thanos.io

What It Is

Thanos is an open-source, highly available Prometheus solution designed to preserve compatibility with the Prometheus ecosystem while adding unified multi-cluster querying and long-term metrics storage. Its official website describes it as a “Highly available Prometheus setup with long term storage capabilities” and lists it as a CNCF Incubating project.

Core Capabilities

From a feature perspective, Thanos focuses on a Global Query View, allowing metrics to be queried across multiple Prometheus servers and clusters. This makes it suitable for multi-cluster setups, hybrid environments, and large cloud-native platforms. It also provides Unlimited Retention by storing metrics data in object storage, with support for GCP, S3, Azure, Swift, and Tencent COS. In terms of compatibility, Thanos supports the Prometheus Query API, so existing tools such as Grafana can continue to be used. For historical data, it offers Downsampling & Compaction, which can improve query speed over large time ranges and enable more advanced retention strategies.

Pricing and Deployment

No commercial pricing is provided in the main website content. The project is explicitly open source and offers entry points such as GitHub, Quay.io, DockerHub, and Download, indicating a stronger focus on self-hosted deployment. The website also includes a Support and Training entry, but it does not provide pricing, service scope, or SLA details, so information on commercial support is limited.

Pros and Cons

Its advantages are that it fits closely with the Prometheus/Grafana ecosystem and does not force users to replace their existing monitoring stack. Its object storage support is broad and includes Tencent COS, which is somewhat friendly to China-based cloud environments. As a CNCF Incubating project, it also has a solid community and ecosystem foundation. The limitations are that the official website’s main content does not go into detail on architecture, deployment complexity, or operational requirements. For small single-cluster Prometheus setups, Thanos may feel relatively heavyweight; for large-scale scenarios, teams will need operational expertise in Kubernetes, object storage, and monitoring systems.

Who It’s For and Access from China

Thanos is suitable for SRE, platform engineering, and cloud-native infrastructure teams, especially organizations already using Prometheus but running into issues with multi-cluster querying, long-term retention, and historical query performance. Access from China is not mentioned in the main content and cannot be confirmed. Alternatives include Cortex, Mimir, VictoriaMetrics, and Prometheus remote storage solutions.