We are facing poor I/O performance using MCS-based VDAs.
The situation is the following : we are hosting a private cloud where each customer has its own infrastructure.
Some customers are using persistent VDAs, others are using MCS.
Our infrastructure is based on 2 datacenters (4 ESXi hosts and 1 SAN each).
Customers using MCS-based machines are complaining about poor performance.
Indeed, we noticed a huge I/O difference using CrystalDiskMark
The conclusion could be easy : why don't you turn on MCSIO for all MCS-based VDAs, as the IO performances are better ? Because we don't see a noticeable impact in practice. It consumes disk space on the datastore with no visible improvement.
Do you have any advice on why the performance is so bad when using MCS vs persitant machines ? Are we missing something ?
Question
thomsi
Hello,
We are facing poor I/O performance using MCS-based VDAs.
The situation is the following : we are hosting a private cloud where each customer has its own infrastructure.
Some customers are using persistent VDAs, others are using MCS.
Our infrastructure is based on 2 datacenters (4 ESXi hosts and 1 SAN each).
Customers using MCS-based machines are complaining about poor performance.
Indeed, we noticed a huge I/O difference using CrystalDiskMark
The conclusion could be easy : why don't you turn on MCSIO for all MCS-based VDAs, as the IO performances are better ? Because we don't see a noticeable impact in practice. It consumes disk space on the datastore with no visible improvement.
Do you have any advice on why the performance is so bad when using MCS vs persitant machines ? Are we missing something ?
Thanks for your help.
0 answers to this question
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now