise.




Warm Regards,

Rizul Khanna

r[hidden email]  |  <a href="tel:%2B91%209501074400" value="+919501074400" target="_blank">+91 9501074400, <a href="tel:%2B91%208595370298" value="+918595370298" target="_blank">+91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  



------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user




--
Robert Randall | [hidden email]


------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: IOmeter Bench-marking Analysis

Robert Randall

I believe that all of the applications you mention will be using a file on the file system as a data store.  They will not be using an RDBMS or similar database server.  If that is the case and I can assume that you will be using NTFS as your file system that will narrow the parameters a bit.  Please let me know if my assumptions are correct and if so I can offer some advice on what to test.  Be sure to run your tests from the guest OS.

Best regards,
Robert


On Wed, Apr 9, 2014 at 12:39 PM, Rizul khanna <[hidden email]> wrote:

Hello Robert, 

Thanks a lot for sharing interest.

We will be hosting financial applications like Quickbooks, Sage, ATX, Peachtree, Drake and many others on these Windows based VMs. These machines will be used for Remote Desktop Connections via VPN/ Secured Public IP access. 

As per our observation based on I/O consumption patterns, these Windows 2008 R2 Datacenter based Guest machines are expected to do more than 60% of Read most of the times.

Please let me know if you would need some more inputs from my side.



Warm Regards,

Rizul Khanna

System Administrator

[hidden email]  |  <a href="tel:%2B91%209501074400" value="+919501074400" target="_blank">+91 9501074400, <a href="tel:%2B91%208595370298" value="+918595370298" target="_blank">+91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  




On Wed, Apr 9, 2014 at 10:59 PM, Robert Randall <[hidden email]> wrote:
You write "Max stress".  That is a very difficult thing to understand.  What applications will you be running on your HyperV guest machines?  If I new that I might be able to suggest how to stress the machine.  Will the guests be file servers, email servers, web servers, database servers, remote desktop servers, etc?


On Wed, Apr 9, 2014 at 5:46 AM, Rizul khanna <[hidden email]> wrote:
Hello All, 

Greetings.

I have one concern. How to decide on the duration of the benchmark test while using IOmeter ? 

I am working on a New HyperV 2012 R2 Host configured with 6 Disks(4 SATA + 2 SSD) with Auto-tiering enabled between the two disk types and the volume derived out of this pool is configured with RAID 1. Thus, no striping is involved. I have been assigned to check for the Max. stress this physical host can handle. 

Can you please brief on what is the best possible roadmap for this, the host has 192 GB of RAM and 2*6 CPUs ? Our applications are Read-intensive, but no idea about the random/ sequential nature of the I/O and also, I have no idea about the I/O Size. What can be the duration of any such test ? 

Please let me know if more details would be required ? Please advise.




Warm Regards,

Rizul Khanna

r[hidden email]  |  <a href="tel:%2B91%209501074400" value="+919501074400" target="_blank">+91 9501074400, <a href="tel:%2B91%208595370298" value="+918595370298" target="_blank">+91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  



------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user




--
Robert Randall | [hidden email]




--
Robert Randall | [hidden email]

------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: IOmeter Bench-marking Analysis

Rizul khanna
Yes, I agree NTFS would be used. 

Warm Regards,

Rizul Khanna

System Administrator

CloudSpace Technologies Pvt. Ltd., New Delhi, India

http://www.myrealdata.com/

[hidden email]  |  +91 9501074400, +91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  

The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, re-transmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or Entities other than the intended recipient are prohibited. If you have received this in error, please contact the sender and delete the material from any computer. The contents of this message may contain personal views, which are not the views of  CloudSpace Technologies Pvt. Ltd., unless specifically stated.



On Wed, Apr 9, 2014 at 11:22 PM, Robert Randall <[hidden email]> wrote:

I believe that all of the applications you mention will be using a file on the file system as a data store.  They will not be using an RDBMS or similar database server.  If that is the case and I can assume that you will be using NTFS as your file system that will narrow the parameters a bit.  Please let me know if my assumptions are correct and if so I can offer some advice on what to test.  Be sure to run your tests from the guest OS.

