Re: JERI vs. JRMP performance penalties

Bob Scheifler <[email protected]> Wed, 19 Dec 2007 11:08:17 -0500
Newsgroups gmane.comp.java.sun.jini
Message-ID <[email protected]>
> However, increasing client.initialInboundRation ration had no dramatic
> effect to throughput with my littele example ( i tried 0x10000, 0x20000,
> 0x40000 and 0x100000 )

It's the server ration that will matter for your example. Setting it
to 0 (infinite) will help some.  But the key to matching JRMP performance
for your example would seem to be the combination of setting the server's
ration to 0 and increasing the client's fragment size (in my testing,
to 8192).

> Increasing server.initialInboundRation to 0x10000 also had no effect.
> However, setting it 0x20000 has the effect that the first call from the
> client NEVER COMES BACK !

Yep, I hit the same problem and filed a bug.

- Bob

--------------------------------------------------------------------------
Getting Started:     http://www.jini.org/wiki/Category:Getting_Started
Community Web Site:  http://jini.org
jini-users Archive:  http://archives.java.sun.com/archives/jini-users.html
Unsubscribing:       email "signoff JINI-USERS"  to [email protected]