By 2019, 25% of global enterprises will have strategically implemented an algorithmic IT operations (AIOps) platform supporting two or more major IT operations function.
Pankaj Prasad, principal research analyst at Gartner, discusses about trends and developments in algorithmic IT operations, and what organizstions should be focusing on this year.
What is algorithmic IT operations (AIOps)?
AIOps platform technologies comprise of multiple layers that address data collection, storage, analytical engines and visualization. They enable integration with other applications via application programming interfaces (APIs) allowing for a vendor-agnostic data ingestion capability. AIOps platforms can thus seamlessly interact with IT operations management (ITOM) toolsets because of the ability to deal with data from any tool irrespective of the data type.
How is AIOps different from data analytics?
Over the past four to five years IT organization teams involved in availability and performance management have made increasing use of big data and machine assisted analytics for improving diagnostic and troubleshooting capabilities of their teams.
While the broader data analytics is concerned with use-cases and models that will be based on an organization’s unique needs, AIOps platform technologies are designed for typical IT operations use case which mainly involves:
* Post-processing of events streams that come from monitoring tools;
* Bi-directional interaction with IT service management tools; and
* Possible integration with automation toolsets for implementing the prescriptive information provided by the platform.
What are the drivers for AIOps adoption?
Early technology adoption of big data and technologies focused on ITOM toolsets, purely around data-centric monitoring and analysis was called “IT operations analytics” (ITOA). The impact that this early adoption has created on the availability and performance discipline is now reshaping ITOM as a whole and beyond. This broader trend is what we now call “algorithmic IT operations”. It has an interesting and important interplay with all disciplines under IT operations and the potential for creating an orchestration across various ITOM toolsets.
What are the challenges inhibiting adoption?
The main challenge is separating the marketing jargon from the actual capability and assessing the effort needed by the technology user. Machine learning – which is a key component in analytics platforms – needs huge amounts of data and more importantly, interactions with humans in a real-world scenario. These two components are critical to extracting value from AIOps platforms, and they take time. Any investment into the toolsets needs to account for the investment in terms of data, human-machine interaction and time.
How is AIOps relevant to business?
The traditional role of analytics is shifting from merely supporting decision making, towards increasingly driving business processes by not only recommending the best possible actions, but triggering those actions in an automated manner. In addition, analytics is being used to predict preferences of customers to drive better and more engaging customer experience.
All of this means, analytics platforms are becoming central drivers of modern business and I&O teams cannot look at analytics platforms like a lab environment anymore. Consistency in performance of data and analytics platforms is key, and only AIOps technologies are equipped to monitor the modern data and analytics platforms.