When customers build new data centers or private clouds, they are looking for a standardized blueprint that can be automated and is considered to be intrinsically secure. They are thinking like cloud service providers (CSP) and would like to deliver a private cloud with public cloud characteristics, where they can build their services (IaaS, CaaS, PaaS, DBaaS etc.) on top. If it is a private cloud stack that I have to take care of, I would like to have automation and built-in security as a design requirement, and it should be something that can be managed efficiently.

Monolithic Private Clouds and Modern Applications

It is very interesting to see that so many organizations talk about modern applications, but are still managing and maintaining what I would call a “monolithic” data center. I see customers talking about a modern infrastructure for their modern (or to be modernized) applications. Modern infrastructure means a public cloud for them.

The word “monolithic” describes that something is very large, united, and difficult to change. Something inflexible. Talking about monoliths, most people immediately think about static, solid, and big applications that need to be modernized to become smaller loosely coupled entities. Therefore, it surprises me that almost nobody talks about monolithic infrastructures or monolithic private clouds. Perhaps this has something to do with the mostly (still) monolithic applications which implies that these workloads are running on a legacy or monolithic infrastructure. That could be the mindset or illusion of enterprises that are starting with the modernization of the low-hanging fruits that can run in the public cloud.

Migration of monolithic applications to modern public clouds

There are so many cases, where organizations have no other choice than to lift and shift virtual machines from their data centers to Azure (aka modern cloud) let’s say, because of their contracts – because of their commitments. No innovation and the same business logic but for 10x the price.

We have seen it more and more over the past few months and years: It is not that easy to move workloads to the public clouds. In most cases, it takes longer than expected and organizations learn from other organizations, which allows them to adjust their plans and “journey to cloud” timeline.

So, what happens to the applications that have to stay in your private cloud, because you cannot or do not want to migrate them to the public cloud (or any other cloud in general)? Some of the applications are for sure still important, need to be lifecycled and patched, and some of them need to be modernized for you to stay competitive with the market and competitors.

What about a modern private cloud?

If you have the same vision and approach in mind, which is putting modern applications on a modern platform, what are the reasons for stopping and not investing in a more modern platform that can host your legacy apps, modern apps, and anything that might come in the future? Where do you deploy your AI-based workloads and data services if such applications/workloads and their data have to stay in your private cloud?

Cloud is an operating model, not an architecture or specific place.

What is the reason that you treat your on-premises cloud differently?

Changes at VMware by Broadcom

I hear and see comments about:

Yes, some of the changes are disruptive. Nevertheless, I believe it is a big step in the right direction.

No company or person in this world is perfect. There is always something or someone that you and I do not like about another person or a company. And that is okay!

Let us take the example of pricing.

If you do not get what you want, you feel frustrated. You feel let down and disappointed. In other words, if you get what you do not want, you are unhappy.

If you (exactly) get what you want, which would be a fair price from your standpoint, you still would be unhappy, because you can’t hold on to it forever.

What are you going to do about it? No vendor on this planet can and will give customers guarantees about future price developments. 

Strategy and Misadventures

I had a chat with the CTO of one of the biggest banks in the world back in December 2022. We were talking about regulations, public cloud concentration risk, application portability, and their defined cloud-exit triggers.

One of their metrics was about “cost increase“:

20% or greater price increase of a cloud provider’s service used by a production application over two consecutive quarters.

In 2023, Microsoft announced a Microsoft Cloud price increase of 9-15% in Europe:

taking into consideration currency fluctuations relative to the USD. […] and move to a pricing model that is most common in our industry.

The Microsoft Cloud continues to be priced competitively, and Microsoft remains deeply committed to the success of its customers and partners. We will continue to invest to enable customers to innovate, consolidate and eliminate operating costs, optimize business performance and efficiency and provide the foundation for a strong security strategy that customers around the world have come to rely on.

Note: This price increase affected services that charge based on usage, such as compute, storage, and networking. Every customer including those with a commitment and a discount in place had to pay the price increase.

What did this customer and many others do? They did not think about a cloud-exit trigger but were finding a way to offset these cost increases.

I think most of us are looking for an approach to put the right app in the right cloud based on the right reasons while not losing track of costs. And then we have the need to maximize performance and business benefits. Oh, and we have to consider regulations and (data) privacy as well.

What to expect from VMware by Broadcom?

Building or modernizing data centers or clouds is about automation and security, and speed can be considered the new security nowadays. Yes, the innovation engine was stalling a bit during the Broadcom acquisition.

But here is what is happening. Let us say that building a data center or in this case, a private cloud is like setting up a tent when camping:

  • You have to select a location to accommodate the size of your tent.
  • Lay out the fabric and sort through all the components to ensure everything is present and in good condition
  • If it is a traditional tent with poles, you need to assemble them according to the instructions provided
  • Then you have to place the tent fabric over the assembled poles and ensure the base is properly secured to the ground
  • After that, you secure the tent by staking down the corners and guy lines to add stability in windy conditions
  • Now that you know that the tent is stable and properly secured, it is time for final adjustments
  • Everyone is happy, we can work on the interior setup and set up other camping gear inside as needed

There are two options to speed up and enhance this process:

  1. Ask for help. Someone with experience who has done this before.
  2. Buy an inflatable tent

The new VMware Cloud Foundation

Picture a self-inflating tent, where every component plays a crucial role in creating a seamless structure. Just like how the framework, fabric, and guy lines work together to make the perfect tent, in a data center, the compute, storage, and network components form a stack, each enhancing the performance, security, and connectivity of the system.

Overall, self-inflating tents offer a convenient, practical, and comfortable camping solution for adventurers of all levels.

This is what we can expect from the VCF division and from Broadcom’s CEO Hock Tan, who has promised “to invest an incremental $2 billion a year to better unlock customer value – with half focused on R&D and the other half focused on helping to accelerate the deployment of VMware solutions through VMware and partner professional services.” 

With the new VMware Cloud Foundation division under Broadcom, we see a strategic consolidation, where diverse expertise converges into a single force working on this self-inflating tent. By integrating the formerly disparate units, customers are going to profit from a better user experience and an enhanced set of products and services.