Updated IO Optimization with Machine Creation Services

With the Citrix Virtual Apps and Desktop 1903 release, Machine Creation Services continues to improve!

The RAM-based write cache feature has been updated and improved. I wanted to see how the new RAM-based write cache capability compared to the previous version and how it compares to a desktop without using a RAM-based write cache.

First, let’s look at the 95th percentile IOPS with a LoginVSI knowledge worker workload running for one hour with different sizes for the RAM cache


Seeing a 15-20% reduction in storage IOPS from the 1811 RAM cache and a 50%+ reduction compared to using no RAM cache.

Let’s now look at the % Disk Time (how busy is the disk)


A 40-60% reduction in total disk activity time when we compare to the previous version. In 1903, we reduce disk write time to under 1% with 256MB write cache.  It took 512MB in the 1811 version to achieve a similar performance.

Daniel (Follow on Twitter @djfeller)
Citrix Workspace Poster
XenApp/XenDesktop On-Prem Poster
XenApp/XenDesktop Cloud Service Poster


7 thoughts on “Updated IO Optimization with Machine Creation Services”

    1. There’s a flow chart that shows when PVS is applicable vs MCS. IMO, I would only be doing PVS if I had physical endpoints I had to stream to, but then again I’m only running under 100 desktops. If you are in the thousands then PVS probably makes sense. As for comparison, PVS also has a similar feature called ram cache overflow


  1. Has anyone had experience running more than 512mb of Cache ram. We are running 2048MB but are experiencing memory exhaustion issues. Although i dont think its the cache rams issue we do believe there is a memory leak at a driver level somewhere. Anyone see any reason why 2048MB would be an issue. our Vm’s have 16GB ram assigned to all.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.