Gumbo Podcast: Your First Steps in Cloud-Native Observability

Last week, I was invited to share my experiences on the Data Protection Gumbo podcast with its host Demetrius Malbrough and we had a nice chat about my journey into cloud-native observability (o11y) since joining Chronosphere.

The title of episode 171 is "Your First Steps in Cloud Native Observability." We covered questions surrounding the series of articles I've been writing called the O11y Guide, which takes you along on the journey as I go from an application development background into the world of cloud-native o11y.

Below is a list of the topics we covered in the 20+ minutes of the podcast, which you can find on all major podcast platforms.

A podcast is always interesting, as you join a virtual call on video, then record the session. In this case, the video was used for a small clip to promote the episode going live and was shared on social media channels.

The host, Demetrius, had stumbled upon my blog series called the O11y Guide and wanted to have a chat about my experiences around the observability world and how that applies to the data angle of his podcast.

Without giving a full transcript of the episode, I thought it might be nice to preview some of the questions that were asked. I'll leave it up to you the reader to then jump over to the podcast for a listen!

We started off with a look at what cloud-native observability means to the listeners, and I gave a bit of background on how I was sharing my journey from application development to observability. This is the foundation of my O11y Guide series.

Next, the transition is made by asking what the 3 pillars of observability are and whether they are really relevant to modern cloud-native observability. I spent some time talking about how you can't just hold on to the technology views in monitoring and expect to be able to massively scale in a cloud-native world. This requires a business view of how to achieve your goals without focusing on the technology used. We cover the 3 phases of cloud-native observability: know, triage, and understand. These involve the process used by observability teams across the globe to solve problems in their cloud-native environment and focus on outcomes rather than underlying data formats (logs, traces, metrics).

Demetrius then asked what data has to do with cloud-native observability, and I shared some thoughts on the massive data being generated in cloud-native platforms at scale and how that's becoming the biggest roadblock to organizations succeeding in the cloud. My answer to this question pointed out a very shocking fact: organizations are paying more for their observability data than for their income-generating business application and customer-generated data.

Next, we covered what a sample application in a cloud-native environment would generate as an observability tooling target. This led to a discussion on average data retention rates and the cost of unused data that are being collected and stored in the cloud.

He then asked me: If companies are paying to retain all their observability data, and if so, what can be done to reduce some of those costs? I shared an architectural overview of how you can use Chronosphere's product, which includes a control plane, to give users the ability to control the amount of data after ingestion, but before storage to save on unused data collection. I mentioned that this can save around 50% on data storage using this control plane.

We covered the players and teams we are interacting with within organizations, such as SREs or observability teams. We then touched on how organizations start out on open source observability tooling where open standards compatibility is very important but find out that it's very complicated to maintain and scale in the cloud native world.

Finally, we dove into the books I'm reading, so tune in to find out!

Click here to listen in your browser

Also worth mentioning: If you have something interesting to talk about in the data world, reach out to Demetrius to possibly join his podcast!