4 min read time

Observability—From Theory to Practice

by   in IT Operations Cloud

This post explores what observability is, what’s driving it, and what it can do for you. You’ll learn some best practices for deployment too—from Micro Focus ITOM Practice Manager Jeff Tokar, who frequently leads customers along their observability journey. Our goal is to shine some real-world light onto this much-discussed but still-misunderstood term so that you can start putting it into practice.

Overwhelmed by Data?

Thanks to the rapid adoption rate of cloud-native applications, ITOps teams are being bombarded by more fast-flying data than ever. The speed, complexity, and vastness of the data is overwhelming. Most teams have no way to collect, process, or make optimal use of it.

That’s where observability comes in. Observability gives you the holistic visibility—and understanding—of the events happening across your IT estate. It brings order to chaos. It helps your organization run.  

What Is Observability, Really?

Industry pundits define observability in different ways, but it generally boils down to this: Observability is the ability to measure the internal state of a system or application by examining the data collected from it to detect problems faster. A system is considered observable if you can determine the behavior of the entire system from just the system’s outputs.

Observability is a two-part operation. The first part is about automatically gathering data from all the different sources that host and deliver business services and applications—no more manual data collection. The second is about gaining insights from that data to inform critical business decisions—for example, capacity planning or root cause analysis. Without observability, you risk flying blind when a disruption happens

What Observability Can Do

Observability is a hot topic because it helps customers:

  • Accelerate mean time to resolution.
  • Reduce event noise.
  • Find root cause faster.

But how do we put observability into practice? For the answer to that question, I interviewed our own Jeff Tokar about his experiences walking customers through the process of deploying observability. Here’s what he said.    

Are customers talking about observability as much as we are?

Observability is a key topic in the industry right now. But that doesn’t mean the industry is talking about it consistently. What we’ve found is that customers aren’t always clear what observability means or what it entails. That’s understandable because it means different things to different people.

What do customers think observability is?

It depends on who you’re talking to. On the DevOps side of the world, they look at observability in a different light than someone on the operations side. DevOps develops applications and wants insights into those applications as they are being developed and deployed. They also want to know how applications are performing and what they’re doing behind the scenes. ITOps teams have a broader definition of observability. They definitely want a deep look at the applications they are running, but this is not the complete picture.

In addition to the underlying elements of those applications, they want to know about the entire application ecosystem—the systems, events, and metrics that surround it and might have an impact on it. For network observability, that could mean understanding traffic flows, network performance, or network availability. Or it could mean knowing which servers host which applications and knowing what their performance and availability is. ITOps wants to see all these observability elements together to understand the health of the digital business and to speed up remediation should issues arise.

Is it safe to say a one-size-fits-all approach to observability doesn’t really work?

Definitely. As you talk to different people, they’re going to have different perspectives based on their specific infrastructure requirements for business services, applications, architectures, and deployment types.

Each organization has different roles and responsibilities regarding how data needs to be consumed so your observability solution needs to fit those requirements.

What is a common outcome of observability?

The most common outcome is the ability to understand the health of your applications. Because at the end of the day, it’s about collecting all your existing monitoring data to understand what is happening with your services and why—all with the aim of keeping digital business operating effectively and delivering for your customers.

Do you have any advice for companies about to embark on their observability journey?

One of the things we frequently hear about is a siloed approach to observability. For example, a cloud vendor might claim that they have observability capabilities, but they’re really focused only on your cloud estate. Or perhaps application tracing is being branded as observability. These are narrow pillars from which to hang the term. True observability comes from the structure built around these various pillars which gives you the complete insights necessary to understand the health of the digital business.

My advice is to focus on centralizing your siloed observability data and driving analytics across it all, rather than working on individual pillars of data. I think that's important and required to ensure the success of your digital business.             

To learn more about observability, check out these analyst reports:


Events

See all the Micro Focus events worldwide. 

Read all our news at the Operations Bridge blog. 

Related items 

Have technical questions about Operations Bridge? Visit the OpsBridge User Discussion Forum

Keep up with the latest Tips & Info about Operations Bridge. 

Do you have an idea or Product Enhancement Request about Operations Bridge? Submit it in the Ops Bridge Idea Exchange

We’d love to hear your thoughts on this blog. Comment below.  

Labels:

Operations Bridge