MUD-Dev
mailing list archive
[ Other Periods
| Other mailing lists
| Search
]
Date:
[ Previous
| Next
]
Thread:
[ Previous
| Next
]
Index:
[ Author
| Date
| Thread
]
[MUD-Dev] Re: OT: Java multithreading test source
On 23 May 98, Mike Sellers wrote:
>
> Wow, hard data. A graph even. Very cool. What is the scale on the y-axis
> though? Are those milliseconds? Seconds? Kwatlus?
>
That should be average millisecs.
> Also, what's the explanation for the falling lag values for the first few
> (80-120?) threads?
>
The first few threads take a "long" time (several hundred ms.) The
average lag seems to be a cumulative of what went before. So we
should naturally see a sharp slope until the raw data starts showing
real lag. This would be due to class loading and invocation of the
JIT.
--
--/*\ Jon A. Lambert - TychoMUD Internet:jlsysinc#ix,netcom.com /*\--
--/*\ Mud Server Developer's Page <http://www.netcom.com/~jlsysinc> /*\--
--/*\ "Everything that deceives may be said to enchant" - Plato /*\--
- Thread context:
- [MUD-Dev] Re: OT: Java multithreading test source, (continued)
[ Other Periods
| Other mailing lists
| Search
]