An eye on observability for February 2022

First up is a fantastic post from the team at Aspectio, giving an excellent overview of what distributed tracing is. It covers how it compares to other tools you might use and why and when tracing could be useful to you.

If you’re interested in digging into more detail, read about one of the best-known tracing tools, Zipkin, in this post from Julia Bellmer.

Finally, OpenTelemetry is one of the many rapidly coalescing standards in the observability space. Giulia DiPietro penned a solid introduction to the operator component (that actively collects metrics). The folks at Timescale created a comprehensive post on using OpenTelemetry with trace data to get you started with the new standard on the block.

Sanjiv Khosla posed a common question in this recent post.

Organizations have invested heavily into their observability infrastructure, yet they fail to respond to outages promptly. Why does it happen? What are the root causes?

The post reminds us that merely collecting more data isn’t enough and that you need to do something meaningful with it, be that analysis or action.

Prometheus may be the de-facto option (and standard) for ingesting and analyzing metrics, but it’s not the only one. SoundCloud recently announced the full open-sourcing of their exception-monitoring service, Periskop.

Optimized for monitoring application exceptions, it has a scraping endpoint, UI, and client libraries for scala, go, python, and ruby.

We’re big fans and users of Temporal’s scalable cloud application solution (watch out for a blog post on how we use it soon). Because of that, we were very pleased to see their series B announcement totaling $128 million. Go team!

It wouldn’t be a tech news roundup without a small handful of takeover news, but there’s a definite and a “maybe” this month.

Akamai, a company you hear little about daily, but whose services you likely use more than you realize, announced a takeover of Linode. Akamai plans to use Linode’s expertise in building developer-friendly platforms in tandem with their massively-distributed network to provide a compelling offer in terms of reach and usability. Time will see if the combination works, but it certainly makes sense.

The Wall Street Journal cites the ever-present “people familiar with the matter” regarding Cisco apparently making a $20 billion bid for Splunk. Nothing is confirmed, but that would firmly bring observability into the corporate world.