@Oksana but... why would anyone do this? lol

@Oksana Thank you for the article. Hopefully i am able to deploy this soon.

@Oksana said in VMware DRS: Smarter Resource Management for vSphere:

SM_VMware Distributed Resource Scheduler (DRS)@2x.png

Managing VMware clusters by hand can quickly become overwhelming as workloads grow and shift. That’s where Distributed Resource Scheduler (DRS) steps in – keeping clusters stable, apps fast, and admins stress-free.
In our latest guide by Dmytro Malynka for StarWind, we break down everything you need to know about VMware DRS: how it works, its requirements, key features, and why it’s a must-have for medium and large clusters. Read more here: https://starwind.com/s/wg

DRS sounds like a great thing on the surface before actually implementing it. In large clusters, VMs with heavy workloads get moved between hosts often with the associated disruptions when memory gets locked. Takes what is already a poor situation and makes it worse!

I formerly worked at an employer that had a very large VMWare cluster implemented with vSphere, vCloud and DRS enabled. They were up to 50 hosts when I left, and constantly ran into issues because of DRS.

@scottalanmiller Thanks for the pushback, Scott, all fair points!

The article ended up blending a few different concepts: remote/published app delivery, application virtualization, full virtual desktops (VDI), and even the underlying server virtualization and containerization layers.

What we actually described mostly fits remote app delivery (X11, Windows RDS/RemoteApp, Citrix/AVD): the app runs on a server and only the UI is sent to the client.

That’s different from application virtualization (MSIX app attach, ThinApp), where the app runs on the endpoint inside an isolated package with little or no traditional install. We’ll add a separate diagram to make that clear.

We’ll rework the article to separate these models properly, add examples, and fix the terminology.

Really appreciate the correction!