I’m pleased to report that today, the Cortex project advanced from sandbox to incubation within the Cloud Native Computing Foundation after a vote from CNCF’s Technical Oversight Committee (TOC). The TOC’s decision is a signal that Cortex has stepped up in maturity, attracting not just innovators but also early adopters among enterprises. To achieve incubation, CNCF projects undergo due diligence and have to demonstrate a healthy level of adoption and community activity. The next and last stage is graduation, joining well-established projects such as Kubernetes, Prometheus, and Envoy.
When I started Cortex with Julius Volz four years ago, our aim was to build a scalable, highly available “clustered” Prometheus implementation. So much has happened since then. Cortex has seen adoption by large enterprises around the world – such as Digital Ocean, Electronic Arts, REWE digital, Gojek, Aspen Mesh, MayaData and Platform9 – who find its centralized architecture and native multi-tenancy well suited for the challenges of running Prometheus at scale.
At Grafana Labs, we use Cortex to power Grafana Cloud Metrics, our Prometheus-compatible metrics and alerting service. Grafana Cloud is the easiest, most scalable and fastest way to get started with Prometheus. We also offer support for self-hosted Cortex installations for some of the largest companies in the world.
The Cortex team recently hit the 1.0 release, marking wider adoption, stability, ease of use, and compatibility guarantees. Cortex has contributions from more than 79 companies and 93 individuals and 3k+ GitHub stars. Its eight maintainers represent four companies (Grafana Labs, Weaveworks, Microsoft, and Splunk); Grafana Labs employs five of them (Goutham Veeramachaneni, Jacob Lisi, Marco Pracucci, Peter Štibraný, and me).
The project continues to evolve and grow. Recent collaborations with the Thanos project (which also recently advanced to CNCF incubation) have brought Thanos’s object-based block storage to Cortex, reducing the number of dependencies and TCO. As part of the collaboration, Cortex maintainers built caching and performance improvements into Thanos.
The collaboration goes both ways, with Thanos now taking advantage of Cortex’s query parallelization, caching technology, and integration testing framework. We look forward to collaborating further with the Thanos team. Marco Pracucci, who is a maintainer for both Cortex and Thanos, gave a talk on this collaboration at KubeCon + CloudNativeCon EU Virtual this week with Thor Hansen from HashiCorp.
Cortex has a bright future with many interesting new features in the works, including downsampling and variable retention, improving scalability and isolation, support for exemplars, and much more.
If you’re interested in trying out Cortex in your organization, sign up for a free 30-day Grafana Cloud trial.