SMB_Cyber

I’ve worked with enough small businesses to know one thing:
Most teams are doing their best with incomplete information.

I break down IT, risk, and operations.

[SMB_Cyber @SMB_Cyber]

After working through a number of #Vmware and Hyper-V migrations to #Proxmox, one thing has become pretty clear:

The actual VM migration is usually not the hardest part.

It’s everything around it.

🧵

1/ Most environments have accumulated years of decisions that made sense at the time.

Old networking configs. Storage dependencies. Backup jobs. VMs nobody is completely sure about. Applications with requirements that aren’t documented anywhere.

That’s where most of the interesting work starts.

2/ We’ve generally had the best results by spending more time upfront than people initially expect.

Inventory the environment. Map dependencies. Understand storage and networking. Test the migration process. Have a rollback plan.

Then start moving workloads.

3/ Another thing I’ve learned: there isn’t one right way to structure these projects.

Some teams want us at Technology Architects to run the migration.

Others want us to migrate the first few workloads with them, establish the process, and then let their internal team take over.

I actually like the second model quite a bit when the customer has a capable IT team.

4/ Storage is probably the area where I see the most assumptions carried over from VMware.

Ceph, ZFS, shared storage and local storage all have their place.

But rebuilding the exact same architecture on a different hypervisor isn’t necessarily the right answer.

5/ Same with backups.

Getting the VM running on Proxmox is obviously important.

But I’d rather call the project complete when we’ve tested restores, monitoring, HA, backups and the operational processes around the environment.

6/ One pleasant surprise has been how quickly experienced VMware admins seem to get comfortable with Proxmox once they spend some time with it.

There are obviously differences, but most good infrastructure people pick it up pretty quickly.

7/ We’ve worked on everything from relatively simple clusters to multi-site environments and larger production deployments.

The technical details vary a lot, but the motivation usually sounds pretty similar:

“We want another option besides VMware, but we don’t want to create a new operational headache.”

8/ My biggest takeaway so far:

Treat a Proxmox migration as an opportunity to rethink the environment rather than just moving VMs from hypervisor A to hypervisor B.

A little extra architecture work upfront can eliminate a surprising amount of technical debt.

9/ Still learning something new on pretty much every migration.

Proxmox has its quirks, just like every platform, but I’ve been impressed with how well it fits a lot of the environments we’ve been moving.

[SMB_Cyber @SMB_Cyber]

@SHOGOx8 @Itsfoss Our team has significant experience migrating enterprise infrastructure from VMware to Proxmox. 100+ migrations this year with 100% success!

[SMB_Cyber @SMB_Cyber]

@CheyCab are you hearing very many grumblings from @Delta loyalists who are jealous of @united Starlink?

I typically take 100+ Delta flights a year and am on United from MSP to SFO and the Starlink WiFi is incredible

[SMB_Cyber @SMB_Cyber]

Technology Architects becomes the first Proxmox Certified Training Partner in the United States!

https://t.co/M1XeVHvPXu

#Proxmox #SysAdmin #Virtualization

[SMB_Cyber @SMB_Cyber]

It is absolutely imperative that you change all of your passwords, physically write them down and lock your credit report at the bureaus. Don’t wait. Do it today.

[Bloomberg @business]

The advent of Mythos and AI models like it that can exploit well-hidden flaws in software points to a faster-moving, less predictable phase of the cyber arms race. Here’s what to know: https://t.co/5uAQ1IueO0

[SMB_Cyber @SMB_Cyber]

This wasn’t just a cost-saving project.

It was a forced moment that led to a better technical outcome than where they started.

Not every platform change makes sense.
Not every “good idea” is worth pursuing.

But when a 550% increase shows up…

It’s worth asking a bigger question:
“Should we still be doing this at all?”

[SMB_Cyber @SMB_Cyber]

We had a client come to us in February with a problem that stopped everyone in their tracks.

Their VMware renewal came in at a 550% increase.

Not a typo. Not a rounding error.
A full-on budget breaker for 2026.

[SMB_Cyber @SMB_Cyber]

Post-migration results:

• Performance met or exceeded expectations
• Management was simpler than anticipated
• Costs dropped dramatically
• Future expansion became predictable again

[SMB_Cyber @SMB_Cyber]

Not every good idea is worth pursuing
Especially in AI, where the cost of getting it wrong isn’t just financial
It’s lost time, lost focus, and lost trust

The goal isn’t to do more AI
It’s to do the right AI for the right reasons

[SMB_Cyber @SMB_Cyber]

One of the hardest parts of AI consulting is sitting across from a client and saying:
“We don’t think this will produce a positive ROI”

Not because the idea is bad
But because the economics, data, or process just don’t support it

And then watching them move forward anyway

[SMB_Cyber @SMB_Cyber]

Good consulting isn’t just about building things
It’s about protecting clients from investing in the wrong things

Sometimes the highest value moment in an engagement is saying:
“We don’t recommend this”

Even when you know it might cost you revenue in the short term