Solved

Testing Nutanix performances with IOMETER

  • 16 January 2017
  • 6 replies
  • 10348 views

Badge +10
Hi,As an introduction, please let me explain why I'm testing My blocs with IOMETER.

Context Of Testing :
Worldwide company. A corp Archtect team that is creating a private cloud product w/ VMware cloud suite and Nutanix / DELL bloc to support them. A 3 year planning w/ 20+ cluster deployment 'til 2018.

Goals :
Giving a way of testing :
  • performance at deployment.
  • Performance w/ NOS / ESX /Other products changes.
It's not a way to qualify a technology against another but giving a rapid way of testing performances.

Exec :
on a 10 nodes XC630 cluster, using DEREK's and other internet resources, executing some performances testing :
  • 8 workers + 8 unformatted drives by master,
  • 1 master by node,
  • ESX 6.02 / NOS 4.6.4 and more,
  • 8 vCPU + 32 Go RAM on CVM,
  • RF3 + dedup and compression inline (real configuration).
  • 4k Random read,
  • 16k 70%random 70%Read
Results :
4k random read, since it targets only RAM cache on CVM is really handsome w/ 185k IOPS by XC630 at 16 outstanding IOs. Scaleout is achieved w/ 1,850 M IOPS on 10 blocs
70% read / 70% random and 16k IO. Totally irelevant w/ a lot of errors / no throughput.

What am I doing wrong ? Do I need to format HDD for Write IOs ?thanks for your help,Thomas CHARLES
icon

Best answer by Jon 17 January 2017, 03:39

NewVirtTom - Thanks for reaching out. Pretty cool that you hit almost 2M IOPS in 10 nodes, great stuff.

Two suggestions, in general:

1) If you're performance testing pre-production, you might as well get to a newer NOS version that 4.6.4.

5.0 is out, available on portal.nutanix.com, and has quite a few enhancements compared to 4.6.4 across the board (lots of new features, as well as a few performance enhancements). If you dont want to go right to the latest release (i.e 5.0), 4.7.3.x has been out for a while.

Dont think that will fix this specific issue as gut feeling is that its an IOmeter setup issue, but its a good idea nonetheless

2) Check out the newer blog on IOMeter + Nutanix here: https://next.nutanix.com/t5/Nutanix-Connect-Blog/Configuring-Iometer-Performance-Tests-on-Nutanix-the-Right-Way/ba-p/13255


Now, with respect to your IOMeter write errors, when I've seen that in the past, thats because something was wrong with the disks IO meter is writing too. Usually that means they are formatted or otherwise in use, which is strange since you noted they were actually unformatted.


Also, do you see this error on ALL the IOMeter instances? or just some? Did you set them all up individually? or from some cloned template?
View original

6 replies

Userlevel 7
Badge +30
NewVirtTom - Thanks for reaching out. Pretty cool that you hit almost 2M IOPS in 10 nodes, great stuff.

Two suggestions, in general:

1) If you're performance testing pre-production, you might as well get to a newer NOS version that 4.6.4.

5.0 is out, available on portal.nutanix.com, and has quite a few enhancements compared to 4.6.4 across the board (lots of new features, as well as a few performance enhancements). If you dont want to go right to the latest release (i.e 5.0), 4.7.3.x has been out for a while.

Dont think that will fix this specific issue as gut feeling is that its an IOmeter setup issue, but its a good idea nonetheless

2) Check out the newer blog on IOMeter + Nutanix here: https://next.nutanix.com/t5/Nutanix-Connect-Blog/Configuring-Iometer-Performance-Tests-on-Nutanix-the-Right-Way/ba-p/13255


Now, with respect to your IOMeter write errors, when I've seen that in the past, thats because something was wrong with the disks IO meter is writing too. Usually that means they are formatted or otherwise in use, which is strange since you noted they were actually unformatted.


Also, do you see this error on ALL the IOMeter instances? or just some? Did you set them all up individually? or from some cloned template?
Badge +10
Hi Jon,

THX for your answer.

upgrading t onewer version is definitly on roadmap ! As I said since my client has quite a legacy way of managing HW / infrastructure life cycle, I know that upgrading will not be an easy thing. I expect a lot of resistancy from admin team / architecting team and management (yes they do lead change management).

My intention here is to gather a lot of materials to prove that this kind of change change is :
  • easy,
  • predictable,
  • game breaker (giving more efficiency and security).
so giving them a baseline for what they bought and what they will have if they change (with the procedure written in their forms) is a way to achieve that ! well at least at my level of decision taking ^_^

So first of all OOB baseline
After 4.7.x baseline
And at least 5.x baseline !!!

I'll check your questions on IO meter to see what's going on and answer later.

THX !

Thomas
Badge +10
  • Check out the newer blog on IOMeter + Nutanix here
That's already done ! as a formal scientist i begin my working set by reading a lot ;)

  • when I've seen that in the past, thats because something was wrong with the disks IO meter is writing too
interesting ! since that was working perfectly on read I didn't suspect this...

  • Also, do you see this error on ALL the IOMeter instances?
yep ! same error...
Userlevel 7
Badge +30
hey NewVirtTom - Sorry, parsing through some old threads and realized I never responded to you here.

Its been a little while, have you made any progress on this?

Jon
Badge +10
Hi Jon,

Time's never enough ! sure i made a lot of progress, my protocol is on heavy testing before giving it as a tool for my client.

We will migrate to AOS 5 soon and after I will create a blogpost series on this,

keep in touch,

Thomas
Userlevel 7
Badge +30
ok NewVirtTom - Thanks.

Feel free to ping me at jon at nutanix dot com, happy to proof read your blog series before hand.

Cheers,
Jon

Reply