From On-Prem to Cloud: Expert Perspectives on Modern IT Solutions

A person standing indoors, wearing a light blue cable-knit sweater over a white collared shirt. The background shows an office environment with blurred details, including desks, chairs, and plants. The individual has their hands clasped in front of them and is positioned slightly to the left of the center of the image.
Discover expert insights on cloud migration, global infrastructure, and the future of cloud computing in our latest interview with Johan van der Merwe, Senior Cloud Solution Architect Manager at Microsoft. Learn from a leader with over 20 years of IT experience.

Introduction

We are delighted to introduce Johan Van der Merwe, a Senior Cloud Solution Architect Manager at Microsoft with over two decades of extensive experience in Information Technology. Johan has deep technical expertise in Azure and a broad perspective across multiple countries. He manages several Cloud Solution Architects across Austria, Finland, Ireland, and Portugal. Born in South Africa and now based in Ireland, Johan brings valuable insights into how different cultures approach cloud adoption. In this interview, we delve into the cloud computing evolution, exploring expert perspectives on the shift from on-prem to cloud solutions. Johan shares his thoughts on cloud computing adoption, cultural differences, and the future of cloud technology. Let’s start by learning more about Johan and his role at Microsoft.

Interview

Introduction to the Interviewee

Azure Navigator

Johan, could you please introduce yourself and tell us a bit about your background, your current role at Microsoft, and what motivates you in your work?

Johan

Thanks, I started in the IT industry almost by accident. Like a lot of 18-year-olds, I started off on a path in university to study what I thought was a good area and career. I quickly figured out that I really did not know exactly what I wanted, but it was not this course. Where I grew up, parents are quite strict, so dropping out led to an ultimatum: if you’re not studying, then you are paying rent. This proved to be invaluable, as during the next two years, I tried a few things and met a lot of interesting people. Ultimately, I got involved in IT by being a tester—sitting down, doing weird things in the program, and writing down what I did. Then I helped write a manual for the program, learned how to deploy it in an enterprise setting, and trained about 140 people on how to use it and train others. After that, I was offered a permanent position, which I accepted. I then went through several delivery and management roles across a myriad of Microsoft technologies and worked with hundreds of different clients. In 2016, I applied for the role of a Premier Field Engineer (PFE) for Azure infrastructure at Microsoft, which I really enjoyed for the next five years. In 2021, the opportunity arose to manage the same team, and I applied for that. It’s quite daunting to manage the people who were your peers just five minutes ago. But over the last three years, we have grown as a team. I’d say my motivation really is to be of service. I like nothing more than solving issues or requests and the satisfaction of helping someone achieve their goals. Every day presents a challenge and the opportunity to share my 22 years of knowledge. It’s not always about giving the perfect answer or solving something for someone; most of the time, it’s about starting someone on the right path, providing your unique perspective, receiving feedback, and adjusting your thinking to fit the situation. Bouncing ideas is where the magic happens. You’re only one person; you need the input and perspective of others.

Cloud Adoption and Benefits

Azure Navigator

Why are many organizations migrating their applications to Microsoft Azure?

Johan

The main reason I personally see is the ability to leverage the newest technologies. Most of what happens in cloud computing is traditional hosting of servers. But the shift to PaaS, SaaS, and others speeds up each year to more agile solutions, where we rather use platforms ready-made for hosting your solutions. These platforms offer agility that would be hard to mimic and manage in other settings and sometimes is proprietary to the cloud. As an admin in a previous life, it was so much easier hosting on Azure and using built-in capabilities. The uniformity of it all makes it more predictable. For example, how do I back up my services in Azure? Well, it all works in the same way. How do I monitor it? Again, it’s uniform across the modules. This means if I have a base knowledge of how the pillar elements of Azure work, then I can easily learn and adopt new functionality that lands.

Azure Navigator

What makes Microsoft Azure unique compared to competitors like GCP and AWS?

