> > And for me too. > > > > I'm testing it (in production). It's definitely performing faster and > > lighter than pperl. > > All sounds good. > > > Now if only I could convince someone else (*cough*) that the qmail > > tcpserver architecture, high performance, and perl don't play along. > > Sounds to me like you are in a position to provide some benchmarking data > to support the argument.
Yes, but that would be a lot of work :). Although when the machine doesn't crawl to a halt when the next big virus hits, that'll be a good sign. In the past 11 hours... 1 Trojan.Dropper.C 1 Worm.Gibe.F.UPX.2 4 Worm.Dumaru.K 4 Worm.Sober.D 5 Worm.SomeFool.O 11 Worm.Bagle.Gen-1 12 Worm.Bagle.P 21 Worm.SomeFool.M 26 Worm.SomeFool.F 32 Worm.Gibe.F 36 Worm.Dumaru.Y 47 Worm.Mydoom.H 110 Worm.Bagle.Gen-zippwd 161 Worm.Bagle.N 198 Worm.Bagle.Gen-zippwd-2 247 Worm.SomeFool.I 259 Worm.Mydoom.F 352 Worm.SCO.A 2336 Worm.SomeFool.Gen-2 6273 Worm.SomeFool.Gen-1 > > There's a lot to be said for minimizing startup cost. > > I assume that you mean per connection, not startup cost overall. The > forking server increases startup cost of the run script (but drastically > reduces per connection costs). Right. Startup cost of the run script is negligible over the lifetime of the process. -R
