Why Sumo Logic Embraced the OpenTelemetry Standard

The big news from Sumo Logic, coming out of KubeCon+CloudNativeCon Europe 2023 in Amsterdam this week, is that the observability company is going all in on OpenTelemetry, according to Erez Barak, Sumo Logic’s vice president of product development for observability.

OpenTelemetry is an open standard for generating, collecting and exporting application telemetry across any number of backend systems.

“It’s the most exciting part I think of all our announcements here is strongly leaning into the world of open telemetry, and really being — I believe from what we see — the leading and most groundbreaking enterprise vendor in terms of leveraging OTO,” Barak said. “[We’re] really, frankly, betting the farm on it because our customers are betting the farm on it.”

OpenTelemetry Use Avoids Vendor Lock-in

OpenTelemetry also solves another problem with observability platforms: Vendor lock-in, according to Vera Reynolds, a senior telemetry engineer with Honeycomb.

“In order to provide richer diagnostic data than traditional monitoring tools, app maintainers would add vendor-specific agents or libraries to their projects,” Reynolds wrote in an article for The New Stack. “While each vendor provided some preconfigured baselines, maintainers would still need to manually add their own instrumentation to ensure that the insights reported to their vendor’s dashboard were relevant to their specific use case. This instrumentation would be added in that vendor’s specific parlance, making it non-transferable to other implementations. The result is upfront time savings when getting started — but that time will inevitably be collected back with an order of magnitude in accumulated interest if you decide to change tooling providers.”

Using OpenTelemetry standards allows Sumo Logic to work without the agent, Barak said. It’s being rolled out with a new customer onboarding process, which the company announced Tuesday, but OpenTelemetry now works all the way into distributed tracing and span analytics of your application.

“What we’re also announcing is take every experience we have big and small … and making that not only OpenTelemetry compliant but OpenTelemetry native,” he said.

The new customer onboarding has streamlined the overall onboarding workflow for new customers and users, resulting in faster times to insights, the Sumo Logic announcement stated.

“Those workflows sometimes introduced friction because there’s new customers coming in,” Barak told The New Stack. “So we reworked those workflows and are only basing them on their ability to bring in OpenTelemetry, which is what our customers are telling us. Same thing for new customers. … Now, first thing I do typically is send in some data, start my analysis, start growing the different usage patterns that we put in place we don’t. What we’ve done is build that in alignment with OpenTelemetry. ”

Onboarding New Users with OpenTelemetry

The company also announced that its Sumo Logic Distribution for OpenTelemetry Collector now supports Windows platforms. Customers using Sumo Logic Distro for OpenTelemetry can gather logs, metrics, and traces from Windows operating systems, which can be configured utilizing the new onboarding workflows, the company stated in its announcement. It already supported MacOS and Linux platforms.

The need for a single, OpenTelemetry-based collection strategy is important for application observability and infrastructure monitoring, the press release noted.

“Onboarding data and setting up data collection from various sources can be complex,” Sumo Logic stated. “The hidden costs include the need to juggle disparate backends or libraries and be limited by vendor or technology lock-in. Through a single app installation and workflow for logs, metrics and traces, Sumo Logic makes it easier for developers to harness data and takes the complexity out of deploying OpenTelemetry as a collection strategy.”

Automating with a Code-First Approach

The company also unveiled its new Predict for Metrics functionality. Predict for Metrics uses linear and autoregressive models to make predictions by harnessing past data points to predict future trends, the company explained in its announcement. It is a metrics query language operator, which allows users to visualize forecasted values and add resulting charts to Sumo Logic dashboards, according to the company.

Predict for Metrics is designed to control production issues, system downtime, and cloud costs. It ties in with a code-first approach with applications and infrastructure that is leading to more automation and for DevOps, Barak explained.

“The idea behind Predict for Metrics is to provide a comprehensive way to really harness observability analytics capabilities, and at the end of the day, be able to predict some change in an application, in the cloud infrastructure, whether it’s about their usage or their resource demands,” he said. “What they’re looking to do is take their code posture, extend it with this ability to predict and again, use what you get in terms of automation and compliance when your code-first so when you look at the world of code-first we’re saying hey, it’s not just about configuration. I think that’s the main message here: It’s also about those advanced AI and analytics capabilities. Those have to be code-first as well, because they come in part and parcel of every configuration, and every setup our customers have.”

As an example, he pointed to a recent customer who was reading the log in a native manner and received 97,000 results. That’s a big number, but it doesn’t really provide the impact and insight the company needed, Barak said.

”What they need to do is use capabilities or AI-powered capabilities, like I mentioned before, like log reduce, what does log reduce do for us? It takes all the signatures compares those patterns, and then builds based on existing models, a view that’s more actionable for the customer. So that’s just one example of how we use AI in the context of all that data,” he said.