[poochkoishi728]

Is there a mistake in the Swift graph of 'Success rate (Y) over concurrent requests (X)'? It shows a near-constant 1.0 success rate, but the description says "[Swift] had higher throughput than PHP yet requests started failing much sooner as load increased".

EDIT: Nm, it must be referring to the slight dip that starts early in the Swift graph.

[bsaul]

That's the pb OP is mentioning quite early : the benchmark didn't make sense.