OpenTelemetry and Prometheus are getting along. What’s still missing?
Welcome to another edition of Road to KubeCon, where we’re tracking the major movements in the Kubernetes and cloud native The post OpenTelemetry and Prometheus are getting along. What’s still missing? appeared first on The New Stack .
In the latest episode of Road to KubeCon, the focus is on the progress being made in cloud-native observability. Two key monitoring tools, OpenTelemetry (OTel) and Prometheus, are working more effectively together. According to a 2026 survey of observability practices, nearly half of respondents use a combination of OTel-style and Prometheus-style instrumentation for infrastructure metrics, while 30.7% use both for application metrics.
The ease of use rating for the OTel-Prometheus combination has improved, with the average rating rising from 3.1 to 3.6, and the proportion of users finding the two tools hard to work together falling from 29% to 10%.
However, there is still room for improvement. Survey respondents are requesting better alignment between the projects' data models, improved handling of resource attributes and metadata, and fewer naming and formatting issues. Meanwhile, Atlassian has successfully migrated its metrics collection from gostatsd, an open-source Go implementation of Etsy's StatsD, to OpenTelemetry.
The team was able to execute this platform-team migration without changing the service-facing interface, resulting in significant performance gains, such as half the CPU usage for the same traffic and a 30% reduction in sidecar costs.
New Relic's 2026 Observability Forecast, based on a survey of 2,575 IT and engineering leaders and practitioners, found that 73% of respondents are standardized on, actively migrating to, or testing OTel. The report also highlights the importance of observability in AI adoption, with 83% of respondents considering it essential for AI-generated code.
Organizations that monitor AI agents are twice as likely to report a threefold return on observability investment compared to those running agents without monitoring. The study also reveals that engineers now spend 37% of their time addressing disruptions, and that outages cost organizations an average of $74 million per year, or $1.85 million per hour.
Written by urgent.news from The New Stack's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
