A Systems Engineer’s Guide to FTTP (Fibre to the Premises) Latency Baselines

As the UK expands its full fibre network coverage, network engineers need to adapt their baseline expectations. Traditionally, FTTC (Fibre to the Cabinet) setups typically exhibited a 10-15ms average ping to major UK data centers, primarily due to VDSL2 interleaving and copper line limitations. However, the FTTP (Fibre to the Premises) environment presents distinct architectural differences.

Unlike its predecessor, FTTP employs a GPON (Gigabit Passive Optical Network) system, which integrates an Optical Network Terminal (ONT) inside the customer's premises with an Optical Line Terminal (OLT) at the exchange. This passive light-splitting mechanism remarkably minimizes physical latency, according to Openreach's technical guidelines.

When performing direct ethernet diagnostics from the ONT (excluding the router to isolate the circuit), systems engineers should anticipate the following: ICMP round-trip times to 1.1.1.1 (London) should fall between 2ms and 4ms, with jitter at or below 0.5ms, and zero packet loss. If an FTTP connection consistently shows latency above 8ms locally, it's almost exclusively attributed to suboptimal ISP routing table configurations or internal home network congestion issues.

The latter, as previously highlighted in a separate guide on diagnosing bufferbloat, can often be resolved with targeted adjustments within the customer's local network infrastructure. By adhering to these established baselines, systems engineers can swiftly pinpoint whether network anomalies stem from passive optical network disruptions or pervasive intranet congestion within the customer's premises.