Digital transformation in hybrid IT landscapes
What does a typical IT landscape look like? How many applications & integrations do a small, medium and large company really need? These questions are raised throughout our daily work and we ask these to business analysts and enterprise architects during discoveries on customer environments. Usually we discover that way more applications and integrations are present than we initially would expect. But how to deal with it?
- Due to digital transformation everything will become connected
- Securing and managing data is getting way more complex
- Lack of insight and overview results in poor governance
A nice example of everything being connected is the rise of IOT devices. This is a market trend which will have a tremendous impact in the future. Next to the usual devices like mobile phones and computers, the number of IOT devices will increase significantly in the near future. Cars, trucks, roads, railroads and buildings are rapidly equipped with IOT devices. So, what are those IOT devices? Physical things that can measure things and send data regarding to their activity on these things. These decentral data streams are collected and stored elsewhere, and to be used later. The increase of connected devices is devastating for traditional IT landscapes, which are not designed to handle these large volumes of data. In the chart below you see a forecast of the growth of IOT devices which results in more IOT devices than traditional devices within the next 3 years. This growth will not stop and continue the next decade.
This is an example of an IT landscape which is getting increasingly complex. As you can see: the picture is spaghetti-like and a tangle of connections. Of course, this an extreme example, but we see that at large companies this view is dangerously getting close to reality. And that could be frightening for us.
In the end it will be very difficult to keep a good overview. Preventing an increase of connections is almost impossible. The digital transformation forces us to facilitate the transition to new IT architectures, where a picture like this is just normal. It’s better to focus on ways to deal with this change and be assured that we use the best tools and approaches. This is not the only challenge.
What’s happening in the business? Our customers deal with a couple of challenges:
- From on-premises to cloud: first we see a movement from on-premises to the cloud and beyond that: serverles, purely services instead of software instances and hardware.
- From single to multi: then we see a transition from single to multi protocol: legacy and old fashioned software communicates different than the new technology. Nowadays you need to handle both, for instance handling files on the file system, but also webservices & API’s.
- Increasing number of applications: the number of applications increases and everything is connected to each other.
- 100% Uptime demanded: everything is supposed to run and perform always, 100% uptime is actually getting very normal. Downtime is not tolerated, there is no room for outage.
- The business wants more: IT departments have less supervision and control on IT related initiatives. Nowadays, the business has more and more a leading role and is also initiator of new solutions. The IT department is expected to keep up the pace of the business, otherwise it will become a bottleneck. Every application needs data and therefore also integration or some way of connectivity. For instance a mobile app that needs master data from your ERP or back office systems. Even the most simple solution needs data.
And that’s a real challenge. We already know that integrations and data related activities are taking most of our time, therefore it’s also responsible for the biggest portion of the costs. It’s not only Gartner who is saying this, we hear this from our customers as well.
Which other movements do we see in IT landscapes?
- Microservices getting more populair: in a microservice architecture applications should be independent and responsible for all communication to other apps in their landscape. See also my blog about microservices for more on this subject.
- Devops, BizDevops or DataOps movement: Integrations and data was formerly an area of the expert or specialist. This is changing. Devops teams are responsible for their own interfaces. How to deal with this responsibility. Do we have the right tools? How do we enforce guidelines and architecture best practices for non-experts? Those are the questions devops teams and architects are struggling with.