Our many conversations with CIOs around the world have yielded one irrefutable fact: today’s tools woefully underserve IT operators.
It’s increasingly difficult for CIOs to assure the overall health of the business while delivering quality service and shortening mean time to resolution. Twenty years ago an enterprise might have had a limited number of systems running all its business processes. A well-trained IT operator could practically keep everything in her head. But in today’s digital economy many businesses have thousands of systems. Meanwhile, the state-of-the-art applications used by IT merely allow them to search for specific system data and graph it.
Take for example the evolution of how personal checks are deposited at a bank. Before ATM machines, a bank’s customers gave their checks to a teller, who would then enter them within the bank’s check clearing system. Problems were simple to diagnose because the entire process was point to point and involved no more than a handful of systems.
Today, checks can be deposited using smartphones and tablets simply by taking a photo. However, this seemingly simple transaction involves dozens of very sophisticated systems, such as various mobile platforms, different versions of the banking application that use the camera on the device, multiple mobile networks, the cloud service receiving the requests, the proxy servers and load balancers, front end application servers, authentication servers, check image processing systems, account reconciliation systems, and a digital clearing process, which in itself is comprised of multiple systems.
If something goes wrong, our unsung IT hero only has today's primitive tools to diagnose and remediate the problem. It’s no longer possible to simply walk back the transaction to learn what went wrong because there are countless variables and interactions between dozens of independent systems. It’s difficult even knowing where to start. In the meantime, customers grow frustrated and possibly move their business to another bank, impacting the bottom line.
On the other hand, the business now has applications that are predictive and personalized, leveraging the many benefits of big data. Unfortunately, IT is forced to use tools from the Stone Age to keep the business up and running.
It’s time to take the same principles used in front line big data applications and apply them to the IT operations domain. What’s needed is a new class of IT operations applications that use machine learning, predictive models and personalization algorithms to keep business systems running optimally. These new business operations applications will look at thousands of systems in milliseconds and proactively alert IT where incremental failures are located, what processes are behaving differently, and how they may be tuned for maximum productivity.
Big data IT operations applications will ultimately streamline how complex multi-system business are run, taking IT from the Stone Age to the forefront of the digital economy.
Stay tuned to this space.