improbable-eng/thanos
Thanos
Thanos is a CNCF cloud native monitoring tool that provides a highly available and scalable extension to the Prometheus ecosystem. It functions as a global query engine, a long-term storage system, and a metric downsampler.
The project enables a unified interface to aggregate and query metrics across multiple distributed clusters from a single view. It maintains historical data beyond local retention limits by persisting time-series metrics in object storage and eliminates data gaps by merging metrics from redundant server pairs.
The system includes capabilities for reducing the resolution of historical data to accelerate queries over massive datasets and distributes query requests across multiple endpoints to ensure continuous availability.
Features
- Distributed Query Engines - Provides a unified interface that aggregates and queries metrics across multiple independent distributed storage locations.
- Prometheus-Compatible Backends - Provides a scalable, long-term storage backend for Prometheus metrics using cloud object storage for unlimited retention.
- Time-Series Downsamplers - Creates lower-resolution versions of historical metrics to accelerate queries over massive datasets.
- Time Series Data Storage - Provides scalable storage solutions optimized for maintaining historical records of numerical performance data.
- Metrics Storage Backends - Configures external cloud object storage as a durable backend for high-volume time-series metrics.
- High Availability Architectures - Eliminates single points of failure by merging redundant data streams to ensure continuous service availability.
- Object Storage Persistence - Offloads long-term time-series data to remote cloud object storage to ensure durability and infinite retention.
- Global Query Interfaces - Merges data from multiple distributed servers into a single unified view for global metric analysis.
- Metric Aggregation & Downsampling - Processes time-series data through downsampling to optimize long-term storage and query performance.
- Monitoring Tools - Provides a cloud-native monitoring solution that scales and extends the Prometheus ecosystem.
- High Availability Clusters - Implements a distributed architecture that ensures service continuity by merging metrics from redundant server pairs.
- Global Prometheus Querying - Provides a unified view by merging metrics from multiple distributed Prometheus clusters for centralized monitoring.
- Metric Downsamplers - Aggregates high-resolution metric data into lower-resolution intervals to optimize long-term storage and query speed.
- Global Query Engines - Functions as a unified global query engine that aggregates metrics from multiple distributed clusters.
- Metric Deduplicators - Implements metric sample deduplication to collapse redundant data points from multiple storage replicas for storage accuracy.
- High-Availability Request Routing - Manages cluster-aware request routing to ensure query availability during individual component failures.
- Load Balancing - Distributes incoming query requests across multiple engine instances to prevent bottlenecks and ensure high availability.
- Push-Based Metric Pipelines - Uses a sidecar process to actively push local metrics to a central remote store.
- Time Series Databases - Components for highly available metric systems with unlimited storage.
