BLOG

AIOPS – News too good to be true. AI

Publication Date

February 27, 2020

Share

I started with the word AI (Artificial Intelligence) and I’m sure “again?” you thought to yourself. Yes unfortunately. Today we are talking AI. Because after 3-5 years, we will not talk at all, there will be AI in every moment of our lives, even in the tea machine, it will be something as important as the socket on the wall and to which we do not pay attention.

While we see emerging solutions to provide the concept of AIOPS, we also see some monitoring and management tools starting to use artificial intelligence to identify operational problems of hybrid IT structures.

Infrastructure monitoring products generally aim to monitor and manage business applications by integrating capabilities such as alarm consolidation or log analysis with process applications, and even achieve efficiency in operations by automating them in some limited areas.

 

Similarly, application performance analyzes aim to monitor the performance of business applications, integrate with infrastructure tools where possible, and expand the perspective to show both performance and errors in more detail.

These tools are not tools that can be used right out of the box, and they are tools that require attention throughout their working lives while being integrated with each other. The reason is not that their capabilities are limited or that the products are not well designed, but that the environments in which they work are constantly changing and developing; therefore, regular attention and effort is required to keep up.

AIOPS: Magic Wand (?)

It is not possible to say, “Throw away the tools you use and replace them with a brand new AI-powered, all-in-one tool.” But it is possible to say that with the tools we have, a concept has emerged that makes it possible to easily manage increasingly complex and growing environments. It is also very pleasing to see that there are solutions that realize this concept. One of the most unpleasant parts of life in IT operations is probably not being able to identify the source of a problem that affects a business application, despite long studies. After defining the problem, it is easy to bury it in history AIOPS NPM IPM APM, expertise and operational capabilities are sufficient for this. As complexity increases, the potential for problems increases, and the time to find the problem also lengthens statistically.

AIOPS helps us find the root cause in two ways.

  • Reactively, when problems occur, it allows us to easily find the root cause despite the extensive IT architecture.
  • It can proactively notify us that a possible business application will be negatively affected by stating the root cause.

The good news is that there are solutions that realize these concepts.

https://info.stackstate.com/gartner-market-guide-for-aiops-platforms-2019