Key takeaways from AppSphere15
Software everywhere or software nowhere?
“Software is increasingly everywhere”, Kirkpatrick explained, “but it’s so seamless that you don’t even see it. You just enjoy new efficiencies and ways of getting things done”.
His insight reminded me of a phrase that we often say at BigPanda: “every company is a tech company”. And it’s true. I challenge you to think of a company – regardless of size or industry – that doesn’t rely on technology to deliver delightful customer experiences. In this day in age, it is simply an inherent fact that a large part of your success hinges on the strength of your technologies. And so it follows that when we talk about uptime, MTTR, continuous delivery, and agility – all of the buzzwords related to IT operations – what we’re really talking about is increasingly key competitive differentiators, and drivers of customer retention and loyalty.
The challenges vs. the desires
It may come as little surprise that a number of common themes emerged amongst the keynotes and sessions. In fact, many of the speakers’ talks underscored pain points and goals that we’ve heard before, whether at other conferences or amongst our customers. Harlan Hagewood, IT Manager for Application Service and Performance Management at Travelport did a great job summarizing some of these key challenges in his talk, “Turning to Unified Monitoring and Real-time Application Analytics”.
Harlan explained that the increasing complexity of apps used by Travelport, married with their use of cloud infrastructures, microservices, and third-party APIs, had amplified deployment challenges, made DevOps collaboration next to impossible, and created a state where performance issues took far too long to resolve. “We need a better way to identify a problem quickly,” Harlan explained, “without 50-60 people needing to jump on a call, all pointing their fingers in different directions.”
What Travelport aimed to achieve also reflects many of the shared visions we’ve heard in talks, from customers, and from our chats with others at the show. Namely: deep and wide visibility into application performance, the ability to quickly identify performance issues and to fix the right problem the first time, and the power to see, act and know with less time and effort. And then on the business side: the ability to measure impact through defined KPIs, to deliver faster by reducing defects and speeding up release cycles, and ultimately, to improve the customer experience.
Monitoring isn’t a nice to have, it’s a must have
The scene set by Harlan’s talk unscored the very reason why monitoring is so critical. As he put it, not only did Travelport have a whole host of different things that needed to be monitored, but “we knew we needed to fix our blind spots through correlation.”
This sentiment was echoed by Paul Pantazis, Manager of IT Ops for Entertainment Partners. Paul was clear: “we didn’t get a monitoring solution because we felt like it, we had to.” Much like Harlan, he underscored that the critical nature of monitoring hinges on the fact that IT isn’t only working to please internal stakeholders. Being held to customer SLAs means that downtime or degraded performance has a real and measureable business impact that IT teams are responsible for.
Success is born of the right tools, executed in the right way
So we find ourselves in an IT world that is increasingly noisy, but the stakes have never been higher. What do we do? The theme at the show appeared to be this: It’s about finding the right tools, and executing them in the right way.
Paul’s story of setting up what he referred to as a “pseudo NOC” was excellent reflection of this: He explained that he questioned the need for the “average IT shop” to spend copious amounts of cash to develop a traditional operations center. “There were already investments, I realized, that could get us most of the way there.” In his trademark dry humor he added, “so we bought some APM solutions, configured some stuff, bought a few TVs (because you have to have TVs)…and we had a pseudo NOC.”
So in short, it’s not about the number of tools that you have or the grandeur of your setup. It’s about investing in the right solutions and integrating them in the right way. Proper integration is not only a question of correctly configuring each system, but also of ensuring that your tools work together in harmony. So as you build out your toolset, consider how you’re going to handle alerts from multiple systems and how they will integrate into your workflows. Defining a strategy from the outset will not only help you avoid noise and chaos down the road, it will also place you in a position of being able to identify, manage, and resolve critical issues faster – protecting those SLAs, and even more important, your reputation.
Luck is on your side
Missed us at AppSphere? You’re in Las Vegas luck: we’re extending our conference offer to get a free drone to 12/18. Find out how to get yours.