4 ms·

I wonder, is this somehow a latency-throughput tradeoff in disguise? This seems roughly analogous to batch processing, where you structure the dataflow to be m

I wonder, is this somehow a latency-throughput tradeoff in disguise? This seems roughly analogous to batch processing, where you structure the dataflow to be mechanically sympathetic to the machine doing the processing. Processing each item becomes more efficient, so you process more items per unit of time, but each item takes longer to process individually. In the same way, it seems like we're structuring the crowd into columns to optimize the layout of individuals with respect to the floorspace. It takes less time to evacuate everybody, but each person spends more time in the building. I'm not sure what to do with this knowledge, if it is indeed correct. EDIT: I think the question I'm orbiting is whether or not this strategy can generalize to people. In the realm of ants, things like self-preservation and the value of the individual are very different from humans.

thaumasiotes 6y ago
It is definitely not a latency-throughput tradeoff in disguise, since the metric is "time taken for everyone to leave the enclosure". > it seems like we're structuring the crowd into columns to optimize the layout of individuals with respect to the floorspace. It takes less time to evacuate everybody, but each person spends more time in the building. As stated, this is impossible. Everyone starts inside.[1] If the time taken to empty the building is lower, it cannot be the case that each person spends more time inside; at least one person takes the full evacuation time, or else the evacuation time would be lower. [1] This is the core difference from the analogy you picked, batch processing, in which jobs arrive at different times. If all the jobs arrived at once, batch processing would be the only possible kind of processing.
- omginternets 6y ago
  >This is the core difference Got it! Thanks for taking the time to point out my error.

thaumasiotes 6y ago

It is definitely not a latency-throughput tradeoff in disguise, since the metric is "time taken for everyone to leave the enclosure". > it seems like we're structuring the crowd into columns to optimize the layout of individuals with respect to the floorspace. It takes less time to evacuate everybody, but each person spends more time in the building. As stated, this is impossible. Everyone starts inside.[1] If the time taken to empty the building is lower, it cannot be the case that each person spends more time inside; at least one person takes the full evacuation time, or else the evacuation time would be lower. [1] This is the core difference from the analogy you picked, batch processing, in which jobs arrive at different times. If all the jobs arrived at once, batch processing would be the only possible kind of processing.

omginternets 6y ago
>This is the core difference Got it! Thanks for taking the time to point out my error.