[omniORB] anomalous behavior of Sequences on Linux

Duncan Grisby dgrisby@uk.research.att.com
Wed, 12 Sep 2001 09:57:47 +0100


On Wednesday 12 September, "Vibhu Rishi" wrote:

> 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 :

That is very very strange. The code is, as you would expect, the same
for all platforms.

Something I think is very interesting is that the in argument test is
faster on Linux than Windows, so it's not as simple as the marshalling
just taking a long time. The only similar thing I've seen is a problem
that only affects the TCP loopback on Linux, where the time for a
transfer fluctuates enormously with different sequence sizes. See

  http://www.uk.research.att.com/omniORB/archives/2000-12/0057.html

But I don't think that's anything to do with the problem here, since
you say the same problem occurs over the network.

If the benchmark code available?  If so, where can I get it?

Cheers,

Duncan.

-- 
 -- Duncan Grisby  \  Research Engineer  --
  -- AT&T Laboratories Cambridge          --
   -- http://www.uk.research.att.com/~dpg1 --