Build or buy. It’s never an easy choice. Do you look for an off-the-shelf application, or do you go in-house with a custom solution? Of course, there are many factors to consider, including time to market, cost of ownership, functionality, and whether you have the knowledge and expertise to build (and support) your own solution.
Time and again, I hear the same build or buy question when IT organizations are looking for an event management solution. Here’s what I tell them.
If you’re a small or mid-market company, you may be able to survive with a basic event management tool that you build yourself. That’s assuming that you have a simple IT environment, with a limited number of applications and not much change. For example, if you have a minimal online presence and a few back-end systems, building your own event management tool is an option. Of course, you need resources and skills to build and maintain a custom system, so it’s still a difficult decision.
However, if you are running IT in a complex environment, then building your own event management platform just isn’t an option. Here’s why.
- There are too many tools to integrate
If you’re like most enterprise IT organizations, you probably have many monitoring tools. There are different tools to monitor servers, applications, networks, virtualized infrastructure, and so on. You will need to integrate all of these tools – and if you build your own event management platform, then you will have to do this from scratch.
And, the cloud makes things worse – not better. The cloud has many benefits, but simplified monitoring isn’t one of them. To take advantage of the cloud’s agility, development teams are increasingly using microservices and containers to develop and deploy applications. These dramatically increase monitoring complexity and volumes. In fact, one recent study showed that container-based applications generate 18 times as much monitoring data as traditional application architectures.
On the other hand, with a commercial platform, you get off-the-shelf integrations with a wide range of monitoring tools – and if you don’t, look for another event management vendor. These integrations are largely plug-and-play, so you’re up and running quickly, and don’t need to invest in up-front development or ongoing maintenance.
- Processing Monitoring Data Is Difficult
Once you’ve got monitoring data into your event management tool, that’s just the start. Unless you want to drown under noise, you’ll need to normalize, filter, deduplicate and correlate that data. And remember, this isn’t just for a single monitoring system. To get an end-to-end view of your applications, you’ll need to stitch together data from multiple monitoring sources.
Doing this without the right tools takes a lot of effort. Unless you intend to build your own commercial-grade event management platform, you’ll end up doing this with scripts. Just creating and testing a script is a lot of effort, but the maintenance overhead is a nightmare. When there is a significant application change, you’ll likely need to update your scripts, creating a constant drain on your resources.
Legacy event management tools also relied on scripts, but now this has changed. Leading-edge platforms now let you configure logic rather than code it, using intuitive drop-down menus and widgets to setup rules. This approach takes far less time and is more resilient – so you spend your time managing your IT environment, not your event management platform.
- Custom platforms don’t scale to today’s operational demands
We’ve already talked about how microservices and containers are dramatically increasing monitoring complexity and volumes. However, that’s just the start. If your enterprise is pursuing digitalization – and most are – then you’re facing an explosion of new applications to manage. If you’re struggling to keep the lights on today, how are you going to cope down the road?
If you have a custom script-based tool, you’re not. On the other hand, best-in-class event management platforms tackle this issue head-on. They use machine learning to automate event-management processes – for example, automatically diagnosing the root cause of service issues, and even predicting service outages. That means you work smarter and faster, making the most of your IT operations staff.
One last word. Even if you only need “traditional” event management – rather than a next-generation intelligent platform – aligning people, processes and tools can be difficult. If you want to understand these challenges and how to address them, take a look at our Event Management Playbook.