Observability

Metrics. Traces. Logs.
Correlated. Analytical.
Centralised.

An OTel-native observability stack on open backends — VictoriaMetrics for hot operational metrics, ClickHouse for long-term analytical intelligence — unified into one dashboard surface covering all four FalconIO features. No proprietary lock-in. No per-host pricing.

OTel Native VictoriaMetrics ClickHouse Analytics 60–80% vs Datadog

Two backends.
One reason: no compromises.

VictoriaMetrics handles your operational hot path. ClickHouse handles your analytical depth. They are complementary — each does precisely what the other cannot. Running both is a deliberate architectural decision, not an oversight.

VictoriaMetrics — The Hot Path
Real-time operational metrics. PromQL-compatible. Handles extreme cardinality without the operational burden of vanilla Prometheus at scale. Powers alert evaluations, KEDA scaling inputs, and live Grafana dashboards. Fast ingest, fast query, lean footprint. Your operational nervous system — always on, always current.
Hot metrics · PromQL · High cardinality · Alert inputs · KEDA scaling
🔬
ClickHouse — The Analytical Depth
Long-term metrics analytics. Distributed trace storage with sub-second query across billions of events. Resource utilisation modelling. KEDA parameter analytics from historical demand data. Cross-signal correlation — infrastructure metrics joined with application traces in a single query. Your intelligence engine — makes every other decision in the platform smarter.
Long-term analytics · Trace storage · Cross-signal correlation · SQL queries
Running both backends is a deliberate architectural decision. The analytical depth that ClickHouse provides — tuning autoscaling models, understanding utilisation patterns, correlating traces with infra events — returns more value than the cost of operating it. We made this choice because it is correct, not because it is cheap.

Observability that
every other feature reads from.

01
Unified Centralised Dashboard

One dashboard surface for all four FalconIO features. IDP request state, infra topology, operational telemetry, BC Manifest coverage, incident queue — no context-switching between tools.

02
Trace-Metric Correlation

Distributed traces stored in ClickHouse alongside infrastructure metrics. Correlate a service latency spike with a CockroachDB replica lag event in a single query. Performance as a consequence of infrastructure state.

03
Resource Utilisation Intelligence

Long-term ClickHouse aggregations on CPU, memory, GPU, network, and storage by workload, namespace, and environment tier. Understand patterns, not just current utilisation.

04
KEDA Tuning — With and Without AI

Autoscaling parameters informed by ClickHouse demand analytics — baseline profiles and burst distributions derived from historical data. AI-assisted recommendation available. Full manual control with analytical visibility if preferred.

05
Operationally Specific Dashboards

Pre-built for SCM, logistics, manufacturing, and industrial stacks. Kafka/Redpanda lag, CockroachDB replication lag, order pipeline velocity, warehouse throughput — correlated with infrastructure state. Not generic cluster health.

06
BC/DR + Incident Integration

Observability is the source of truth for BC/DR trigger thresholds and incident context. When an alert fires, the incident ticket is created with the triggering metric, historical baseline, and affected services — automatically, before an engineer touches a keyboard.

60–80% cost reduction
at scale. A consequence,
not a campaign.

We do not lead with cost. We lead with resilience, analytical depth, and operational intelligence. The 60–80% cost reduction vs Datadog at comparable telemetry volumes is a consequence of choosing open backends with no per-host, per-metric, or per-user pricing.

For enterprises running hundreds of Kubernetes nodes across US and European production environments, that consequence is material. It is also irreversible — once you own your telemetry data and the query infrastructure to analyse it, there is no reason to go back.

Stack Comparison
Hot MetricsVictoriaMetrics
Long-term AnalyticsClickHouse
Trace StorageClickHouse
Telemetry PipelineOTel Collector + Vector
DashboardsGrafana (pre-built)
Pricing modelOpen · No per-host pricing
Data ownershipYours. Always.