The story of the evolution of Observability to encompass APM and other IT performance management capabilities would not be complete without discussing the monumental impact of open source.
Start with: APM and Observability - Cutting Through the Confusion - Part 8
Open source is transforming how organizations approach APM and observability by providing vendor neutral standards for collecting and exporting telemetry types, says Mimi Shalash, Observability Advisor at Splunk, a Cisco Company.
Solutions like OpenTelemetry simplify integration across platforms, reduce vendor lock-in, and improve interoperability in complex environments, Shalash continues. Prometheus enhances this approach with robust metrics and alerting, especially systems like Kubernetes. And together these tools enable flexible, cost-effective stacks designed to scale and evolve with modern infrastructure.
“Open source tools like OpenTelemetry and Prometheus are becoming essential building blocks for observability in modern, cloud-native environments,” explains Andreas Grabner, Fellow DevRel and CNCF Ambassador, Dynatrace. “They empower organizations with greater flexibility and standardization in how telemetry data is collected. The broader industry trend is moving toward interoperability and data unification—using open standards for collection while relying on more advanced platforms to contextualize, analyze and act on that data at scale. This hybrid model allows teams to preserve their existing investments in open source while benefiting from automation, AI and enterprise grade observability.”
“The observability space is a prime target for OSS,” Sven Delmas, VP of Research at Mezmo, agrees. “Between dealing with a tech-savvy and curious audience, constant pressure on cost control, and the need for transparency and avoiding vendor lock-in, there has been — and will be — an ever-increasing push to OSS.”
Driving Observability's Evolution
Open source is changing the center of gravity in observability from tools to telemetry, according to Brian Douglas, Head of Ecosystem, Cloud Native Computing Foundation (CNCF). Developers are adopting Prometheus, OpenTelemetry, and Fluent Bit not just because they're free or flexible, but because they represent an open, portable foundation. These tools make it easier to switch vendors, build internal platforms, and innovate on top of shared standards. They're not just part of the observability conversation; they're shaping the future of how observability is defined.
APM is one specific implementation of observability, not its full scope, Douglas continues. It answers questions like, 'Is this app performing within expected parameters?' Observability, in contrast, supports deeper exploration: 'Why did latency spike in a downstream service for certain regions?' Projects like Prometheus and OpenTelemetry enable this broader context by collecting high-dimensional metrics, distributed traces, and logs which gives teams the raw, interoperable data needed to connect the dots.”
Observability supports cross-signal correlation and open-ended investigation, Douglas adds. Rather than focusing solely on applications, it lets teams visualize the full stack, from container runtimes and infrastructure to network topology and business-level SLIs.
- Prometheus provides robust, flexible metrics, while Cortex scales them across environments.
- Fluent Bit and Fluentd handle log aggregation and routing across edge and core environments.
- OpenTelemetry standardizes telemetry collection and enriches it with context, making it easier for tools and teams to interoperate without reinventing the wheel.
“What's important is interoperability,” Douglas of CNCF explains. “With standards like OpenTelemetry and protocols like the Prometheus exposition format, teams can adopt a modular approach: instrument once, analyze anywhere. This lets them use best-in-class components rather than be locked into a monolithic solution. Observability isn't a single tool, it's a strategy backed by open, composable tooling.”
With OpenTelemetry, users can build a composable observability stack where each tool plays to its strengths: one might excel at exploratory debugging, another at automated root cause analysis, and a third at cost-effective long-term storage, Severin Neumann, Head of Community & Developer Relations at Causely, elaborates. This flexibility lets teams get the best outcomes for their specific needs without duplicating instrumentation or locking themselves into a one-size-fits-all solution.
OpenTelemetry: Reshaping APM and Observability
OpenTelemetry, in particular, has already profoundly reshaped the APM market and the broader observability field, explains Juraci Paixão Kröhling, Software Engineer at OllyGarden. “Initially, some established players might have overlooked it, but strong customer demand has made OpenTelemetry support almost table stakes now; it's rare to find a vendor unable to ingest the standard OTLP format.”
OpenTelemetry is an open source standard, framework and suite of tools facilitating the generation, collection, and exporting of telemetry data.
“OpenTelemetry is having a huge impact on the industry with studies showing that nearly half of organizations polled are using OpenTelemetry with another 25-percent-plus looking to adopt in the near term,” says Harald Burose, Director, Product Management, Research & Development – Engineering, OpenText.
Kröhling from OllyGarden continues, “I expect vendors will increasingly embrace OpenTelemetry more natively, treating its semantic conventions not just as data points but as first-class citizens for richer understanding. The era of requiring proprietary agents for basic data collection is closing; customers now expect tools not only to handle open formats but to do so meaningfully, respecting the common language defined by standards like OpenTelemetry. This shared foundation allows everyone to cultivate better systems.”
OpenTelemetry provides teams with greater flexibility, standardization, and control over their telemetry data and has become the de facto standard for data ingestion, according to Bahubali Shetti, Senior Director, Product Marketing, Elastic. Whether deployments use standard OTel SDKs, auto-instrumentation, OTel Collectors, or a combination of these, users can avoid vendor lock-in and reduce the need for future retooling.
“OpenTelemetry isn't just shaping the future of observability, it's quickly becoming the standard that modern, scalable systems are built on,” concludes Shalash from Splunk.
Go to: APM and Observability: Cutting Through the Confusion — Part 10, discussing AI's impact on APM and Observability.