[omniORB] anomalous behavior of Sequences on Linux

Vibhu Rishi vibhu.rishi@wipro.com
12 Sep 2001 11:13:41 +0530


hi 

I ran the Benchmark that comes with ORBacus/E with omniORB and I find
that there is some speed differences w.r.t. the platforms I run it on
!!!! This seems very strange to me and I do not know the reason as to
why this is happening ! I have compiled the code on both Linux and
Windows platforms and here is what I get :

On LINUX ( RH 7.2 kerver 2.4.2)
1000 S2Seq_UB in: 1.653000 seconds (604.96/sec)
1000 S2Seq_UB out: 40.111000 seconds (24.93/sec)
1000 S2Seq_UB i/o: 79.999000 seconds (12.50/sec)
1000 S2Seq_UB ret: 40.000000 seconds (25.00/sec)

On Windows ( Win2k)

1000 S2Seq_UB in:               2.077940 seconds (481.25/sec)
1000 S2Seq_UB out:              2.305242 seconds (433.79/sec)
1000 S2Seq_UB i/o:              4.822260 seconds (207.37/sec)
1000 S2Seq_UB ret:              2.518529 seconds (397.06/sec)

Just look at the speed loss when on Linux ! this is observed always when
the server is running on Linux. The client can be linux or windows.
Also, I have done interworking testing with JacORB and MICO and the same
result is got. If the server is omniORB on Linux , I am getting this
huge time gap ! No such problem with the server on Windows.

Is the code for handling the Sequences different for different platforms
?? 


vibhu..