Johan

Our vast ecosystems work together. A rather large chunk of what runs in GCP, AWS, and on-premises are Windows-based systems and programming. This makes Azure ideal for hosting the services we build, working with experts on how these systems operate. Differentiators like our Azure Virtual Desktop, which offers Windows 11 multi-session, our best-in-class AI services, and the ability to bring your third-party service and use it via the marketplace preconfigured or using the license mobility that some offer. You protect your investments in Azure and other cloud players using the security services running from Azure. It’s all built-in and ready to monitor what you point it at. Importantly, it’s always a choice, and there are so many options available to our customers across the trusted brands we host and make available.

Azure Navigator

Some organizations are still hesitant to move their existing applications to the public cloud. Why is that? Do you understand their standpoint?

Johan

Yes, like with everything, there are always pros and cons. I once went to a customer who decided to move the system that controls their lift systems to Azure. At the time, they had a VPN running into Azure awaiting their ExpressRoute to be installed. The lift system was not built to handle latency well, and it offered a bit of a humorous situation where the lifts might deposit you on random floors. It became a bit of a game to check where you might end up. Of course, there are always new advancements in each area to enable solutions to do better in the cloud. We also hear a lot about “control,” the ability to run down and “hug” your server when something goes wrong. The reality is, on average, the cloud recovers much quicker from faults than many could because of the scale and monitoring in place. The commitment of SLAs keeps providers in check. And what we see finally is regulatory influence. Some industries have strict regulations on where data is kept and how it’s handled. Microsoft actually does a lot of work to ensure our services are designed to adhere to these regulations. So very few situations like this still exist, but they do. So yes, certain workloads are not yet for the cloud, but every day these become less and less. Do I understand their point? Of course, everyone has to take accountability for their decisions and actions, and you want to be sure you are choosing the right solution.

This diagram illustrates the architecture of Microsoft Azure ExpressRoute, detailing secure and reliable connections between on-premises networks and Microsoft cloud services through an ExpressRoute circuit.

Global Infrastructure and Regional Considerations

Azure Navigator

Microsoft is building more data centers across the globe. What is the benefit of having a data center near customers instead of just using West Europe if you are a customer in Europe?

Johan

I’m glad I used the lift story already. This is an example, as latency would be one reason. Also, as we discussed before, regulation of your data might require it to be in-country or even in a country with similar laws. There are many benefits, including potential cost savings, to hosting in your local data center. It also gives you the ability to use the local data center as the main one and then use another region as failover. Resiliency should be a strong consideration when designing solutions. An area that might make the decision for you is the availability of the service you plan on using. Of course, some of those might land in areas outside of your region, but the design of an Azure landing zone to accommodate multiple regions allows you to pivot easily and include this new region into your hub-and-spoke design. The essence of it is we have regions close to you so you can get the best experience from your provider. We add new regions at an astonishing rate.

Diagram illustrating Hub-Spoke Architecture on Microsoft Azure. The diagram includes sections for management groups, subscriptions, resource groups, and resources. It outlines processes for organizing resources, managing access and policies, monitoring and reporting, and ensuring security compliance. The hub-spoke topology is highlighted, showing a central hub connected to multiple spokes, simplifying network management and security.
Hub-Spoke Architecture on Microsoft Azure

Azure Navigator

Do you argue that customers should consider more than one region when setting up their infrastructure in Azure?

Johan

Yes, as mentioned above, resiliency and regulations are the main drivers for this argument. While cloud providers do everything they can to ensure local resiliency, you are always exposed to some risk. Understanding what risks you are comfortable with and which ones you are not is crucial for proper planning.

Azure Navigator

How is Azure different when it comes to backup and failovers? Let’s assume a customer has a contract with a provider that hosts a data center close to the customer, and the data center is affected by a flood.
 

