[MUD-Dev] OT: Java multithreading test source

Jon A. Lambert jlsysinc at ix.netcom.com
Fri May 22 17:52:37 New Zealand Standard Time 1998


Here are my results in graphic format running Vadim's Java thread 
test program. 

The machine was running Win95 on a Pentium II at 300mhz with 64mb of 
memory.

The MS test was run using the latest VM supplied with IE 4.01 with 
the JIT enabled.  It was run from JView supplied with Visual J++ 1.1.

The Sun test was run using the 1.1.6 JDK with the Symantec JIT 
enabled.

Some notes: 

The first 4 results where thrown out because of really wild 
variations in the first few MS threads in order to get my graphics 
program to scale theY axis for readability. 

Both tests took less than 6 minutes to run to completion although
completion was termination due to out of memory errors.  

The Sun test outputted raw data with negative lag values every dozen 
or so lines.  So Sun results had a built-in bias in the average 
calculation.  
(<hehe> I'm sure this is merely a bug and not evidence of a vast 
anti-MS consipracy :P)

I tried to run the test without the Sun JIT but it logged errors 
after the first 100 threads.

I'm attempting to get FreeBSD installed on the exact same machine and 
if I ever figure it out I'll run some comparsions using Kaffe.  

My thoughts:
Vadim ran out of processor before running out of memory at 128mb.
I ran out of memory before I ran out of processor at 64mb.
In the past MS's JIT has whipped the pants out of Sun's VM (even the 
one with the Windows performance pack.  Sun's decision to package the 
Symantec JIT is a good thing.  It's very competitive. :)

The raw data is available on demand. ;)
--
--/*\ Jon A. Lambert - TychoMUD     Internet:jlsysinc at ix.netcom.com /*\--
--/*\ Mud Server Developer's Page <http://www.netcom.com/~jlsysinc> /*\--
--/*\   "Everything that deceives may be said to enchant" - Plato   /*\--




More information about the MUD-Dev mailing list