[e-lang] Missing (?) E concurrency mechanism

Karp, Alan H alan.karp at hp.com
Mon Nov 29 13:21:48 EST 2004


Shap wrote:
> 
> In the vast majority of cases, this is a fine model, but in 
> cases where
> concurrent handling of requests is desired, this model unifies two
> things that ought properly to be separated: the message 
> rendezvous queue
> and the thread of control that receives the message.
> 
E-speak did something similar.  Messages were addressed to Inboxes, not
processes.  A process could, given the right capability, connect to more
than one Inbox.  A process could also detach an Inbox so that another
process could start processing its messages.  I don't know if there were
any determinism issues.  Most clients only used one Inbox, and those
that had more than one used them for disjoint activities.
 
We did it this way because of my bad experience with PVM message queues.
There was only one queue per process, which made it hard to handle
messages of different priorities.
 
________________________
Alan Karp
Principal Scientist
Virus Safe Computing Initiative
Hewlett-Packard Laboratories
1501 Page Mill Road
Palo Alto, CA 94304
(650) 857-3967, fax (650) 857-7029
https://ecardfile.com/id/Alan_Karp
http://www.hpl.hp.com/personal/Alan_Karp


-------------- next part --------------
A non-text attachment was scrubbed...
Name: Karp, Alan H.vcf
Type: text/x-vcard
Size: 433 bytes
Desc: Karp, Alan H.vcf
Url : http://www.eros-os.org/pipermail/e-lang/attachments/20041129/21dddcd4/KarpAlanH.vcf


More information about the e-lang mailing list