25Gbit/s Ethernet PCS

Overview

The Chevin Technology 25GPCS provides Ultra low-latency 25Gbit/s Ethernet connectivity in Xilinx Virtex® UltraScale™ FPGAs. Ultra-low latency is achieved by using only the PMA function in FPGA Multi-Gigabit transceivers, and moving all PCS functions to code that is optimized for 25GBASE-R. This allows the data to take the shortest and lowest latency path, to and from the wire. The 25GPCS /PMA core can be used directly with Multi-Gigabit Transceivers (SerDes & CDR logic) in Xilinx Virtex® UltraScale FPGAs for the lowest possible latency.

Key Features

  • Designed to IEEE 802.3by
  • Low Latency PCS/PMA (RTT) 99 ns
  • 5250 LUTs
  • Integrated 64b66b encoder/decoder
  • Integrated Scrambler/Descrambler
  • Gearbox for rate conversion
  • Fault management, BER monitoring
  • Clock Rate adaptation, TX/ RX
  • Detailed traffic analysis statistics collection
  • XXVGMII interface, 64bit @ 390.625MHz
  • Connect directly using SFP28
  • Use 4-channel for QSFP28 interfaces

Benefits

  • Ultra Low Latency to reduce round trip time
  • High performance for exceptionally fast, reliable data delivery
  • Easy to integrate with other Chevin Technology IP cores

Block Diagram

25Gbit/s Ethernet PCS Block Diagram

Applications

  • HPC / Big Data systems
  • Trade execution & monitoring
  • Data Storage & Capture systems
  • Signal processing systems
  • Data Mining

Deliverables

  • Encrypted RTL/VHDL source code
  • Encrypted compiled netlist
  • Datasheet & User Guide to assist integration
  • Reference Design for Alpha Data’s ADM-8V3 board
  • Simulation Test bench
  • Build scripts for Vivado
  • Support for integration into FPGA

Technical Specifications

Availability
Now
×
Semiconductor IP