[realreality]

Oh, that’s even less-expected behavior! In that case, one would never run into the size limitation for range reads. I think the docs should clarify that only the first and last keys count toward the transaction size.

Yes, the 5 second limit could be a problem.

[ryanworl]

There are two reasons for this I can see:

1) All mutations and key ranges are stored locally and submitted when the transaction commits. Lots of data to transfer.

2) the optimistic conflict checker can process one range (even if it is a lot of underlying keys) a lot easier than each individual key in that range