1.53 Petabits per Second Transmission in 55-Mode Fiber

Re:1.53 Petabits per Second Transmission in 55-Mode Fiber by rowanG077 on Tuesday November 15, 2022 @06:27AM (#33607879)

Insane. I wonder how you would even design a memory subsystem for this. Back of the envelope calculation says you'd need at least a 212500 bit wide DDR5-7200 bus to even approach that throughput.

[ Parent | on HN ]

Re:1.53 Petabits per Second Transmission in 55-Mode Fiber by wbl on Tuesday November 15, 2022 @06:44AM (#33607997)

My guess is that it gets treated as 55 separate optical channels each with their own receivers and that there is considerable multiplexing on top.

[ Parent | on HN ]

Re:1.53 Petabits per Second Transmission in 55-Mode Fiber by krisroadruck on Tuesday November 15, 2022 @06:50AM (#33608031)

These things don't go to a single system. The signal is optically split at the ends and there are big racks of transceivers, each transceiver deals with a single mode. With 55 modes you are looking at dealing with 27 terabits or 125 terabytes per second on any given system.

[ Parent | on HN ]

Re:1.53 Petabits per Second Transmission in 55-Mode Fiber by MayeulC on Tuesday November 15, 2022 @09:16AM (#33609402)

I get 28.5 Tib/s, so 3.6 TiB/s, not sure how you went from 27 to 125?

[ Parent | on HN ]

Re:1.53 Petabits per Second Transmission in 55-Mode Fiber by krisroadruck on Tuesday November 15, 2022 @09:55AM (#33609830)

I am also not sure, it's not even a case of multiplied instead of divided. Bizarre.

[ Parent | on HN ]

Re:1.53 Petabits per Second Transmission in 55-Mode Fiber by nine_k on Tuesday November 15, 2022 @07:30AM (#33608317)

I suppose this is not about larger bandwidth from/into one system, but about fewer long cables (say, transoceanic).

[ Parent | on HN ]