The White Rabbit ecosystem
CERN-derived sub-nanosecond fibre-distributed timing — for the applications where IEEE 1588 PTP is no longer precise enough. Open standard, commercially mature, and built for the networks that need accuracy below one nanosecond, continuously.
< 1 ns
Accuracy
24.67 ps
RMS jitter
Open
Standard
What it is
One precision tier above PTP
White Rabbit is an extension of IEEE 1588 PTP that adds hardware-level phase tracking and synchronous Ethernet frequency distribution over fibre. The combination delivers sub-nanosecond accuracy continuously — not during holdover, not under ideal conditions, but every nanosecond of every day for as long as the fibre link is up.
Developed at CERN in the late 2000s for the Large Hadron Collider, published as an open standard, and now deployed commercially in financial trading, AI training clusters, quantum networking and defence. White Rabbit is not in competition with PTP — it sits one precision tier above it, for the use cases where tens of nanoseconds is still too much drift.
How it works
Three layers that PTP can’t do alone
01
PTP message exchange
Standard IEEE 1588 PTP runs between every node in the White Rabbit fabric, measuring round-trip delay and clock offset — the same protocol PTP networks already use.
02
SyncE frequency lock
Synchronous Ethernet recovers the master's frequency from the physical-layer clock on the fibre link. Every node locks its oscillator to the upstream master's frequency, continuously.
03
Phase tracking
A hardware phase detector on each node measures the sub-nanosecond phase offset between its local clock and the recovered SyncE clock. This is what pushes precision from tens of nanoseconds into single digits and below.
At the edge of the White Rabbit fabric, a WR boundary clock outputs standard IEEE 1588 PTP to the rest of the network — so a White Rabbit core delivers PTP-compatible time to ordinary PTP slave devices without those slaves needing White Rabbit hardware.
Hardware
Two products. One precision tier.
A production-ready grandmaster node for rack deployments, and an OEM mezzanine module for system integrators who want to embed sub-nanosecond timing into their own platforms.
Where it earns its place
Six sectors pushing past the PTP floor
White Rabbit is the right answer when the application’s failure mode involves something happening with sub-nanosecond temporal asymmetry. If your worst-case requirement is in the tens of nanoseconds, PTP is enough — save the budget.
Interoperability
Precision core, PTP edge
White Rabbit doesn’t replace your existing PTP infrastructure — it sits inside it as a precision-tier core. At the boundary, a WR node outputs standard IEEE 1588 PTP messages. Your existing PTP slave devices lock to the WR core exactly as they would to a conventional grandmaster, with no changes required on the slave side.
The result is a deployment where the inner fabric runs at sub-nanosecond precision and the outer edge runs at standard PTP precision — the best of both architectures with no interoperability trade-offs.
Read the PTP grandmaster guide →Architecture
WR Master
GNSS-disciplined reference. Sub-ns accuracy to UTC.
WR Fibre Fabric
Single-mode fibre between all WR nodes. Sub-ns phase lock across every link.
WR Boundary
WR node at the edge. Outputs standard IEEE 1588 PTP to the rest of the network.
PTP Slave Devices
Standard PTP slaves. No WR hardware needed. Lock to the WR boundary node as if it were a conventional grandmaster.
Go deeper
Engineering guides on White Rabbit
Talk to us
Design your WR fabric
The engineering team can walk you through architecture, fibre requirements, PTP interoperability and the operational realities of running a White Rabbit fabric in production. Whether you’re evaluating WR for a new deployment or upgrading an existing PTP fabric, the conversation starts here.
- Sub-nanosecond continuously
- CERN-derived, commercially mature
- Interoperable with existing PTP
- Runs the TimeBeat Agent
Below one nanosecond
When PTP isn’t precise enough.
The TimeBeat White Rabbit ecosystem is shipping, open, and production-ready. The networks that need sub-nanosecond precision already know who they are.