Faster only because the distances involved are short enough that the PHY layer adds significant overhead. But if you somehow could wave a magic wand and make optical computing work, then fiber would be faster (& generate less heat).
> Faster only because the distances involved are short enough that the PHY layer adds significant overhead.
This specifically mentions the 7130 model, which is a specialized bit of kit, and which Arista advertises for (amongst other things):
> Arista's 7130 applications simplify and transform network infrastructure, and are targeted for use cases including ultra-low latency exchange trading, accurate and lossless network visibility, and providing vendor or broker based shared services. They enable a complete lifecycle of packet replication, multiplexing, filtering, timestamping, aggregation and capture.
It is advertised as a "Layer 1" device and has a user-programmable FPGA. Some pre-built applications are: "MetaWatch: Market data & packet capture, Regulatory compliance (MiFID II - RTS 25)", "MetaMux: Market data fan-out and data aggregation for order entry at nanosecond levels", "MultiAccess: Supporting Colo deployments with multiple concurrent exchange connection", "ExchangeApp: Increase exchange fairness, Maintain trade order based on edge timestamps".
Latency matters (and may even be regulated) in some of these use cases.
The PHY contributes only 1ns difference, but the results also show 400ps/m advantage for copper which I can only assume to come from difference in EM propagation speed in the medium.
No. Look at the graph -- the offset when extrapolated back to zero length is the PHY's contribution.
The differing slope of the lines is due to velocity factor in the cable. The speed of light in vacuum is much faster than in other media. And the lines diverge the longer you make them.
It's true, but also if you go look at their product catalog you will see none of their direct attach cables are longer then 5m, and the high bandwidth ones are 2m. So, again, it's true, but also limiting in other ways.
