A unified destination, or all-in-one, combines observability backends and frontends into one logical product. It stores and queries telemetry while providing interfaces for dashboards, alerts, and ad hoc investigation. Its frontend is designed for the provider’s backend rather than independent use. section 6.0
Unified destinations are useful when teams want an integrated observability workflow, but selection still requires practical trade-offs:
- Support multiple signal types and provide ingestion and query capabilities, sometimes through pluggable backends or nonpersistent in-memory storage. section 6.2
- Offer a close approximation of a single pane of glass by joining storage, querying, visualization, and alerting concerns. section 6.3
- Trade integration for coupling: self-hosting can add patching, scaling, on-call, security, storage, ingestion, and query costs. section 6.3
- Reduce vendor lock-in when they support open standards such as OpenTelemetry; open-source choices still require checks on project health, licensing, and governance. section 6.3
Typical unified destinations
CloudWatch, X-Ray, Azure Monitor, and Google Cloud’s operations suite fit the all-in-one category. Dedicated observability offerings such as Splunk, Datadog, Dynatrace, Honeycomb, New Relic, Lightstep, AppDynamics, and Sumo Logic commonly integrate frontends and backends. Open-source projects can also support two or more signal types. section 6.2 section 5.0
