I thought I would make an article about something that has been widely talked about and been the bane of just about every IT organisation’s existence at the moment.

Yes, I am going to talk about the Broadcom takeover of VMware.

I don’t think that I can remember a company takeover so big that has been handled so poorly ever in my time in global IT during my 25-year career.

It almost seems like Broadcom bought VMware and then once the sale was approved only then started to think about how the two companies would come together.

I would have thought if you spend $67 billion US on buying a company, you would at least start to have some idea of a plan before the takeover occurred.

Look—there could have been one in place but it has been extremely hard to see from the outside in partner and customer land.

I personally have been a longtime supporter of VMware and they are in my opinion by far the leading virtualisation provider in the market.

Their breadth of products and services plus the resiliency of their products are why they have been so successful on the world stage.

Now Broadcom has bought them out and it is anyone’s guess where VMware is going to end up.

Someone in my network called Broadcom the ‘killers of dying tech’ when the VMware takeover happened.

VMware is anything but dying tech and given the market share they have, they won’t die anytime soon.

However, if the current direction from Broadcom continues coupled with their very poor communication to their global partners and customers, they are certainly going to take a big hit.

I have seen many a post from Nutanix employees getting stuck into VMware as well as hyperscaler staff getting excited about the potential migrations to public cloud services.

Let’s also not forget the many small and niche virtualisation players are now smelling blood and looking to gain market share.

However, as a lot of us know, while hypervisor migrations can occur, it is also not an easy thing to do.

On top of that, it is not just a case of migration from one platform to the other.

There are plenty of other considerations you need to think about before moving.

Let’s break this down.

First, let’s talk about cost because this is the first and foremost pain point just about every customer has with Broadcom at the moment.

VMware licenses are no longer perpetual and are now subscription based.

A lot of customers have been screaming about this but if we are being honest, this is the way just about most major software vendors in the world are heading.

This is nothing new and even if Broadcom didn’t buy VMware, this probably would have happened anyway.

Also affecting costs has been Broadcom making the subscriptions core based around 16 cores per subscription unit per ESXi host compared to the legacy per socket model by VMware.

Now on the surface, this does seem like customers are getting shafted big time but let’s not forget, while VMware called their vSphere licensing per socket, each socket license was up to 32 cores.

You could also call this a core based license based around 32 cores per subscription unit.

Yes, the Broadcom change means more subscription units need to be accounted for to cover ESXi fleets but when you break it down like that, it doesn’t sound so bad.

There is also the new four-tier subscription model Broadcom have created for vSphere—Essentials, Standard, Foundation, and VMware Cloud Foundation.

What this has done is group most of VMware’s old products together into these four tiers and make them available to customers depending on the tiers.

This has led to calls of customers paying for ‘shelfware’ because they don’t use those features.

This is a very fair point but could we put another spin on this for customers and look at these features—could they be licensed for now and could they make use of them?

I am not going to go into the specifics of the four tiers but I suggest to customers once they are told what new subs tier they are on at renewal times, look at the functionality now available to you and see if you can use it.

There still may be ‘shelfware’ customers will be stuck with but you never know, there may be VMware products you could use that will benefit your VMware environment that you never considered before.

I call this making the best out of a bad situation.

However, going to these new tiers has come at a cost increase—a substantial cost increase.

So substantial it does give the impression of gouging customers and, if I am being honest, it really is.

However, I would like to remind everyone of another vendor that made substantial cost increases when they changed from a per socket to a per core licensing cost model for their database engine.

Yes, I am talking about Microsoft when they decided to drive the knife into customers who used SQL Server.

I had customers who felt serious pain with this and like what we are hearing with VMware licensing / support renewals now, I had customers have cost increases of up to ten times with this.

Like Broadcom, Microsoft were very unapologetic about this and just said ‘pay up’.

I pose a question here—did customers move away in their droves from Microsoft SQL Server?

Not really—why?

Because a lot of applications are tied to using Microsoft SQL Server DB backends and getting an application to use the same data in a different database engine (if even possible) is a very hard thing to do!

So what happened?

Customers shouted, vented, stomped their feet—but paid up.

It seems everyone has forgotten this and this wasn’t even a corporate take over—this was simply a vendor wanting to make a lot more money from an existing product knowing customers would have no choice but to pay up.

Similar case here—yes, it sucks, but Broadcom are not the first company to do this and they won’t be the last.

Don’t get me started on Oracle …

Let’s now consider you are a customer, you have looked at the new subscription tier you have been told to move to, see very marginal benefit from the extra features in vSphere you ‘could’ use and now want to do market assessment.

Yes, there are cheaper options out there but the first thing to consider is what the feature set is.

Is it as feature rich as the current VMware stack you are using at the moment?

Also, how similar is the new virtualisation platform to VMware?

What will be the effort to upskill your technical team?

To be honest—any decent, competent, and knowledgeable technical team should be able to pick up another hypervisor provider but this is still a consideration.

Next item of consideration—does your hardware vendor certify this hypervisor?

This is something a lot of the niche virtualisation players are going to have to work through and fast if they really want to grab market share.

Some engineers will say a lot of the niche players use a KVM Linux based hypervisor and all hardware vendors support KVM so it shouldn’t be a problem?

Right? Sorry, wrong.

It is not a case of whether it will technically work—you would think anything based on KVM should work.

But working with your hardware and whether your hardware vendor certifies this hypervisor and more importantly SUPPORTS it is another question.

I ask all of you—do you really want to be in a position where the hardware vendor does not support your hypervisor?

What if you face issues once the new hypervisor is installed and it is looking like some sort of interaction with the hardware?

The hardware vendor won’t help or support you.

What does this also mean for drivers and firmware? Updates to drivers and firmware? Will they work?

There is a lot of risk right there.

Onto our next consideration—does your current backup product support and certify your potentially new hypervisor?

Again, the same rationalisation can be used that we talked about above—well, if it works for KVM, then it ‘should’ work for this new hypervisor.

Maybe, but what if it doesn’t?

Can you go to the backup vendor then for help? No, you can’t.

Backups are a company’s last line of defence for cyber breaches, accidental deletions, and so on.

Do we really want to put a platform that is so critical to your business in a potentially unsupported state?

I know I wouldn’t.

Now, I stress this is NOT an article supporting the Broadcom direction to date—it is certainly not!!

Here at my company I own, we are just as angry as the next partner out there and like all of you we are trying to navigate this mess as best we can.

I also want to stress the VMware staff who transitioned to Broadcom who didn’t lose their jobs are doing the best they can with what they know.

Just remember, they too are trying to figure out what this all means and people you have been dealing with happily for years don’t become your enemies overnight.

However, I won’t make any apologies for Broadcom at all nor should anyone.

I think we as a global IT community need to keep at them and see what can possibly be salvaged from what has been and what still could be a great virtualisation vendor.

I am not encouraging you to stay with VMware nor am I trying to discourage you from looking at alternatives.

Given the uncertainty Broadcom have created, if you are in customer land, you would be doing your employer an injustice if you didn’t.

However, what I AM trying to do for anyone reading this is to take the emotion out of the situation and try to look at the facts at hand from a different perspective.

If we do this and try to consider this problem analytically and pragmatically considering the above, then maybe some of us will make the best we can out of a bad situation.

This is just my humble opinion from years in the field of technical roles but I wanted to put this out there to try to help people make some sense out of the carnage that has been created.