Johan

Well, Microsoft has local zones designed to offset localized issues that might arise. Within our services, we have either built-in capabilities or the ability to activate zonal protection for free in most cases. Beyond that, we offer regional resiliency built into services like EntraID, which can be activated for other services to ensure your workloads are always available to your customers. If you run backup services, the storage for this can be geo-redundant as well, allowing you to restore from anywhere to the closest next region unaffected by disaster. As a client, there are many options available to reach your desired RTO, from local instance protection all the way up to geo-safe solutions.

An image showing a timeline with RPO on the left, indicating the period of lost data before a disaster, and RTO on the right, indicating the acceptable downtime after a disaster. The timeline is marked with hours and features a red “Disaster!” label at the point where data loss ends and downtime begins.
A visual representation of Recovery Point Objective (RPO) and Recovery Time Objective (RTO) in a disaster recovery plan, highlighting the periods of acceptable data loss and downtime.

Cost and Security Considerations

Azure Navigator

Would you argue that enterprises move their applications to the cloud because it’s cheaper?

Johan

I would, but with everything in this life, “it depends” comes to mind. You do have the ability to get this right or wrong. Most customers are quite accustomed to how cloud services operate these days, and Microsoft offers a lot of help in this regard. Tools like Azure Migrate and our TCO tools allow customers to plan and make the right decisions when moving to the cloud. When I was working for a partner many years ago, around 2015-ish, we set the ROI at just over 3.5 years. It’s much closer now for most customers. It’s all about planning—using the right people and the right tools to plan. The bigger question is whether we maybe delay our savings a bit and modernize now to gain so much more in the long term. A hard question for many to ponder. Compared to traditional offerings, the agility of our services, discounts on licenses, SLAs, and the joint responsibility model mean customers can move to Azure for “cheaper.”

This flowchart outlines the Azure Migrate process, detailing the stages of Decide, Plan, and Execute, which encompass discovery, assessment, planning, and execution of migrations to Azure.

Azure Navigator

Some argue that the cloud is not as secure as having their infrastructure in their own data center. What is your opinion about that?

Johan

Microsoft makes a significant investment in security; for a while now, it has been our number one priority. This means our products, our people, and our premises are all striving to be the most secure they can be. Our data centers are actually quite impressive in the way they operate, with multiple layers of responsibility and access, using top-of-the-line AI for monitoring networks, video, and other signals to ensure security. The scale at which Microsoft approaches hosting means that the resources they can dedicate for this purpose are impressive. Most companies cannot achieve this level of security within their budget. And yes, on-premises you might have an air-gapped environment, and those are needed. We go back to a few questions ago on the workloads that work for the cloud and those that do not. Security and access to data is an everlasting battle between secure and usable. I’ll close the point by saying that we are one of the most secure services out there today with world-class products, services, and facilities.

Gartner Magic Quadrant for Security Information and Event Management (SIEM) as of January 2024. The quadrant categorizes companies based on their “Ability to Execute” and “Completeness of Vision.” Microsoft is positioned in the Leaders quadrant, along with IBM, Securonix, and Exabeam. Other companies are placed in the Challengers, Visionaries, and Niche Players sections.
In 2024, Microsoft is again named a Leader in the Gartner Magic Quadrant for Security Information and Event Management (SIEM), highlighting its strong execution and comprehensive vision with Microsoft Sentinel.

Cultural Differences in Cloud Adoption

Azure Navigator

You manage Cloud Solution Architects across Austria, Finland, Ireland, and Portugal. How do different cultures approach cloud adoption differently?

