Scaling Reggie

Dennis Reedy <[email protected]>
Newsgroups gmane.comp.java.sun.jini
Message-ID <[email protected]>
Hi,

I've been asked to evaluate a deployment [*] that at the moment has
800 clients performing simultaneous discovery of a lookup service,
with subsequent discovery of services that have active service
registrations [**].

We are getting out of memory exceptions when the clients start up and
do the discovery (this occurs within a 5-6 second window). Each
client correlates to a physical machine. The immediate thought was to
add a random wait for the clients to initialize, this may not be
possible due to other (assumed) application constraints.

Is the right place to look would be the
com.sun.jini.reggie.taskManager setting? What would be a recommended
setting for this property?

Another approach is to partition the clients to separate lookup
instances, perhaps 200-300 per lookup. Are there any rough ideas on
what the "right" number here could be

Has anyone scaled reggie to this level (or beyond)?

Thanks

Dennis

[*] Sorry for incomplete info at this time, I dont have much to go on
other then general description of the problem
[**] There are approximately 300 services registered.

--------------------------------------------------------------------------
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]
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.