Open Source Sflow Collector

| Metric | Value | |--------|-------| | Max datagrams/sec | 50k – 200k (depending on decoding complexity) | | Flow exports/sec | 500k – 2M | | Memory usage | <2GB (with 1h time‑window) | | CPU cores | 4‑8 (efficient if per‑core poller) |

Each sFlow datagram contains:

sFlow is an industry-standard technology that provides continuous, real-time visibility into network traffic. But to turn that stream of data into actionable intelligence, you need a collector. While commercial options exist, the open-source ecosystem has matured into a powerful, flexible, and cost-effective solution. open source sflow collector

If you are running an enterprise network or a large cloud infrastructure, vendor lock-in is a real risk. Proprietary collectors often charge by the volume of flows or the number of interfaces—a cost that explodes as your network grows.

sFlow-RT doesn't export to Prometheus by default without a little configuration, but it has a dedicated app for it. You can install the prometheus app within the sFlow-RT UI. | Metric | Value | |--------|-------| | Max

Typical 5‑tuple: (src_ip, dst_ip, src_port, dst_port, protocol)

Enter .

The sFlow (Sampled Flow) architecture is built on three main components: