Zero-instrumentation monitoring

See everything your
servers are doing.

One Helm command, 30 seconds to deploy. No code changes, no external databases, and no cloud account. Your telemetry data never leaves your network. Apache 2.0.

View on GitHub Install →

What is aacyn?

A monitoring tool that watches your servers from the operating system level — no SDKs to install, no code changes, no cloud account required.

1

Monitor

See every request, error, and slow response across your services. aacyn uses eBPF to observe traffic at the kernel level — no agents or SDK installs needed.

2

Self-hosted

Runs on your hardware — a mini PC, a server rack, or a Docker container. Your telemetry data never leaves your network. No cloud vendor lock-in.

3

Free and open source

No per-host fees. No per-GB surcharges. No surprise invoices. Apache 2.0 licensed — use it, modify it, ship it. All features included, no tiers.

Deploy in 30 seconds with Helm

One command installs the eBPF probes on every node and the dashboard aggregator. Works alongside any CNI — no migration needed.

terminal
helm repo add aacyn https://charts.aacyn.com helm install aacyn aacyn/aacyn \ --namespace aacyn --create-namespace

Ingest millions of events per second

Most monitoring tools parse JSON on every incoming request, which limits throughput. aacyn accepts a compact binary format that skips parsing entirely — the bytes from your HTTP request go straight into storage.

The result: over 5 million events per second on a single box. That's 16× faster than JSON-only ingestion.

5,089,364 events/sec on a Ryzen 9 mini PC — no cloud cluster needed

Query 5 million events in under a millisecond

aacyn stores data in a format optimized for analytics — instead of reading entire rows, it reads only the columns you need. Combined with modern CPU acceleration, this makes searching, filtering, and aggregating nearly instant.

286μs to scan 5 million events using CPU vector instructions (SIMD) on a single core

Your data never leaves your network.

aacyn runs on your hardware and stores everything locally. There is no cloud backend, no telemetry, and no phoning home. Your data stays on your machines.

Features

Everything included. No tiers, no locked features, no enterprise upsells.

Golden Signals

RED metrics (rate, errors, duration) per service, computed automatically from kernel probes. No configuration required.

Distributed Tracing

Automatic span generation with W3C traceparent propagation. Trace waterfall view for every request.

Service Topology

Auto-discovered service dependency map from kernel-level TCP connection tracking. Pod names, namespaces, deployments enriched via K8s API.

SLO Tracking

Error budgets and burn rate alerts per service. Google SRE workbook model. Multi-window burn rate detection.

OTLP Ingest & Export

Bridge existing OpenTelemetry instrumentation. Accept OTLP traces and export to any OTLP-compatible collector.

Forwarders

Push metrics to Datadog (Metrics API v2), Splunk (HEC), or any OTLP collector. Use your existing observability stack.

gRPC Visibility

Kernel-level gRPC protocol parsing — service name, method, status code — without sidecars or proxies.

Grafana Plugin

Native Grafana data source plugin. Query aacyn topology and golden signals from your existing Grafana dashboards.

Pre-configured Alerts

5 rules out of the box: latency spikes, error rate thresholds, connection failures, SLO burn rate, and ring buffer drops.

aacyn is free and open source.

Apache 2.0. No tiers, no limits, no phone calls from sales.

View on GitHub Install →