Pulse
Infrastructure visibility without the complexity
Monitoring tools either do too little — a ping every minute — or demand a platform team to operate, like self-hosted Prometheus federation. Pulse sits exactly in between: one static Go binary per host, a dashboard that answers “is everything okay?” at a glance, and alerting that respects your sleep.
We built Pulse because we needed it for our own client infrastructure. It monitors the machines that monitor your machines — and it’s launching soon for your fleet too.
<10MB
Agent memory footprint
1s
Metric resolution
13mo
Retention on Team
Install in one line
curl -sSL https://get.servelium.com/pulse | bash -s -- --key YOUR_API_KEYKey features
Agent-based
A single static Go binary. No dependencies, ~10MB RAM, installs in seconds.
Real-time metrics
CPU, memory, disk, and network streamed live — 1-second resolution when you need it.
Smart alerting
Threshold and anomaly alerts to email, Slack, or webhooks — with flap suppression.
Status pages
Public or private status pages generated straight from your real metrics.
Multi-host fleet view
Servers, VMs, and bare metal across regions in one dashboard.
Dark industrial UI
Built for NOC screens and 2 AM incidents. High contrast, zero clutter.
Where it shines
Startups without an SRE
Get production-grade visibility before you can afford a platform team.
Agencies & MSPs
Watch every client’s fleet from one place, with per-client status pages.
Edge & bare metal
Monitor machines that don’t fit neatly into cloud-native tooling.
Under the hood