Re: Using ServiceStarter to launch a "proxy is service" service
Mark Brouwer <[email protected]>
| Newsgroups | gmane.comp.java.sun.jini |
|---|---|
| Message-ID | <[email protected]> |
Dennis Reedy wrote: > I think it would be fairly straightforward to produce some benchmarks > showing colocated vs remote invocations, intermingling Bob's > experimental endpoints package if interested. FWIW in the past the Memory(Server)Endpoints didn't provide a performance benefit, however there is a new version that outperforms JRMP at least on one benchmark, see https://user-rscheifler.dev.java.net/servlets/ReadMsg?list=users&msgNo=2 It appears as long as you want to keep RMI semantics (pass by value) in place you won't gain that much (although the serialization improvements discussed lately may have a positive impact here). There are however some good usages for the memory endpoints, such as when you want to provide a business interface of a service to another service in the same JVM where you want to keep the RMI semantics (which you should) and don't want to expose that service to the network for whatever reason you might have. Note although at this time these endpoints provide no support for constraints. -- Mark -------------------------------------------------------------------------- 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]