Messages in this thread | | | Date | Fri, 19 Jun 1998 00:48:22 +0200 (CEST) | From | Thomas Heide Clausen <> | Subject | RE: Remote fork() and Parallel Programming |
| |
-----BEGIN PGP SIGNED MESSAGE-----
On 19-Jun-98 mshar@vax.ipm.ac.ir wrote: > > Thomas Heide Clausen <voop@cs.auc.dk> wrote: > >>Hello all. > > Hi, > >>As this discussion has turned into a discussion about which >>paradigm is better/worse and how to implement them, I would >>like >>to point Your attention to a strictly userspace solution (even >>writen in Java) called the "Actor Foundry". A (free) >>distributed >>active object system, based on the "Actor" model. >> >>It - and information about it - is available from: >> >> http://osl.cs.uiuc.edu/foundry > > Such systems are great for developing new applications. > Object Oriented > programming is learned by nearly all programming students, so > they are > familiar with the techniques. What is needed in this realm is > a transparent > Remote Method Invocation mechanism, and preferably, object > migration with > dynamic load balancing (I am already hearing protests :-)
Funny you should mention it - that's something we are working on related to the Actor Foundry. Transparent "remote method invocation" IS in fact allready present....as is migration....... :-)
The mechanisms for dynamic load balancing thus are in place (I implemented a simple prototypic heuristics for load balancing in the Actor Foundry a while back).
- --thomas
-----BEGIN PGP SIGNATURE----- Version: 2.6.3ia Charset: noconv
iQCVAwUBNYmZMcQLb2bL5bWVAQGHIwP5AX5cFTsaz9MWI7FtBxvpfdNqbaK3sBaR 8cWDL/8SJrifS5LsJj5yDqByUcSEDeOC/p5m05uEEh32cWk1aq9uXHZeuyKwX6Sy 5Jgc3p6F6gYGoTR63nLDLW68s9iJ9PBRhLMlfYFezm+viIVklPOANZ+PEg0MlTo9 vEMxt4atNpI= =sPpy -----END PGP SIGNATURE-----
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu
| |