2 ms·
Not my area, just curious. If I am getting this (still reading the paper... on a phone :/)... Stupid fast latency due to no processing overhead but slow switch
by scrps 3y ago
Not my area, just curious.
If I am getting this (still reading the paper... on a phone :/)... Stupid fast latency due to no processing overhead but slow switching speeds due to how it physically switches, this is basically good for any type of reconfigurable cluster of systems that need low latency without a fixed topology, not strictly ML specific or have I misunderstood that?
- cavisne 3y agoCorrect, google also has a paper on this at the DC level https://arxiv.org/pdf/2208.10041.pdf https://arxiv.org/pdf/2208.10041.pdf