Best regards,
Robert


On Wed, Apr 9, 2014 at 12:39 PM, Rizul khanna <[hidden email]> wrote:

Hello Robert, 

Thanks a lot for sharing interest.

We will be hosting financial applications like Quickbooks, Sage, ATX, Peachtree, Drake and many others on these Windows based VMs. These machines will be used for Remote Desktop Connections via VPN/ Secured Public IP access. 

As per our observation based on I/O consumption patterns, these Windows 2008 R2 Datacenter based Guest machines are expected to do more than 60% of Read most of the times.

Please let me know if you would need some more inputs from my side.



Warm Regards,

Rizul Khanna

System Administrator

[hidden email]  |  <a href="tel:%2B91%209501074400" value="+919501074400" target="_blank">+91 9501074400, <a href="tel:%2B91%208595370298" value="+918595370298" target="_blank">+91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  




On Wed, Apr 9, 2014 at 10:59 PM, Robert Randall <[hidden email]> wrote:
You write "Max stress".  That is a very difficult thing to understand.  What applications will you be running on your HyperV guest machines?  If I new that I might be able to suggest how to stress the machine.  Will the guests be file servers, email servers, web servers, database servers, remote desktop servers, etc?


On Wed, Apr 9, 2014 at 5:46 AM, Rizul khanna <[hidden email]> wrote:
Hello All, 

Greetings.

I have one concern. How to decide on the duration of the benchmark test while using IOmeter ? 

I am working on a New HyperV 2012 R2 Host configured with 6 Disks(4 SATA + 2 SSD) with Auto-tiering enabled between the two disk types and the volume derived out of this pool is configured with RAID 1. Thus, no striping is involved. I have been assigned to check for the Max. stress this physical host can handle. 

Can you please brief on what is the best possible roadmap for this, the host has 192 GB of RAM and 2*6 CPUs ? Our applications are Read-intensive, but no idea about the random/ sequential nature of the I/O and also, I have no idea about the I/O Size. What can be the duration of any such test ? 

Please let me know if more details would be required ? Please advise.




Warm Regards,

Rizul Khanna

r[hidden email]  |  <a href="tel:%2B91%209501074400" value="+919501074400" target="_blank">+91 9501074400, <a href="tel:%2B91%208595370298" value="+918595370298" target="_blank">+91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  



------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user




--
Robert Randall | [hidden email]




--
Robert Randall | [hidden email]


------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: IOmeter Bench-marking Analysis

Robert Randall
Hi Rizul,

The next step would be to run fsutil from the guest OS and see what transfer sizes NTFS is using.  Here is an example:

C:\Windows\system32  Thu 04/10/2014  7:33:30.12
>fsutil fsinfo ntfsinfo c:
NTFS Volume Serial Number :       0x48bc80c6bc80b04c
Version :                         3.1
Number Sectors :                  0x0000000037df7ff8
Total Clusters :                  0x0000000006fbefff
Free Clusters  :                  0x000000000474d529
Total Reserved :                  0x00000000000007c0
Bytes Per Sector  :               512
Bytes Per Physical Sector :       4096
Bytes Per Cluster :               4096
Bytes Per FileRecord Segment    : 1024
Clusters Per FileRecord Segment : 0
Mft Valid Data Length :           0x000000002c0c0000
Mft Start Lcn  :                  0x00000000000c0000
Mft2 Start Lcn :                  0x0000000000000002
Mft Zone Start :                  0x000000000246a940
Mft Zone End   :                  0x00000000024701a0
RM Identifier:        50D667E4-E903-11E2-9405-0021CC5C9374

C:\Windows\system32  Thu 04/10/2014  7:33:33.88
>

The data above shows a logical sector size of 512 and a physical sector size of 4096.  Assuming that the applications you are hosting are using typical file system access (buffered IO) if may be a challenge to get application performance measures form IOMeter.  IOMeter always opens files in an unbuffered mode which will bypass all file system caching performed by Windows.  Do you know if these applications use single threaded access to their files or multithreaded access to their files?

