Data Observability Overview

Docs > Data Observability Overview

Cette page n'est pas encore disponible en français, sa traduction est en cours.
Si vous avez des questions ou des retours sur notre projet de traduction actuel, n'hésitez pas à nous contacter.

Overview

Data Observability (DO) helps data teams improve the reliability of data for analytics and AI applications and optimize the performance and costs of data pipelines. By unifying quality and jobs monitoring from production to consumption, teams can detect and remediate issues faster while optimizing cost and performance.

Datadog Data Observability end-to-end lineage with Spark job traces.

Key capabilities

Detect failures early: Catch bad data in warehouses like Snowflake, Databricks, and BigQuery through ML-powered monitors before dashboards, stakeholders, or AI models are impacted. Detect upstream pipeline failures in jobs run on Databricks, Spark, Airflow, or dbt.
Accelerate remediation: Triage faster using end-to-end lineage to pinpoint root causes, assess incident blast radius, and route to the right owner. View which job in the pipeline failed or was delayed, and pivot into job execution traces and logs to determine why.
Optimize cost & performance: Get visibility into the cost and efficiency of Spark and Databricks jobs and clusters, and use recommendations to optimize cluster configuration, code, and queries.
Unify end-to-end observability: Correlate data quality, pipeline execution, and infrastructure signals in one place, spanning the entire data lifecycle.