Since I could walk, spending time on my Dad’s cycle has been one of my fondest memories. My love for motorcycling has grown over the years to become one of my favorite ways to relax and enjoy some quiet time – as quiet as a Harley can be, I guess. Much like my love for cycling, my love for Application Performance Management (APM) has also grown.
As a pre-sales advisor at CA specializing in APM, I’ve learned a lot over the years – this blog is my attempt to combine my love of cycles with my love of APM. Yes, it’s love. There are actually several concepts that apply to both!
The First Try May Ne a Little Scary
Despite being a passenger on motorcycles my whole life, taking control and operating a cycle myself was scary the first time. What if I stalled? Popped the clutch too quickly? Crashed? Luckily, I had a perfectly smooth start!
Think about how that applies to deploying an Application Performance Management solution. What if you enable too many settings? What if not enough? What if it slows down or crashes your app? I guess I lucked out with my first APM install as well over ten years ago – smooth sailing there too.
In both cases, once you get through that first try, each “next time” becomes easier, your confidence increases, and your operator skills improve.
Remember, choosing the proper equipment for your task at hand is essential to ensuring that success. Your first rollout should be focused on initial visibility into your application – and maybe even just into end user experience. You can add visibility across app tiers as you become more comfortable but don’t go overboard your first time.
Training Is a Great Way to Get it Right, the First Time
You can find instructions online for just about anything these days. So, I knew the steps to operate a bike but working with an instructor enlightened me to details that make all the difference between a safe ride and an unpleasant ending.
Likewise, the steps for deploying APM are clearly outlined. The tasks involved are not difficult. The challenge comes in ensuring you are monitoring the right things, i.e. the components and metrics that your app teams care about most. “Actionable items” we like to call them – if a light turns red, does someone know which action to take?
Training provides that actionable insight as well as insight into the APM infrastructure itself, including how to achieve optimal performance, how to make adjustments if the need arises, and how to avoid impacting your applications. One of the easiest ways to become an expert is to work with an expert.
Practice Makes Perfect
Mastering any new skill takes time. Repetition is a great way to build confidence and repeatability.
For me, I frequently practice stopping and starting on a hill. It’s actually not complicated, but it does take practice to become second nature.
The same goes with Application Performance Management. Deploying and using APM is not difficult. You can use default settings, customize instrumentation, and even write your own plugins if you’d like. The more time you spend using APM in different applications, understanding what it can do for say a Java application in AWS versus a Node.js app running in Docker containers comes with practice and experience. Soon it becomes second nature.
Stay Focused to Avoid Harm and Don’t Take Anything for Granted
The first time I scraped a peg was scary. Learning just how low you can lean into a turn is pretty surprising and is based on not just the bike but the skill of the rider. Braking is another example. Some newer bikes come with anti-lock braking systems (ABS) but many do not.
In application monitoring, taking for granted that one size monitoring “fits all” apps can be equally dangerous. While the same Application Performance Management tool can be used for many different apps, your app teams will likely benefit from you staying focused on the unique attributes of each app environment to ensure optimal visibility and to avoid any harm or overhead.
APM tools can instrument apps like a profiler and in turn introduce performance overhead. The same tools can also run with minimal footprint, essentially “lost in the noise” of the operating environment, and provide critical visibility in a high-volume production application. Paying attention means you can navigate the right balance for your monitoring needs.
Don’t Be Afraid to Upgrade
My first bike was a Harley Sportster 883. Fate intervened and I now ride a Harley Davidson Custom Sportster 1200.
While there are many Application Performance Management vendors to partner with these days, there are only a few, when it comes to offering that right mix of ease of use, powerful capabilities, and the flexibility to extend and customize your monitoring, to suit your needs today, without holding you back in the future.
As I mentioned earlier, many APM vendors take a “one size fits all” approach to monitoring. We know that rule doesn’t work for t-shirts, obviously not for motorcycles, and certainly not for monitoring. Don’t make the mistake of sticking with an underpowered, entry-level application performance monitoring tool once your organization outgrows it.
Yes, that could mean migrating from an entry-level tool to an upgraded one that supports true enterprise-scale Application Performance Management. That may sound daunting. But you’ll know when you’re ready.