I was reading the paper and got stuck somewhere when I saw this:
> With Mio, realistic HTTP servers in Haskell scale to 20 CPU cores, achieving peak performance up to factor of 6.5x compared to the same servers using previous versions of GHC. The latency of Haskell servers is also improved: <snip> under a moderate load, reduces expected response time by 5.7x when compared with previous versions of GHC
