1. Summary
A single 188-minute run of the full release suite against one physical BNG with
two 40GbE dataplane ports and two load-generator hosts, at a working set of
5,000 concurrent subscribers.
| Subscribers onboarded | 5,000 of 5,000 — 100% |
| Onboarding rate | 398.7 per second, 5,000 bound in 12.5 s |
| In-rush completion | 391.8/s cold, 405.5/s warm |
| Downstream throughput | 11.7 Gbps mean, 16.5 Gbps peak, 30.2 GB moved |
| QoS compliance | 480 of 480 measured subscribers within contract |
| Churn integrity | 1.0 — zero leaked sessions, zero orphans |
| Forwarding (2 workers, 64B) | 13.57 Mpps, platform-limited |
| Per-packet cost | ~330 cycles at 64B |
| Checks passed | 4,663 of 4,668 |
Read the hardware line before any throughput figure. These numbers come
from a processor that launched in 2013. They are floors, not best-case
results — see §3.1.
Four of the twenty forwarding cells are maximum-throughput numbers. A
further thirteen are wire-speed results — the box forwarded everything offered,
at 90% of link, with zero loss — and three are floors where the rig never filled
the link. Every cell is labelled with which it is. §5.7 explains how to read
them, and no floor is promoted into a product claim anywhere in this report.