- Ethernet switches
- IP routers
- Data center switches
- Communications equipment
- To keep up with increasing CPU and storage bandwidth, rack or blade servers must support aggregate throughputs faster than 50 Gb/s (single lane) or 100 Gb/s (dual lane) from their Network Interface Card or LAN-on-Motherboard (LOM) networking ports.
- Given the increased bandwidth to endpoints, uplinks from Top-of-Rack (TOR) or Blade switches need to transition from 100 Gb/s (four lanes) to 400 Gb/s (four lanes) while ideally maintaining the same per-lane breakout capability.
- Due to the expected adoption of 400GBASE-CR4/KR4/SR4/LR4, SerDes and cabling technologies are already being developed and deployed to support 100 Gb/s per physical lane, twinax cable, or fiber.
In addition to typical Coupled MAC+PCS application for up to six ports (up to 12 ports for 50 Gb/s FEC-only), the DCMAC Subsystem supports MAC client monitoring applications for up to 40 channels.
The DCMAC Subsystem provides support for OIF Flex Ethernet (FlexE) by exposing a channelized MAC interface (MAC I/F) that can be connected to the client side of a FlexE shim. Likewise, the DCMAC Subsystem flex interface (FLEX I/F) provides access to the PHYs and can be connected to the FlexE group side of a FlexE shim. The FlexE shim is implemented in user logic.
The DCMAC Subsystem supports ITU-T Flexible OTN (FlexO) through the FLEX I/F. This provides access to the FEC and lane deskew logic and includes support for additional deskew using alignment buffers in user logic.