High-Cardinality Infrastructure Metrics.
Sub-Second PromQL on ClickHouse.
Prometheus remote-write and native OTLP metrics ingestion with zero cardinality throttling. Execute PromQL aggregations across billions of time-series points at a fraction of legacy monitoring costs.
Deep Dive
Core Capabilities & Architecture
Explore how Metrics & Infrastructure instruments, processes, and correlates high-cardinality telemetry at scale.
Prometheus Remote-Write & Native OTLP Ingest
Point your existing Prometheus or OpenTelemetry Collector remote_write configs directly to aiAxonIQ with zero agent re-architecting. Buffered through partitioned Kafka streams with guaranteed delivery semantics.
- Full Prometheus remote_write 1.0 and 2.0 wire protocol compatibility
- Native OTLP metrics ingestion for OpenTelemetry Collector and SDK pipelines
- Kafka-buffered streaming ingestion with multi-tenant rate limiting and zero data drops
- Automatic discovery and schema registration of dynamic metrics and dimension labels
Familiar PromQL Syntax with ClickHouse Analytical Scale
Run standard PromQL range queries, rate calculations, and histogram percentiles directly on ClickHouse. Pre-computed 1-minute and 1-hour materialized view rollups keep multi-month queries sub-second.
- PromQL-compatible AST compiler translating expressions to optimized ClickHouse SQL
- Interactive autocomplete with real-time label index discovery and cardinality scoring
- Continuous 1-minute and 1-hour automated rollups via materialized view engines
- Configurable tiered retention policies per namespace, cluster, or tenant
High-Density Composite Dashboards with Zero Latency
Build enterprise monitoring boards visually with nine native panel types, or import standard Grafana dashboards via JSON with template variables for cluster, namespace, and service environments.
- Visual drag-and-drop builder with 9 panel types: time-series, stat, heatmap, gauge, and table
- Out-of-the-box production templates for Kubernetes nodes, JVM runtimes, and ingress gateways
- Multi-dimensional template variables supporting cascading regex label filters
- Secure public snapshot permalinks with tokenized read-only guest access
Built for every team that cares about reliability
One platform, tailored to how your team actually works.
Unified Multi-Cloud Host & Cluster Health
Monitor bare-metal servers, Kubernetes daemonsets, and serverless compute from a single unified control plane with zero blind spots.
High-Cardinality Dimension Freedom
Attach high-cardinality container IDs, git commit hashes, and tenant IDs to metrics without incurring exponential billing multipliers.
Real-Time Deployment Canary Signals
Track error budget burn rates, latency deltas, and automated rollback thresholds in real time during continuous deployment stages.
Columnar Time-Series Architecture Built for Cardinality Explosions
Incoming metrics stream into a distributed Kafka cluster and land into ClickHouse columnar storage partitioned by tenant and timestamp. Continuous materialized views pre-aggregate series into 1-minute and 1-hour summaries, ensuring long-term capacity trend queries execute in milliseconds regardless of raw data volume.
Start metrics monitoring in 5 minutes
Your existing OpenTelemetry instrumentation works unchanged, so there is nothing to rewrite and nothing to migrate.