[omniORB] Latencies - Oneway vs. Roundtrip

[Gary D. Duzan]

In Message <3711F5E0.6BBEDC4E@datarespons.no> ,
   "Helge Penne" <Helge.Penne@datarespons.no> wrote:

=>My theory is that omniORB buffers my oneway calls and sends them in larger
=>packages every 200ms (5Hz).  This explains every symptom that I have seen, and
=>also fits in nicely with my previous experience with VisiBroker.  VisiBroker
=>also buffers such requests by default, but there is an option in BOA_init that
=>lets you turn off this buffering and force these requests to be sent
=>immediately.
=>
=>This leads up to a very important question: How do we turn off this buffering
=>in omniORB?  This is quite an important question for my project.  Has anybody
=>done this before?

   My guess would be that it is your TCP stack doing the buffering. On
AIX, at least, there is a TCP_NODELAY option (for use with setsockopt())
which turns off the "Nagle algorithm for deciding when to send data".
You might look to see if the NT socket API has something similar. If
so, OmniORB could be modified to use it. (I should probably put together
a patch for AIX to use TCP_NODELAY, actually.)

					Gary Duzan
					GTE Laboratories