Sumo Logic Inc., maker of an analytics-based platform for application performance management and observability, today introduced a new reliability management feature of its Sumo Logic Observability platform that’s based on a service level objective methodology.

Reliability management focuses on the performance and reliability of applications from the end user’s perspective rather than monitoring infrastructure. That’s a fundamentally different approach from that of most observability platforms, which analyze interactions among services, networks, databases and other application elements to identify the source of slowdowns and outages.

“We look at reliability not from the context of a CPU or middleware or storage but how end users are doing,” said Erez Barak, vice president of product development for observability at Sumo Logic. “We believe that’s the only heartbeat that matters.”

User-driven analytics

SLOs are defined by the business and refer to certain milestones and thresholds that define acceptable performance, such as a login failure rate of less than 1% or a dropped shopping cart rate of less than 50%. “These insights help keep business leaders out of the weeds and inform their technology decisions based on business metrics,” Chief Strategy Officer Bruno Kurtic wrote in a blog post.

Sumo Reliability Management is the first commercial iteration of slogen, an open-source tool the company created to generate SLO dashboards, monitors and service-level indicators based on configurations derived from OpenSLO, a project intended to define open specifications for SLOs.

“Slogen allows you to, using a code-first approach, define your SLOs and then it takes those definitions and applies them using the Sumo back end,” Barak said. “Slogen creates the entities, SLOs, thresholds and dashboards within Sumo.”

Experience measurement lacking

Sumo Logic’s platform uses artificial intelligence to recognize patterns, align connections and move up and down the infrastructure stack working from the user perspective. The company conducted a survey of 3,500 information technology professionals on LinkedIn and found that most said user experience is vitally important to them but they don’t have a consistent way to measure it, Barak said.

Sumo monitors user sessions to look for anomalies such as slowdowns and errors and works back from there to identify the services and applications in play. The intelligence “comes from real users. It’s not proxied,” Barak said. “When you click on an app and it’s observed by Sumo we’re looking not at how you are tied into the infrastructure but rather what you’re trying to do and from that connect it to SLOs.”

The company said its approach enables operations teams to gather real-time reliability and performance metrics for data-driven decisions while also providing proactive alerts on SLOs and error budget consumption. Teams can monitor SLIs based on existing Sumo Logic queries and, when combined with the Terraform open-source infrastructure-as-code management, tool, effectively manage service levels as code.

Its toolset also enables system reliability engineering teams to uniformly adopt concepts such as SLIs, SLOs, service-level agreements and error budgets and apply them to reliability management as a business management discipline.

Photo: Sumo Logic