What we’ve learned about observability in 2019

The pillars of observability are crumbling

2019 was a banner year for observability. To increase the security and quality of their products, both SecOps and DevOps sought out observability of their systems in growing numbers. But gaining observability is like trying to hit a moving target. Even since the beginning of the year, there have been changes to how we understand it.

To look at the current state of observability, we look at the four biggest insights we’ve discovered about it in 2019.

1) The “three pillars of observability” are outdated

The traditional view is that if you have the three pillars of observability — logs, metrics, and traces — you have observability and you can understand all you need to know about your system. But really these are just broad starting points for beginning to find the answers to your questions.

Real observability means understanding your user’s experience and being able to answer your own business questions. It means applying tools to understand the meaning behind your logs, metrics, traces, and other event data available from your system.

If your organization is just beginning to seek observability, gaining access to comprehensive system data is key, but you cannot simply capture data and claim observability. You need tools that let you search your data, and ask questions both in real-time and from historical data as you find out what is going on in your system.

2) You can’t buy observability

According to Grant Schofield, Head of Infrastructure at Humio, “you can’t purchase a tool to solve and answer your observability questions. You need to use the tools to answer your specific observability questions.”

How you apply the tools and how you build a culture of using them within your organization is what really matters. Choosing a tool that you want to engage with is important, as well as choosing a tool that adapts to your needs. Humio meets both of these needs with an easy-to-use interface and customizable dashboards. Humio engineers can help you adjust it to suit your use cases.

3) Distributed traces are important

The proliferation of microservices and cloud services has led to an increased need for distributed traces to discover the root causes of problems in multi-server varied environments. The rise of service meshes like Jaeger makes distributed traces even more appealing because it allows users to run distributed traces without adding libraries to their applications.

Humio is able to interface with service meshes and give users a streamlined way to visualize their trace data.

One of our Fintech clients, Sparebank, developed an integration that allowed Jaeger to query Humio.

4) People want one platform to rule them all

In an attempt for more observability, many users are putting all their tools on one platform. Because of this, the choice of which platform you use becomes even more important. If you choose a traditional index-heavy option, you may end up actually sacrificing observability because it will struggle with real-time searches and you’ll experience delays in responding to issues.

Humio handles historical and real-time searches with ease, and integrates with a variety of other services, making it an outstanding candidate for building a one platform system.

Humio gives you deeper access to your data

Humio encourages you to collect all relevant event data, and lets you interactively explore your data, find answers to questions, and reach observability of every component of your infrastructure. Its easy-to-use search language and lightning-fast ingest speeds allow you to search your entire system and answer questions you’ve been trying to ask for months in seconds. It can produce live dashboards you can share with other teams such as sales and support. Humio’s 5-15x data compression allows you to store more and search deeper into your history, looking back months instead of weeks. Real-time alerts help your teams respond immediately to errors and intrusions, reducing downtime and saving you money.

Humio is an adaptable and accessible tool that can meet the ever-shifting needs of observability. Get a free 30-day trial, or schedule a demo to see how Humio can create an impact within your organization.

Start your free trial now, available On Premise and Humio Cloud, or request a demo.