As developments in artificial intelligence and kubernetes technology fuel industry-wide transformations, tracking the performance of AI models and cloud applications is vital.

“From an observability perspective, a lot of customers [are] obviously themselves experimenting with … generative AI, even building their own large language models, and so, they want to observe those workloads,” said Ian Smith (pictured), field chief technology officer of Chronosphere Inc.

Smith spoke with theCUBE industry analysts Savannah Peterson and John Furrier at KubeCon + CloudNativeCon NA, during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio. They discussed the impact of AI and Kubernetes on observability and the importance of open telemetry to a company’s data strategy. (* Disclosure below.)

Finding an observability strategy

With tools such as OpenTelemetry, a framework for capturing metrics and logs from cloud-native applications, Chronosphere offers customers greater flexibility for data analysis.

“OpenTelemetry … allows you to own your data and do things that you couldn’t do when you were sort of single vendor before,” said Smith, who likens advancements in AI to GraphQL, a data query and manipulation language. Both created a “massive explosion of data” that required detailed observability.

When it comes to observing the data from large language models and an increasingly complex architecture of cloud services, it’s not as simple as finding the right vendor, argues Smith.

“It’s a double-edged sword,” he said, “Observability strategy no longer can be just tool selection. And so, there’s a lot of choice out there, and I don’t think there are some hard and fast rules, but I think you really have to look at it through the lens of how does my tool selection feed into my observability strategy?”

The ratio of cost to value is a priority for crafting one’s strategy, highlights Smith. Instead of trying to “spend money infinitely,” companies should focus on how these observations serve their business.

As companies enter an era of AI and cloud-native innovations, being able to own their data and establish a concrete observability strategy is crucial, believes Smith. “They can be empowered. [Their] limited resources can be used more effectively.”

Here’s the complete video interview, part of SiliconANGLE’s and theCUBE’s coverage of KubeCon + CloudNativeCon NA:

(* Disclosure: Chronosphere Inc. sponsored this segment of theCUBE. Neither Chronosphere nor other sponsors have editorial control over content on theCUBE or SiliconANGLE.)

Photo: SiliconANGLE