When Prometheus Can't Take the Load Anymore

Explore a comprehensive analysis of scaling Prometheus for high-load environments in this conference talk. Discover the journey of Riskified's SRE team as they encountered limitations with their initial Prometheus setup and embarked on a quest to find the optimal solution for multi-cluster, high-availability, and long-term metrics storage. Compare the features, advantages, and disadvantages of Thanos, Cortex, and M3 as potential alternatives. Gain valuable insights into the performance, cost-effectiveness, and operational aspects of each tool, enabling you to make informed decisions for your own use case. Learn about the Cortex architecture, its advantages, and potential drawbacks. By the end of this presentation, acquire a deeper understanding of advanced monitoring solutions and their applicability in cloud-native environments.