By Clayton Ching
Gartner and other IT pundits have continually pushed the concept that the app is king. In many respects, it is. Application developers remain one of the "hottest IT jobs," according to the trade press, though more of these jobs are being outsourced than ever before and online job boards are full of less glamorous titles such as "IT operations manager" and "network architect." Apps make money and they propel startups to IPOs. Apps get people to buy $600 tablets. But they aren't always the best focus for the people who are responsible for keeping the lights on in IT. Without the infrastructure optimized, even the greatest apps don't deliver their value because they aren't performing efficiently.
For years, the IT service and the apps beneath it have dominated discussions around management and performance. App performance management (APM) has emerged to dominate the IT landscape, and large IT vendors have continued to push their pricey, legacy APM suites. Yet the app, albeit center stage from the business perspective, is not where IT needs to focus today: it's the infrastructure events that run beneath and all around those apps. Understanding the connection between the various components and technologies is critical in managing IT operations today.
Monolithic applications are taking the back seat to applications running in highly distributed, cloud or hybrid cloud environments. The modern, "transient" app consists of virtual components that appear and disappear with little notice; people are also using different methods to access them, such as tablets or smartphones.
This constantly shifting environment is driving many IT organizations to dig a little deeper to see critical performance trends. Today, managing applications and service levels means managing all of the events across every part of the infrastructure, in real time.
First, you ask, what's an event? An event is any alteration in the infrastructure, whether that's a traffic spike or a server going offline or a database configuration change. Events, unlike incidents, are not necessarily negative. Yet with the thousands of events occurring every day across a company's infrastructure, it's impossible to manually track and prioritize these events without some sort of automation. That's where some of the newer, SaaS tools for modern operations management come into play. This sector of software has been growing quickly and attracted higher than ever volumes of funding in the past year, with startups such as AppDynamics and Splunk making a splash with investors.
Why look at events?
But let's get back to the central question: why should IT and the business give a hoot about the arcane infrastructure event?
Think about the number of applications companies are managing today versus 10 years ago: it's likely more than twice the number, plus all the data coming from both virtual and physical machines, mobile devices and even multiple hosting providers. With the cloud, event management has become all the more complex. If you want to understand how your IT environment is performing at any given time (especially the revenue-generating components such as your public website and customer facing apps), you've got to first get a handle on these events.
Companies can't afford downtime on their sites, nor poor customer service from an agent that can't open the CRM application while a customer hangs on the line, nor a mobile app that continually goes offline, causing service technicians to be in the dark about their daily schedule. Those capabilities are table stakes; without reliable IT, there's no room for any innovation much less mind-blowing market disruptions.
Now as for the CIO, he or she has got to pay attention to events and the practice of event management, because if they don't know how to properly manage this new virtual, hybrid cloud environment, they're out of a job. Users are much more in tune today; they begin to complain the instant an app or site is down. They will certainly start tweeting about it, if the CIO doesn't catch it and fix it pronto. Here's the other scary thing for CIOs: if they can't fix response time and application issues in due order, newly IT savvy business managers will run out and provision servers somewhere else with a new app, thank you very much.
This transition from an app perspective to an event perspective is not a big deal for IT workers in the trenches. In fact, it makes a heck of a lot more sense to them. IT organizations have long struggled with understanding all the applications the company is using. Focusing on events is in some ways easier and also provides a better analytic, allowing a more detailed view of what's going on across multiple servers and tiers.
By focusing on event management, IT obtains a high-level picture of your entire environment to effectively manage risk. They can also drill down when they need to diagnose an issue. Organizations that do not adequately understand, monitor and manage events, run the risk of alienating customers and the business alike, with poorly-performing, unreliable systems. Putting an eagle's eye on IT events is less about preventing disasters, than it is about keeping users and end customers happy. Can you afford not to change your IT Ops viewpoint?
About the author: Clayton Ching is Director of Products at Boundary.