Solved

Determine IOPS

Posted on 2014-09-24
7
155 Views
Last Modified: 2014-10-04
My boss asked that I determine the current IOPS on a server and recommend alternative solution.

He doesn't need specific metrics just a general idea of current iops and what new server should be able to handle the current needs.

From what I am reading I keep getting these extremely detailed breakdowns of number. I have used perfmon to collect the metrics per a website which  I thought could give me this information. I am now staring at the numbers but have no idea what to do with them. I assumed there was a general formula I could use based on my collections but I don't know what to do. Any suggestions would be help. Please see attached file of what I have and if anything else if need. Again I just need a general idea of the IOPS as this is not a mission critical server. Basic understandable explanation would be great. Thanks.
Perfmon.jpg
0
Comment
Question by:Tim OBrien
[X]
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
  • 4
  • 3
7 Comments
 
LVL 2

Expert Comment

by:Calvin Paxson
ID: 40343010
From a cursory glance it looks like you are using 57 IOPS.  From wikipedia:

7,200 rpm SATA drives      HDD      ~75-100 IOPS[2]      SATA 3 Gbit/s      
10,000 rpm SATA drives      HDD      ~125-150 IOPS[2]      SATA 3 Gbit/s      
10,000 rpm SAS drives      HDD      ~140 IOPS[2]      SAS      
15,000 rpm SAS drives      HDD      ~175-210 IOPS[2]      SAS      

Realistically you want to get an average over time:

http://blog.synology.com/blog/?p=2086 

and then use the above chart to determine how many disks you need at max.  There is a penalty for IOPS when you are using RAID.
0
 

Author Comment

by:Tim OBrien
ID: 40343051
Could you let me know which metrics you referred to when determining that number? I am fairly certain the metrics I provided were an average over 1 day which would be enough time to get good numbers.

Reading the documentation of what configuration I may need for the new server was more understandable. My main issue was filtering thru the values I provided and determining current IOPS. Thanks for your initial help!
0
 
LVL 2

Expert Comment

by:Calvin Paxson
ID: 40343061
DiskWrites/sec
DiskReads/sec
DiskTransfers/sec

Those should be the values you are interested in.  My math was bad but i'm sure you've already corrected it!
0
Comparison of Amazon Drive, Google Drive, OneDrive

What is Best for Backup: Amazon Drive, Google Drive or MS OneDrive? In this free whitepaper we look at their performance, pricing, and platform availability to help you decide which cloud drive is right for your situation. Download and read the results of our testing for free!

 

Author Comment

by:Tim OBrien
ID: 40343080
I don't expect you to do all the work. I will review those numbers and hopefully figure it out. Thanks so much!

I will review the numbers and mark it as a solution shortly, don't know if it closes out question  if I mark as solution now so will hold off.
0
 

Author Comment

by:Tim OBrien
ID: 40344825
Ok I added up the numbers and came to about 105 IOPS
DiskReads/sec - 4.6
DiskTrasnfers/sec  - 57.6
DiskWrites/sec - 53

105 IOPS is the limit this server can handle, from what I understand this doesn't indicate what the the new server should be planned for as this 105 IOPS is again the current limit of the server.

I read that referring to the Current Disk Queue Length will provide an indication to how much IOPS I should plan for so somehow the Total IOPS of 105 and the Current Disk Queue Length Metric of 8.9 will indicate how many IOPS I should plan for but I don't understand how the Current Disk Queue Length should be applied when forecasting.
0
 
LVL 2

Accepted Solution

by:
Calvin Paxson earned 500 total points
ID: 40350299
Current disk queue length should be 2 plus 1 for each additional spindle in the machine.  If this machine has only one spindle then it is definitely over worked.  As for forcasting what is the workload?  SQL?  Exchange?  Other?  The application vendor should provide some sizing guidelines so that an informed decision can be made.
0
 

Author Closing Comment

by:Tim OBrien
ID: 40361680
Thanks for your feedback.
0

Featured Post

Free Tool: IP Lookup

Get more info about an IP address or domain name, such as organization, abuse contacts and geolocation.

One of a set of tools we are providing to everyone as a way of saying thank you for being a part of the community.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

Network ports are the threads that hold network communication together. They are an essential part of networking that can be easily ignore or misunderstood, my goals is to show those who don't have a strong network foundation how network ports opera…
PRTG Network Monitor lets you monitor your bandwidth usage, so you know who is using up your bandwidth, and what they're using it for.
This video gives you a great overview about bandwidth monitoring with SNMP and WMI with our network monitoring solution PRTG Network Monitor (https://www.paessler.com/prtg). If you're looking for how to monitor bandwidth using netflow or packet s…
There's a multitude of different network monitoring solutions out there, and you're probably wondering what makes NetCrunch so special. It's completely agentless, but does let you create an agent, if you desire. It offers powerful scalability …

688 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question