I belong to an amateur soccer club that could raise enough funds to purchase Automated External Defibrillators (AED's) for all its teams that compete in the Over 35's Leagues. For those not familiar with an AED, it's a portable electronic device that can be used to treat cardiac arrest for heart attack victims. I recently attended a training session on how to use the AED and was impressed by the way it provides prescriptive verbal instructions and automates critical decisions such as when to apply an electrical shock to the victim. In stressful situations, it can be hard to think clearly and determine the right decision, so technology such as this does literally save lives.
It's the combination of human intervention and automated systems that provides this life saving service. So, it got me thinking about why there is still much skepticism towards automation in some professions. Why is it tools that manage business-critical functions, such as communications networks, still rely mostly on operators using them to look for the proverbial needle in the haystack when troubleshooting? There is this belief humans can make better decisions than computer systems and yet, one by one, industries and professions are disrupted as hard evidence proves this wrong. Despite the compelling facts, many people remain skeptics. I've heard friends say they would never want to be operated on by a robot, and yet they are quite happy to put blind faith in a surgeon that might be having an off day. How quickly we have forgotten the pilots that fly us around the world, for the most part, are only there in case the automated systems fail.
At IR, we have added prescriptive guidance into Prognosis for UC, our unified communications performance management tool. This has been well received by our customers as, unlike most other solutions on the market, Prognosis not only identifies and diagnoses the problem, but also provides advice on the likely cause and recommended remedial actions. If we were to take this a step further, why not let Prognosis automatically execute the necessary action? Such a capability has in fact been developed and publicly demonstrated. It leverages Software Defined Networking (SDN) by sending instructions to a software controller designed to make the final decision. However, while I have witnessed a lot of interest in this capability, I still hear many reasons why IT departments feel these important decisions must remain with humans.
Now granted, I understand network automation technology is still maturing but what if we could adopt the same philosophy as the AED? Whereas the UC management tool could determine if the situation was critical, such as all customer communications lines down, and it could automatically intervene and takes drastic action such as rebooting a device? In reality, we are some ways off putting our faith in this approach but there could be a middle ground. Perhaps this is lighting up a big red button but leaving the final decision to press it with the IT Operator.
I am grateful for the investment my soccer club has made in the AED and have complete faith in the automated actions it might take if I ever have to use one. I also believe it is a matter of time before the technology matures enough for IT professionals to put their faith in the systems automating critical network actions. A this point we will be one step closer to realising the dream of self-healing IT systems.