WeAreDevelopers LIVE • Oct 30, 2024

Observability with OpenTelemetry & Elastic

Iulia Feroli

Are you still debugging complex Python microservices with native print statements? Discover how to achieve vendor-agnostic observability by combining OpenTelemetry's standardized signals with Elastic's powerful dashboards.

Pause
Mute Enter Fullscreen
#1 about 2 min

The increasing complexity of modern application observability

As applications and generative AI solutions become more complex, standardized monitoring processes are required to diagnose unexpected failures.

#2 about 6 min

Limitations of native debugging and logging in Python

Relying on print statements or the built-in logging module creates manual overhead and lacks compatibility across robust software systems.

#3 about 2 min

Defining OpenTelemetry as a standardized observability framework

OpenTelemetry provides a rapidly growing open-source toolkit to standardize data collection across all vendor ecosystems and programming languages.

#4 about 4 min

Core concepts of application instrumentation and telemetry

Instrumenting application code enables the continuous emission of telemetry signals that detail system behavior without inspecting internal mechanisms.

#5 about 5 min

Differentiating between metrics, logs, and distributed traces

Telemetry data breaks down into numeric metrics over time, discrete event logs, and sequence-based distributed traces for mapping request paths.

#6 about 3 min

Preventing vendor lock-in with an independent observability layer

Implementing an open standard separates data collection from specific cloud providers or analysis tools to ensure architectural flexibility.

#7 about 3 min

Implementing manual and automatic instrumentation in Python

Python developers can embed specific library calls manually or use an automated toolkit to seamlessly track common frameworks without altering core code.

#8 about 4 min

Integrating Elastic dashboards with OpenTelemetry data schemas

Routing standardized telemetry into Elastic's unified schema provides powerful search capabilities and comprehensive dashboards for advanced debugging.

#9 about 4 min

Live dashboard demonstration of tracing errors and system metrics

Running a containerized Python demo highlights how observability platforms surface latency bottlenecks, crash logs, and complex request transactions automatically.

#10 about 2 min

Future trends for monitoring complex software and language models

Embedding open-source observability practices simplifies long-term maintenance and model optimization for emerging machine learning applications.

Matching moments

3:31 min

Distinguishing between telemetry, monitoring, and observability

Juraci Paixão Kröhling Juraci Paixão Kröhling · World Congress 2026 Europe

12:33 min

Exploring advanced observability stacks and distributed infrastructure challenges

Pawel Piwosz · LIVE

2:01 min

Integrating OpenTelemetry pipeline platforms with embedded third-party observability providers

Nočnica Mellifera · LIVE

1:24 min

Importance of OpenTelemetry in front-end web applications

Carly Richmond Carly Richmond · Europe 2026 Virtual

2:15 min

Maintaining code quality with open standard observability and pipelines

Alex Laubscher Alex Laubscher +3 · World Congress 2025

17:03 min

Introduction to metrics and observability challenges in monitoring

Liam Hurrell · LIVE