I've been wondering, why is there a 40 Gigabit Ethernet standard? Is there a reason why it's 40GbE and not 50GbE or why is there any step in between from 10GbE to 100GbE?

Unfortunately my own research didn't lead to an answer. The only thing someone wrote in a forum was, that 40GbE was the highest speed, which could be achieved with copper, when this standard was introduced, but as far as I understand 40GbE and 100GbE was proposed at the same time?

  • A twisted-pair cable has four pairs of cables. If each pair could do 10Gb/s, ... WARNING: this is an unconfirmed claim I have heard somewhere, don't trust me!
    – corny
    Aug 26 '16 at 19:28

There are two main reasons, and they are closely related.

One is that the use of four data paths is common and 10Gb PHYs are available. That would make 40GE a natural step.

Another is that it was expected to be common for backplanes to have a 40Gb path that would be broken out in a line card to four 10GE links. A 40GE standard would allow a line card that plugged into that same backplane connection.

  • Further, 25 Gb came about because it was cheaper than 4x of the PHYs to make a 100 Gb. The state of the art single lane will be popular, particularly if it can fit in a RJ45 or SFP form factor. Aug 28 '16 at 15:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.