Imagine a car where the steering wheel and drivers are optional extras. A fully autonomous vehicle which unlike its level 4 predecessor that needed to work in carefully managed environments (usually geofenced urban), is now able to self-drive anywhere. An advanced stage where cars have become sentient like – continuously perceiving, feeling and reacting to a myriad of conditions. If faults are predicted the level 5 car will order the part and coordinate repairs (probably via software updates as you sip your morning Latte). It's also likely that you'll be renting the car rather than owning it, so the vehicle will be learning and adapting to new behaviors, routes, and conditions – continuously.
Many pundits claim that level 5 cars are still a decade away. That's probably true, but the potential to leverage AI and machine learning to leapfrog auto manufactures and disrupt transportation markets is huge. So, we can expect to see more variants of Google WayMo prototype in a suburb near you and millions of autonomous cars on the market the not too distant future – 21 million by 2035 according to the analyst firm IHS automotive.
So what will a level 5 continuous AIOps system look like and what will be the business and operational impact?
Well, let's start with that word "continuous." Like a fully autonomous car, advanced AIOps systems will continuously process massive amounts of information at tremendous scale and apply real-time machine learning modules to gain new and deeper insights. No data will be off-limits, with logs, metrics and application performance instrumentation, user-experience data and IoT data all enriching the system with additional context.
With these systems the human cognitive overhead associated with lengthy data gathering, cleansing, correlation and interpretation has been eliminated. All replaced with a system that uses massive learning sets to increase intelligence and deliver new capabilities over time. Like the level 5 car builds an understanding from past behaviors and conditions, then so will AIOps – continuously fixing and tuning without operator intervention, and better still, without customers even realizing there's been a problem.
The impact on IT operations will be profound. Now, the cost / human capital overhead associated with dedicating valuable staff to low value activities will be reduced dramatically. Rather than tie staff up on mundane and interrupt-centric tasks that increase technical debt and atrophy organizational capability, level 5 AIOps will build strength. This means eliminating the heavy-lifting and augmenting their human co-workers with optimization knowledge and learnings – so that everything and everyone get stronger.
So, imagine working with an advanced AIOps system as a partner to use AI and machine learning in ways that are far more beneficial to you and the business. Like for example:
■ Rather than fighting fires and fixing repeat problems, applying AIOps analytics across the continuous delivery pipeline to determine which apps, code, functions, practices – whatever – correlate to the best performance and business outcomes.
■ Not waiting until the end-of-year to guestimate infrastructure investment budgets but using AIOPs to continuously determine the optimum placement of workloads over elastic infrastructure in real-time.
■ Never having to hunt for the root-cause of problems in a haystack of needles but using AIOps to advise business on the cost-benefit of a 100ms requested performance improvement.
And just as a fully self-driving car will service many drivers in many conditions, level 5 AIOps will support multiple stakeholders and use cases; extending beyond their normal monitoring purview.
Like for example:
■ Automatically correlating application performance data, security information and event management (SIEM) with external threat intelligence inputs to proactively identify potentially malicious activity and risks within applications and remediate them – automatically.
■ Gathering all risk metrics associated with a particular public cloud application architecture, projecting failure points and outlining the appropriate exit strategies to executives.
■ Applying AIOps against production cloud workloads – continuously recommending and tuning the optimum instance types, network configurations, storage placement, serverless cold and warm starts.
Realizing that traditional methods can no longer scale to support the unpredictability of modern dynamic systems, organizations must replace low-value labor intensive monitoring with more advanced AIOps systems. Use the guidance laid out in this blog series to start that journey with advanced AIOps systems that become the self-driving apps for IT – fully automated learning and intelligence to self-heal and optimize applications. Learn more about how CA can help you on our autonomous journey in this recent press announcement on self-driven autonomous remediation.