The Evolution of Artificial Intelligence (AI) and DevOps

If you work in the IT department of a company with SLA that asks for 99.6 percent availability, here is a scenario that may sound familiar to you. Customer calls at odd hours with some failing transaction issue or an information being processed wrongly, you ask your specialist from Ops team to look into the system immediately. What happens afterwards is known to everyone - more than a hundred thousand messages logged at the set timeframe – a data set impossible for a human being to review line by line. A trial and error resolution starts. With concept of DevOps emerging in recent time this problem is getting into gigantic proportion. The dramatic pace in which the processes, technologies, and tools are changing, it has become quite problematic to cope with such scenarios. Moreover, the pressure business users have been putting on DevOps teams is manifold, expecting that everything should be solved within minutes. The big data generated out of the logs are not easy to crack with manual interventions.

 

So, what do we do with it and how sleepless nights can be overcome and find the critical triggering event without much of time pressure? Basically, we need something automatic,which can pinpoint the problem location in the plethora of logs and help in resolving the current issue as well as ‘Reducing the PBI cycle time’ for fixing it properly to avoid recurrence. This is where Analytics with artificial intelligence capability finds a foothold.Combining Big Data, AI, and Domain Knowledge; technologists have been able to create big breakthroughs and opportunities for the DevOps teams.

AI is no more a buzzword or fiction. It is substantiating the incapability of human mind to handle speed, complexity and quantity of the data and helping to optimise the root cause analysis. So how it is working?- It does basic algorithm stuff like how a user investigates, monitors, and troubleshoots events, and allows it to develop an understating how humans interact with data and conclude. Some higher level solutions are being propounded as well with bringing in the concept of cognitive thinking in system. This technological concept offers to use machine-learning algorithms to match human domain knowledge with log data, along with repositories and discussion forums threads. Using all this information, it makes a big data of relevant insights that may contain solutions to a wide range of critical issues, faced by IT DevOps teams on a daily basis. The core of these cognitive insights is based on the ELK stack, it sorts and simplifies the data and makes it easy to have clear picture.  

Using AI driven log analytics systems, it becomes considerably easy to efficiently solve issues. Such a system will definitely have a considerable impact on DevOps. ELK is fast becoming the trend and many reputed vendors are proactively researching and testing AI to bring in automation and providing logging solutions. It has become a good solution for companies to create a setup without incurring astounding upfront cost. 

In summary, integrating AI with log management is way to go for DevOps and can bring efficiency, risk management, optimisation in data management and trouble shooting and in the end customer delight. The endless wait of solution and root cause analysis, daily chorus of ticket priorities and SLA breaching could be a history!