Grafana Tempo

Grafana Tempo is an open-source, easy-to-use, and high-scale distributed tracing backend, developed by Grafana Labs. Released in 2020 and now with over 4,000 stars as of 2026, Tempo is designed to store and query distributed traces with minimal operational overhead. Unlike other tracing backends (like Jaeger or Zipkin) that index trace metadata for search, Tempo takes a different approach: it stores only trace IDs as indexes and keeps the full trace data compressed in object storage. This means search is limited to trace ID lookup, but storage costs are dramatically lower (often 10x cheaper than indexed backends) while supporting virtually unlimited trace retention. Key features include: trace ingestion via OpenTelemetry OTLP, Jaeger, and Zipkin protocols (accepting traces from any instrumented application or tracing agent), trace ID-based search (retrieving full trace data by trace ID with sub-second latency regardless of trace age), TraceQL query language (a purpose-built query language for finding traces based on span attributes, durations, and service names, introduced in Tempo 2.0), metrics generation from traces (deriving RED metrics - Rate, Error, Duration - from trace data without needing a separate metrics source), object storage backends including S3, GCS, Azure Blob, and local filesystem, high compression ratio using parquet columnar format (achieving 10-100x compression compared to raw trace data), microservices architecture with separate components for ingestion (distributor, ingester), querying (querier, query-frontend), and compaction (compactor), horizontal scalability for both ingestion and querying, multi-tenancy support via X-Scope-OrgID header, seamless integration with Grafana (visualizing traces alongside metrics from Prometheus and logs from Loki in unified dashboards), and service graph generation (automatically building service dependency maps from trace data).