Best regards,
Robert


On Wed, Apr 9, 2014 at 1:01 PM, Rizul khanna <[hidden email]> wrote:
Yes, I agree NTFS would be used. 

Warm Regards,

Rizul Khanna

System Administrator

CloudSpace Technologies Pvt. Ltd., New Delhi, India

http://www.myrealdata.com/

[hidden email]  |  <a href="tel:%2B91%209501074400" value="+919501074400" target="_blank">+91 9501074400, <a href="tel:%2B91%208595370298" value="+918595370298" target="_blank">+91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  

The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, re-transmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or Entities other than the intended recipient are prohibited. If you have received this in error, please contact the sender and delete the material from any computer. The contents of this message may contain personal views, which are not the views of  CloudSpace Technologies Pvt. Ltd., unless specifically stated.



On Wed, Apr 9, 2014 at 11:22 PM, Robert Randall <[hidden email]> wrote:

I believe that all of the applications you mention will be using a file on the file system as a data store.  They will not be using an RDBMS or similar database server.  If that is the case and I can assume that you will be using NTFS as your file system that will narrow the parameters a bit.  Please let me know if my assumptions are correct and if so I can offer some advice on what to test.  Be sure to run your tests from the guest OS.

Best regards,
Robert


On Wed, Apr 9, 2014 at 12:39 PM, Rizul khanna <[hidden email]> wrote:

Hello Robert, 

Thanks a lot for sharing interest.

We will be hosting financial applications like Quickbooks, Sage, ATX, Peachtree, Drake and many others on these Windows based VMs. These machines will be used for Remote Desktop Connections via VPN/ Secured Public IP access. 

As per our observation based on I/O consumption patterns, these Windows 2008 R2 Datacenter based Guest machines are expected to do more than 60% of Read most of the times.

Please let me know if you would need some more inputs from my side.



Warm Regards,

Rizul Khanna

System Administrator

[hidden email]  |  <a href="tel:%2B91%209501074400" value="+919501074400" target="_blank">+91 9501074400, <a href="tel:%2B91%208595370298" value="+918595370298" target="_blank">+91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  




On Wed, Apr 9, 2014 at 10:59 PM, Robert Randall <[hidden email]> wrote:
You write "Max stress".  That is a very difficult thing to understand.  What applications will you be running on your HyperV guest machines?  If I new that I might be able to suggest how to stress the machine.  Will the guests be file servers, email servers, web servers, database servers, remote desktop servers, etc?


On Wed, Apr 9, 2014 at 5:46 AM, Rizul khanna <[hidden email]> wrote:
Hello All, 

Greetings.

I have one concern. How to decide on the duration of the benchmark test while using IOmeter ? 

I am working on a New HyperV 2012 R2 Host configured with 6 Disks(4 SATA + 2 SSD) with Auto-tiering enabled between the two disk types and the volume derived out of this pool is configured with RAID 1. Thus, no striping is involved. I have been assigned to check for the Max. stress this physical host can handle. 

Can you please brief on what is the best possible roadmap for this, the host has 192 GB of RAM and 2*6 CPUs ? Our applications are Read-intensive, but no idea about the random/ sequential nature of the I/O and also, I have no idea about the I/O Size. What can be the duration of any such test ? 

Please let me know if more details would be required ? Please advise.




Warm Regards,

Rizul Khanna

r[hidden email]  |  <a href="tel:%2B91%209501074400" value="+919501074400" target="_blank">+91 9501074400, <a href="tel:%2B91%208595370298" value="+918595370298" target="_blank">+91 8595370298  |  http://www.linkedin.com/pub/rizul-khanna/39/81/a3b  |  http://virtualizationforyou.blogspot.in/  



------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user




--
Robert Randall | [hidden email]




--
Robert Randall | [hidden email]




--
Robert Randall | [hidden email]

------------------------------------------------------------------------------
Put Bad Developers to Shame
Dominate Development with Jenkins Continuous Integration
Continuously Automate Build, Test & Deployment
Start a new project now. Try Jenkins in the cloud.
http://p.sf.net/sfu/13600_Cloudbees
_______________________________________________
Iometer-user mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/iometer-user
Loading...