They make a big point of using HD-FEC rather than SD-FEC.
To me, that makes no sense - the extra power usage of SD-FEC can be tiny, even at high data rates. The FEC problem can be parallelized (if the protocol is designed for this) so it doesn't need to run at line rate too.
(SD-FEC lets you get substantially more data throughput in a given channel).
To me, that makes no sense - the extra power usage of SD-FEC can be tiny, even at high data rates. The FEC problem can be parallelized (if the protocol is designed for this) so it doesn't need to run at line rate too.
(SD-FEC lets you get substantially more data throughput in a given channel).