An open-source monitoring system with a dimensional data model, flexible query language, efficient time series database and modern alerting approach.

Prometheus stands as a cornerstone in the realm of modern observability, providing a robust toolkit for monitoring and alerting in dynamic, cloud-native environments. Delving into Prometheus, readers can uncover the intricacies of time-series data collection, metrics instrumentation, and scalable monitoring architectures. With Prometheus, readers gain the tools and knowledge needed to ensure the reliability, availability, and performance of their distributed systems.

Prometheus

Prometheus is shifting from storing composite metrics (histograms, summaries) as multiple primitive float samples to native composite samples in its TSDB. This architectural change improves efficiency by 30%+, enables transactionality across metric components, and simplifies the data model. The migration includes native histogram custom buckets (NHCB), upcoming native summaries and statesets, OpenMetrics 2.0 with composite values, and Remote Write 2.0 support. Compatibility modes will allow transparent migration by expanding native types back to classic representation for existing PromQL queries and consumers.

Modernizing Prometheus: Native Storage for Composite Types

Classic Representation: Primitive Samples