A trigger signal provided via a pulse-per-second input port of a network interface controller is detected. In response to the trigger signal, an internal hardware clock value of the network interface controller is recorded. The recorded internal hardware clock value is reported, wherein the reported internal hardware clock value is reported for use in determining a timing error of the network interface controller based at least in part on a comparison with a time value of another device that also received the trigger signal.

A low-latency network device and method for treating serial data comprising an oscillator generating a device-wide clock; a receiving physical medium attachment (PMA) having an internal data width, a symbol timing synchronization module configured to receive the parallelized sample stream; and detect therefrom synchronized bit values corresponding to bit values of the received serial data; and a physical convergence sublayer (PCS). The PMA is configured to receive the serial data, deserialize the serial data based on the device-wide clock and internal data width, whereby the received serial data is oversampled, the oversampling of the received serial data being asynchronous relative to a timing of the received serial data, and output a parallelized sample stream. The PCS is configured to receive the synchronized bit values; and delineate packets therefrom to provide packet-delineated parallelized data. The PMA, the symbol timing synchronization module and the PCS are all driven by the device-wide clock.

Technologies for calibrated network interlink access. In some embodiments, a system can calculate a first communication latency of a first link between a first processing element in a first switch and a second processing element in a second switch, and a second communication latency associated with a second link between the first processing element and a third processing element in a third switch. The system can determine a delta between the first communication latency and the second communication latency, and whether respective clock rates of the first switch, second switch, and third switch have a clock rate variation, to yield a clock rate variation determination. Based on the delta and clock rate variation determination, the system can determine an offset value for synchronizing the first communication latency and second communication latency. Based on the offset value, the system can calibrate traffic over the first link and/or the second link.

In a transceiver, the accuracy of a packet time stamp can be improved by compensating for errors introduced by processing of the packet. A received packet can be received via multiple lanes. A packet time stamp can be measured using a start of frame delimiter (SFD). A last arriving lane can be used to provide a recovered clock signal. A phase offset between the recovered clock signal and the system clock of the transceiver can be used to adjust the time stamp. A position of the SFD within a data block can be used to adjust the time stamp. A position of the data block within a combined group of data blocks can be used to adjust the time stamp. Also, a serializer-deserializer delay associated with the last arriving lane can be used to adjust the time stamp.

In one embodiment, a device includes a hardware clock to maintain a clock value, a hardware counter to maintain an estimation of a dynamic error bound of the clock value, and a clock controller to intermittently discipline the hardware clock responsively to a remote clock, advance the hardware counter at a rate responsively to a clock drift, and adjust the hardware counter responsively to the hardware clock being disciplined.

A network element includes a port; and a device with circuitry configured to encode data for communication to a second device via a plurality of physical channels, and utilize one of the plurality of physical channels as a dedicated timing channel with encoding thereon different from encoding on the other plurality of physical channels, and interface encoded data via the plurality of physical channels with the port for transmission and reception with a second device.

A system includes a plurality of line cards and a timing card. A clock generation circuit on the timing card generates a clock signal which is pulse width modulated according to information to be transmitted. A clock line supplies the pulse width modulated clock signal to the line cards. The timing card sends a first control word to the plurality of line cards over the clock line after sending a beacon. The first control word includes a size field specifying a first length of first data following the first control word. The timing card sends time of day information over the clock line to the line cards following the first control word. The time of day information may be encrypted. A second control word follows the time of day information. One or more additional control words can follow the second control word before the next beacon.

The Digital Time Processing using Rational Number Filters (DTP RNF) disclosed herein is contributing methods, systems and circuits for using a Precision Time Protocol (PTP) such as IEEE 1588 for distributing a master time secured by a master unit to slave units by utilizing slave clocks, synchronous to referencing frames communicated with PTP messages or compatible with them data receiver clocks, for maintaining a local slave time which is increased to a local master time by adding to it an estimate of a transmission delay derived by processing PTP messages or by other means, wherein such distribution of the master time includes filtering out phase noise of the timing referencing signals with the Rational Number Filters in order to produce accurate and stable timing implementing signals such as the slave clock, local slave time and local master time.

In one embodiment, a network device, includes a network interface port configured to receive data symbols from a network node over a packet data network, at least some of the symbols being included in data packets, and controller circuitry including physical layer (PHY) circuitry, which includes receive PHY pipeline circuitry configured to process the received data symbols, and a counter configured to maintain a counter value indicative of a number of the data symbols in the receive PHY pipeline circuitry.

Tenants in data centers may want access to high precision clocks without having to run their own PTP stacks or reference clocks. Furthermore, different tenants may want their workloads synchronized to their own secured clock domain. PTP, the currently dominant synchronization protocol, allows for only 256 clock domains (CDs). Virtual CDs (vCDs) virtualize the concept of clock domains by maintaining a hardware clock within a host computer, receiving a network clock domain packet that includes a clock domain identifier and an origin timestamp produced by a reference clock, using the network clock domain packet to synchronize the hardware clock to the reference clock, and using the hardware clock to provide a hardware timestamp value to a virtual machine (VM) running on the host computer or to a process running on the host computer, wherein the hardware clock is secured from manipulation by the VM or by the process.