Johan

This is where it gets interesting, for me at least. Finland is at the forefront of innovation in this set of countries, with the public sector using cloud services for a lot of innovative services. It’s unusual for us to see quick adoption of cloud services in the public sector for reasons discussed earlier, but in Finland, they make good use of these products. Austria is quite interesting in the way they approach cloud and always strive to make it super secure. This leads to sometimes interesting design decisions that push the boundaries of capability and can lead to new feature requests from our product team. They are also quite strict on roles and responsibilities, so it’s important to have a good cloud operating model defined. Ireland is a bit more of an all-rounder. Here we host most of the world’s tech companies, a lot of pharma, and such, but the main theme is multinationals. That means working with customers who are located across time zones and cultures. Maybe that’s what helps me adapt to managing four different countries to start with. Ireland’s customers are more about the relationship with their providers, so they form a close relationship and really put a lot of trust in you as an advisor. That trust we work hard for in all countries, and we want to ensure we give the right advice for the customer. Portugal is almost a mix of all of these in one place across different customers. This is, of course, a bit of a generalization. When we get into the detail, there are more similarities, and that means our team across these countries is well-suited to deliver anywhere.

A diagram illustrating the Cloud Adoption Framework methodologies. It includes eight steps: Define Strategy (understand motivations, business outcomes, business justification, prioritize project), Plan (digital estate, initial organization alignment, skills readiness plan, cloud adoption plan), Ready (operating model, landing zone concepts, design area guidance, implementation options), Adopt (migrate, modernize, innovate), Secure (risk insights, business resilience, asset protection), Manage (business commitments, operations baseline, operations maturity), Govern (business risks policy and compliance governance maturity). Each step is represented by an icon and brief description.
The Cloud Adoption Framework: A comprehensive guide to cloud adoption, detailing strategies from defining business justifications to managing and securing cloud environments.

Azure Navigator

What are some of the unique challenges and opportunities you have observed in these different regions?

Johan

This is a tough one. I’d skip the challenges because they are quite specific and would be lost without context. But the opportunities, on the other hand, are vast. The first thing that comes to mind is the ability to bring the learnings from one region to another. I used to really enjoy this when I was a Cloud Solution Architect (CSA) myself. We’ve seen what works and what doesn’t. Because different regions have different rates of adoption, it also means that our skills are exercised in an ever-changing way. It ensures the CSAs are never bored. For us as a region, it’s the opportunity to help our clients expand into areas of tech they have not been exposed to yet. We can be the advisors that bring the eureka moment to the design team, providing constant innovation and keeping our clients as close to the edge of technology as they feel comfortable with.

The Future of Cloud Computing

Azure Navigator

Cloud computing is often considered a disruptive technology. Why is that?

Johan

Cloud computing is often seen as a disruptive technology because it fundamentally changes how businesses and individuals use and manage computing resources. Take scalability as an example. While the cloud is not new, the fact that it can scale to epic proportions on demand is well known by most now. But it still disrupts an industry that today still has a large reliance on sometimes static on-premises workloads. It brings flexibility in planning. It allows us to scale based on cost, demand, and other factors we might not have been able to foresee with 100% accuracy. It allows for more innovation within your team. Instead of trying to get a new product that might have very specific hardware demands installed on-premises to test or Proof of Concept (POC), you can simply spin it up for the time you need it. Like I said, nothing earth-shattering, but still, these concepts disrupt the everyday way things are done or were done.

Azure Navigator

Where do you see the future of cloud computing? Will we even need our own hardware if everything is in the cloud?

Johan

I’m an old-school infra guy, and honestly, I can still see scenarios where you need your own hardware. There are still limitations like access to the internet or security concerns where services need to be air-gapped. So yes, in the medium term at least, my opinion is that there is still a place for these. I’m from Africa originally, and I don’t see some of the mining equipment being controlled at a rural location from the cloud. So those PLCs still live there, onsite, connected to a 1987 machine (true story). But in these examples, there is always an opportunity to connect the relevant data to the cloud that might not need real-time reporting or assessments using IoT, for example. So it’s a case of having both for the right purposes.

Azure Navigator

Where do you see potential for customers moving to Microsoft Azure?

