MUD-Dev
mailing list archive
[ Other Periods
| Other mailing lists
| Search
]
Date:
[ Previous
| Next
]
Thread:
[ Previous
| Next
]
Index:
[ Author
| Date
| Thread
]
Re: [MUD-Dev] Text Parsing
J C Lawrence wrote:
> I've scaled using simulated users under very artificial conditions
> and a rather minimal world to just over 700 simultaneous players on
> a fairly mid-range box. Performance at that point really became a
> kernel function due to poor MT work on IO under Linux (single lock).
When you speak of simultaneous users, how active were they? Did they issue 1
command per second, a constant stream of commands, or what?
I do frequent benchmarking on my moo, but I've so far been using a
simplistic test that calculates the average number of commands per second.
(Perlmoo can manage about 170/sec right now on a P-II 350, I figure that
corresponds to more than 5x that many normal active users.) Of course this
eliminates any overhead due to threading or network IO, etc, but that can be
a good thing when you have little control over those factors anyway.
--
see shy jo
_______________________________________________
MUD-Dev maillist - MUD-Dev#kanga,nu
http://www.kanga.nu/lists/listinfo/mud-dev
- Thread context:
- Re: [MUD-Dev] Text Parsing, (continued)
- Re: [MUD-Dev] Text Parsing,
Petri Virkkula pvirkkul#iki,fi, Fri 11 Jun 1999, 03:46 GMT
- Re: [MUD-Dev] Text Parsing,
Adam Wiggins adam#angel,com, Fri 11 Jun 1999, 18:12 GMT
- Re: [MUD-Dev] Text Parsing,
Katrina McClelan kitkat#marcus,pants.nu, Fri 11 Jun 1999, 19:34 GMT
- Re: [MUD-Dev] Text Parsing,
J C Lawrence claw#varesearch,com, Fri 11 Jun 1999, 02:46 GMT
- Re: [MUD-Dev] Text Parsing,
Joey Hess joey#kitenet,net, Thu 10 Jun 1999, 06:14 GMT
- Re: [MUD-Dev] Text Parsing,
J C Lawrence claw#varesearch,com, Fri 11 Jun 1999, 02:11 GMT
Re: [MUD-Dev] Text Parsing,
Caliban Tiresias Darklock caliban#darklock,com, Tue 08 Jun 1999, 03:31 GMT
[ Other Periods
| Other mailing lists
| Search
]