In our previous Pulses, we’ve followed a familiar thread, from the false choice between hybrid and multi-cloud, to the OpenStack vs VMware dilemma, to the search for an abstraction layer beyond Kubernetes and PaaS. 
And through it all, we’ve told Marco’s story: hidden costs, growing complexity, and a Total Cost of Ownership that never quite added up. 

Last week, Marco called me. 
"I’ve understood what you meant," he said. "I’ve already moved a few workloads to a managed environment. It works better, and life’s easier - fewer incidents, more efficiency, more time for the business. But now I have a bigger problem." 

That “bigger problem” is the same one facing most IT leaders in 2025: 

How can you simplify operations without giving up control? 

----- 

The Starting Point: A Fragmented Infrastructure 

Marco has already made real progress. 
He migrated several critical applications to a managed solution and he saw immediate benefits: lower TCO, 40% fewer operational incidents, and a team now focused on higher-value projects instead of nightly patching and troubleshooting. 

But his reality remains fragmented. 
Most workloads still live across a hybrid patchwork: legacy VMware, private clouds, isolated VMs across multiple hyperscalers, and manually orchestrated containers. 

"The thing is," Marco told me, "I’d love to move everything to a single managed model — I’ve seen how simplification and automation change everything. But I can’t lose control of my infrastructure, and I have to keep an eye on costs. I can’t move everything to hyperscalers. Some applications need to stay sovereign — close, compliant, and fully under our governance." 

—-

The Great Dilemma 

Marco’s recurring question (and probably yours too) is inevitable: 
Where and how should I migrate my dispersed workloads - and how can I control such a mixed environment? 

He wants: 

• The operational simplicity of a managed solution. 

• The full control of a sovereign infrastructure. 

• The flexibility to switch providers without lock-in. 

• The ability to govern infrastructure and applications from a single point. 

The problem is: these needs have always been in conflict. 

—-

The Options on the Table (and Their Limits) 

Marco spent weeks comparing options, consulting peers, and building evaluation spreadsheets. Here’s what he found: 

VMware: the “safe choice” that’s no longer safe 

• Pros: Proven reliability, mature ecosystem, widely available expertise. 

• Cons: Exploding costs (+300–500% post-Broadcom), growing lock-in, rigid contracts. 
  "I can’t afford it anymore — but migrating everything away is a nightmare." 

  
  

OpenStack: the road to sovereignty 

• Pros: Full control, no vendor lock-in, zero licensing costs. 

• Cons: High operational complexity, requires in-house expertise, steep learning curve. 
  "It’s perfect for me — if only I had the people to run it." 

Hyperscalers (AWS / Azure / GCP): power at a price 

• Pros: Unlimited scalability, advanced services, massive ecosystem. 

• Cons: Architectural lock-in, unpredictable costs, distributed governance. 
  "Great for new projects — but how do I bring my legacy workloads without tearing everything down?" 

Proxmox: simplicity that doesn’t scale 

• Pros: Easy to set up, cost-effective, low learning curve. 

• Cons: Limited enterprise adoption, fewer managed services (no DBaaS, LBaaS), less mature APIs for complex automation. 
  "Solid for traditional setups — but not built for cloud-native ecosystems." 

——-

The Impossible Matrix 

Marco built this chart to visualize his dilemma: 

Looking at this table, the pattern is obvious: every option has a red flag somewhere. 
There’s no perfect solution. 

But Marco sketched an even clearer version of his dilemma: 

The top-right quadrant — high sovereignty + high management capability — is where the ideal solution should live.  

It seems impossible, but in reality, some paths point in that direction: enterprise-grade Cloud Management Platforms, customized platform engineering stacks, or complex combinations of specialized tools. 

The catch?

They usually require significant investments - in licenses, specialized skills, and dedicated teams - resources that many mid-sized companies like Marco’s can’t easily afford. 

—-

The Hidden Frustration 

"What really frustrates me," Marco admitted, "I know that enterprise CMPs can technically do what I want — but they’re too complex and too expensive for us. I just need something simpler and more accessible, something that lets me unify IaaS and PaaS. I want to say, ‘I need this app to run with these requirements,’ without worrying if it’s on OpenStack, AWS, or anything else." 

And that’s the crux of it. 
The issue isn’t just where to deploy workloads — it’s how to manage them cohesively once they’re running. 

—-

The Missing Links 

What Marco — and the market — are missing are the connectors between worlds: 

• Between infrastructure governance and application governance.

• Between legacy VM workloads and containerized applications. 

• Between ease of use and technical control. 

• Between centralized management and geographic distribution. 

• Between advanced automation and operational transparency.  

—-

The Question That Changes Everything 

Marco had nailed the real problem: unified management. 
But his story raised an even deeper question. 

After a while, I asked him: 
"What if the real problem isn’t what or how to manage — but how do you interact with all this complexity?" 

"What do you mean?" He asked. 

"Think about it. Those enterprise CMPs you mentioned deliver what you need — but they still force you to think in their terms: complex dashboards, layered configurations, predefined workflows. 
What if you could just describe what you want to achieve, without learning yet another technical language? What if there was a way to get OpenStack’s sovereignty, a hyperscaler’s simplicity, and the control of a private infrastructure — without having to choose, and without becoming an expert in each?" 

Marco paused for a few seconds. 
"If that really existed, most of my problems would solve them. But does it?" 

—-

The Answer? In the Next Chapter 

In our next Pulse, we’ll explore how some companies are already solving this exact dilemma. 
Spoiler: the answer isn’t a new technology, but a new way of thinking about the interface between people and infrastructure. 

The future of the cloud won’t be measured in clicks or configurations — 
but in conversations. 

The next frontier isn’t management. 
It’s dialogue. 

—-

And you? Do you recognize yourself in Marco’s dilemma? 
How much of your team’s time is spent managing infrastructure instead of creating value? 
In the next Pulse, we’ll see how some organizations are already embracing infrastructure that responds to intent — not just commands.