Johan

Everywhere, ha ha… Here I’m sure a bit of bias will creep in. Honestly, this is a bit of an open question, and the answer differs for every single customer or startup. But we have wonderful assessment tools like “Azure Migrate” or third-party tools on the marketplace like “Dr. Migrate” that will deliver these assessments to customers and help them make the decision where it makes sense in the short and medium term. It allows you to plan for change to PaaS or just good old lift and shift. And of course, speaking with your account teams and CSAs.

Workforce Impact and Change Management

Azure Navigator

People are sometimes afraid of moving their applications to the cloud because roles might change, and some might fear losing their job in the new setup. What is your opinion on that?

Johan

I remember a long time ago, I was giving a session, and the PowerPoint slide suggested you could now have fewer people doing X, Y, and Z. The look on everyone’s faces was unforgettable. Over the last eight years, I haven’t seen a company downscale their personnel (I was mostly operating in EMEA). Everyone can be reskilled, and the base knowledge you have from on-prem is exactly what you need to skill up for the cloud. There are certain principles of computing that stay the same, and all the companies I’ve worked with used training from partners or Microsoft CSAs to ensure their teams can successfully transition. It’s great new skills for them, and the company benefits from having a person who can support both the old, which we all know will linger, and the new. So, it’s not something I’ve seen.

Azure Navigator

 What strategies can organizations use to manage this transition and support their employees?

Johan

I keep answering questions before they are asked. I’ve said it already, but there are wonderful training programs. But it starts with a person who is curious and willing. Before I interviewed at Microsoft for my technical interview, I set up a station with four screens in my dining room. I would read up about the basics of, say, an Azure landing zone and move from there to concepts I did not know or understand. I would open videos, documents, and threads on each screen. And it was easy because I wanted to know more; I wanted to see how something works and why it works that way. So bring with you a sense of wonderment and curiosity when attending training.

Conclusion

Azure Navigator

 Thank you so much for your time and insights, Johan. Is there anything else you would like to share with our audience?

Johan

Azure has been a part of my life for the last 10 years now, and I’ve seen many ways that it is employed and used. The next big shift is here in the form of AI, and Azure will form a big part of that future. I would encourage companies to speak to the relevant professionals and investigate with interest the possibilities that this brings to their businesses. Looking forward to seeing more of these interviews, Gregor. We have so many great professionals to share their views.

Share the Post:

Leave a Reply

Your email address will not be published. Required fields are marked *

Related Posts

A photorealistic illustration showing the transition from traditional on-premises operations to a modern, cloud-centric approach. On the left side, the image depicts an old-fashioned scene with people using horses for transportation, symbolizing outdated methods. On the right side, a futuristic and vibrant scene showcases advanced technology and infrastructure, representing the adoption of Cloud Center of Excellence (CCoE), DevOps, and Site Reliability Engineering (SRE) practices. The contrast highlights the evolution towards a more efficient, scalable, and innovative future.

Transitioning On-Prem Operations to Cloud Center of Excellence (CCoE) and DevOps Teams

Transitioning on-prem operations to a cloud-centric approach is a critical step for organizations aiming to enhance scalability, reduce maintenance costs, and accelerate innovation. Adopting a Cloud Center of Excellence (CCoE), DevOps, and Site Reliability Engineering (SRE) practices can significantly contribute to successful digital transformation. This article provides a comprehensive guide to understanding these concepts and outlines the steps involved in the transition process.

Read More

Transitioning from Legacy AD Tier Model to Modern Enterprise Access Model: Challenges, Solutions, and Implications

Transitioning from the legacy Active Directory (AD) tier model to a modern enterprise access model is a critical step for organizations aiming to enhance their security posture and ensure compliance with regulatory requirements. This shift involves addressing significant security, compliance, and operational challenges to safeguard sensitive data and maintain regulatory compliance. By adopting a modern access model, organizations can achieve greater flexibility, scalability, and robust security, ultimately protecting their valuable assets in today’s complex digital landscape.

Read More