Day after day, we see companies around the world embrace a cloud-native approach to software delivery. Why? To meet today’s needs, apps need greater resilience, scale, and adaptability. But this transition to cloud-native isn’t easy. Just buying new technology won’t make it happen. Smart companies focus on improving their internal skill set while also engaging with outside experts to accelerate their learning. The folks at ITQ are the type of outside experts worth talking to.
Based in the Netherlands, ITQ is a new Pivotal partner helping Dutch companies become cloud-native. I reached out to Ruurd Keizer of ITQ to learn more about what they’re doing, why it’s working, and what you can learn from their journey.
Richard: You've recently set up a practice with ITQ to help customers deliver cloud-native applications. Given that many companies are trying to add this practice themselves, can you explain how you went about building your team? What skills did you look for? What kind of training did you offer your employees?
Ruurd: Our cloud-native practice has two components. The first is aimed at cloud-native platforms in general and Cloud Foundry in particular. We offer Professional Services in the form of consultancy and (platform)engineering to mostly enterprise/government customers: both in early stages where we help customers identify what products best suit them, and in later stages where they have a running platform and wish to onboard internal customers by enabling services on the platform that don’t come out of the box. The second part of the practice is about replatforming applications to run well in a cloud environment. This part also comes with a two pronged approach: before we rush into the engineering we help customers identify their real business needs to make sure the replatforming effort is well spent.
The step from traditional enterprise IT to working with a DevOps enablement platform – which Cloud Foundry is – is a cultural leap which is not for everyone, at least not initially. So we built our initial team by identifying individuals in our organization that were ready for the next step, and are revolutionary instead of evolutionary thinkers. These internal champions of the practice were then surrounded by other skilled engineers and consultants.
For the platform engineering part we looked for people with a strong background in infrastructure (VMware vSphere) and cloud management/orchestration, and with experience with configuration management and continuous delivery tooling (Puppet, Ansible, Chef, Jenkins, Concourse). They had to understand the developer mindset, and preferably be able to code some small things themselves (Service brokers, shell script). For the replatforming part of the practice, we identified strong all-round developers. Their exact toolset and framework of choice mattered less to us than the ability to understand how to work with abstract concepts like distributed applications, CQRS+ES, API first, and personality traits like being pragmatic, empathic, and customer oriented.
ITQ is by trade a consultancy company. We intentionally don’t sell hardware or licenses, so our work isn’t finished when we delivered and installed a box, but when we leave a happy customer by helping them with our knowledge and experience. Therefore, we invest a lot in people every year to make sure our consultants stay at the top of their game. We offer them to get certified in the VMware, Pivotal and Microsoft programs (we have a number of VMware VCDX’s), follow trainings online (Coursera, Pluralsight) and offline (VMware Livefire, Pivotal Platform Acceleration Lab), and regularly go to conferences. Also we help organize community events and usergroups, and encourage our consultants to contribute to the community (writing blogs, speaking at conferences, contributing code to OSS).
Richard: Do your customers actually ask for cloud-native apps, or do they ask for the attributes they're after? Things like resilience, rapid changes, and improved scale?
Ruurd: Definitely the latter. On more than one occasion we still have to tell customers what ‘cloud-native apps’ actually are, but when you explain that it’s a set of patterns by which you can move code changes (read: business value) into production on a daily basis day while keeping the confidence that stuff won’t break, it’s often exactly what they want.
Richard: What's some tangible advice you have for those looking to modernize their application portfolio? What should they stay away from to start with? Where is an area of underrated impact?
Ruurd: The most important thing is to identify and understand your core business before you start. This may sound obvious, but over time Enterprise IT has moved through waves of increasing confusion: ‘outsource everything’, ‘insource everything’, ‘do everything yourselves except own the hardware’, ‘move everything to the cloud’. It’s not hard to find people who are fundamentally confused about how their work is adding value to the business.
Once you understand your core business, buy – as a service - everything that’s not contributing directly. It will be somebody else’s core business to provide these tools, and they will be better at it than will you ever be. Adapt your processes to the few things that are different in the software, don’t customize until it fits the process, it’s not worth it as it’s not your core business.
The remainder will be (software that is critical to) your core business. You should never be able to buy that, as it would mean your core business doesn’t have a unique selling point. If you find out you can buy it anyway you have another challenge entirely.
Start by modernizing the apps that make the most business impact, and work your way through.
Richard: ITQ covers a handful of industries. Are there ones that are surprisingly forward thinking with regards to technology? Something we might not expect?
Ruurd: You might traditionally expect the high-tech vertical to be the most forward thinking with regards to technology. While they definitely push the tech they sell, we see they aren’t nearly always as far as other industries in adopting cloud technology and culture. The most forward thinking industries we see today are banking and government. Everyone is familiar with the term ‘FinTech’ by now: the banking sector has really embraced digital transformation, especially on the customer facing end. On other levels there is still much to be gained: how to deal with data, and how to replace at least part of your compliancy culture with rugged software (see the manifesto).
The unexpected factor is in government. With a traditional dusty image, and all the excitement going to the startup and enterprise space it’s really cool to see the pockets of government that are moving at least at the same velocity adopting cloud-native platforms and way of working.
Richard: You recently won a regional partner award from VMware. With the obvious growth of public cloud offerings, where do you see the future of private cloud solutions?
Ruurd: Private cloud will continue to be around for a very, very long time. While public cloud has some very tempting propositions, especially given the ‘buy everything except your core business’ that we touched on above, there are a number of problems that prevent it from being a brainless move to go all-in on public yet. Cost is one: unless your applications are cloud native, and use the cloud in the best ways possible, public cloud is going to be more expensive every single time – also in terms of Total Cost of Ownership. Another one is vendor lock-in: while the large public cloud providers provide amazing services to connect to your apps they have one thing in common: they are all engineered to be ‘just non-standard enough’ to keep you from migrating to another cloud. Cloud Foundry is awesome in this respect as you can host it on every major platform, and the industry certification guarantees your apps can be moved without any effort. That is, if you don’t depend on a vendor specific backing service. Finally, there’s the problem of security and data gravity. Industries like banking and high-tech will not move their most important data to the public cloud, be it because of regulations or concerns about the safety of company secrets and data. That also means the first layers of services around that data will always be located in a private cloud.
About the Author
Richard Seroter is a Senior Director of Product for Pivotal, a 10-time Microsoft MVP for cloud, an instructor for developer-centric training company Pluralsight, the lead InfoQ.com editor for cloud computing, and author of multiple books on application integration strategies. As a Senior Director of Product at Pivotal, Richard heads up product marketing and helps customers see how to transform the way they build software. Richard maintains a regularly updated blog (seroter.wordpress.com) on topics of architecture and solution design and can be found on Twitter as @rseroter.Follow on Twitter More Content by Richard Seroter