Cortex, a Prometheus Spinoff, Boosts Data Monitoring
An open source monitoring system released this week by Grafana Labs is compatible with and inspired by the de facto standard Prometheus machine data storage and monitoring platform for cloud-native deployments.
Grafana Labs help launch development of the Cortex monitoring system in 2016, graduating to a Cloud Native Computing Foundation (CNCF) “sandbox” project in 2018. The stable version of Cortex aimed at enterprise customers allows them to query metrics from Prometheus servers in a single place, eliminating gaps in graphs due to a server failure.
Cortex also stores Prometheus metrics that can be used for applications such as capacity planning and performance analysis. (Prometheus was the second project accepted and “graduated” from CNCF after the Kubernetes cluster orchestrator.)
New York-based Grafana Labs currently uses Cortex in production to monitor its Prometheus cloud backend used to deliver the data visualization and monitoring vendor’s managed logging and metrics platform.
Among the vendor’s contributions to Cortex are improved query performance, a horizontally scalable alerting and rule evaluation service and backends for Google Cloud Storage and its Bigtable NoSQL database service designed for analytics workloads.
“Cortex is more for enterprises that want a centrally managed monitoring experience for their internal teams,” Grafana Labs’ Goutham Veeramachaneni noted in a recent blog post. “It’s for companies that have 25 million or more series across all their clusters and want to store that data for years.”
Cortex Version 1.0 is available now.
Along with Prometheus and Loki, Grafana Labs’ cloud service offers Graphite, another open source monitoring platform.