Solved

ESX Server Disk Performance - Interpreting the data

Posted on 2009-03-30
2
3,877 Views
Last Modified: 2013-11-14
I am trying to interpret storage utilization data from our ESX servers but I am not sure what
counters to look at, and when I get that data determining exactly what it means.  

Currently I am going through Virtual Center and looking at the performance tab.  I have Disk Usage (Avg)  pulled up and it shows a maximum of 9,279 KBps with an average of 2,022KBps.  I truthfully dont know if those are high numbers or not, or if ESX storage counters need to be interpreted with a formula based on hardware.
This server is a local storage server with (6) 15,000RPM SAS drives.  

What data should I be looking at for the counters to determine a disk io problem, and at what point in time should I be concerned that I have overloaded the server and need to migrate to another storage device or server ?

Attached: Stacked Graph of  Disk usage (average) over the past week.
vmware-disk-utilization.JPG
0
Comment
Question by:heliontech
2 Comments
 
LVL 30

Accepted Solution

by:
Duncan Meyers earned 250 total points
ID: 24025501
Broadly speaking, disk bandwidth is a spectacularly useless metric to look at in a VMware environment. Disk bandwidth is high where you have a sequential I/O load, and low where the workload is largely random in nature. ESX Server, by its nature, presents a highly random workload to disk, so take a look at Disk Read Requests and Disk Write Requests. The total of Read + Write shouldn't exceed about 1200 - 2400 (give or take, depending on RAID config etc etc etc).

An important thing to monitor is response time, which should be less than 20ms at worst, less than 10ms in your config. High response times indicate either a heavy workload or a configuration issue.
0
 
LVL 18

Assisted Solution

by:larstr
larstr earned 250 total points
ID: 24028607
There are several factors that will impact your disk performance such as your disk controller type, raid level, the amount of ram on your controller and whether it has a onboard battery or not. With directly attached storage (das) you should normally not have too high latency (response time), but it also depends on how powerfull your disk controller is.

For measuring latency you can use esxtop from the esx console.

A better way of measuring the performance of your disk system is to use iometer and compare your results with what other people have gotten. I did some testing here after following this thread.

Good luck!

Lars
0

Featured Post

Free Tool: Port Scanner

Check which ports are open to the outside world. Helps make sure that your firewall rules are working as intended.

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

If we need to check who deleted a Virtual Machine from our vCenter. Looking this task in logs can be painful and spend lot of time, so the best way to check this is in the vCenter DB. Just connect to vCenter DB(default DB should be VCDB and using…
Many businesses neglect disaster recovery and treat it as an after-thought. I can tell you first hand that data will be lost, hard drives die, servers will be hacked, and careless (or malicious) employees can ruin your data.
Teach the user how to edit .vmx files to add advanced configuration options Open vSphere Web Client: Edit Settings for a VM: Choose VM Options -> Advanced: Add Configuration Parameters:
Teach the user how to use create log bundles for vCenter Server or ESXi hosts Open vSphere Web Client: Generate vCenter Server and ESXi host log bundle:  Open vCenter Server Appliance Web Management interface and generate log bundle: Open vCenter Se…

820 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