Subject: RE: PID's etc.. (please excuse possible duplications of this)..
To: 'Scott.Burns@Netcontech.Com' <Scott.Burns@Netcontech.Com>
From: Andy Sporner <>
List: tech-kern
Date: 04/08/2000 10:01:02
From what I read of the project, the answer would be yes.
I will look more completely into this in the next few days.

Even though this has been done for the most part, as it
seems, I still think there is value in pursuing this for
NetBSD, as it is a rapidly moving target and Sprite seems
to be largely dead.

Does anyone have any input on where to bury the node ID
in the process structure?  Or would it be better to have
an encapsulation structure that is in the cluster region
of code that provides a pointer to the process structure
and contains the other cluster related stuff (As I write
this, this seems actually like a good idea).

What are your thoughts...


> -----Original Message-----
> From: Scott Burns [mailto:Scott.Burns@Netcontech.Com]
> Sent: Friday, April 07, 2000 9:17 AM
> To: 'Andy Sporner';
> Subject: RE: PID's etc.. (please excuse possible duplications 
> of this)..
> Would this be simlilar to something like Sprite ?
> Scott Burns
> -----Original Message-----
> From:	Andy Sporner []
> Sent:	Thursday, April 06, 2000 11:30 PM
> To:
> Subject:	PID's etc..   (please excuse possible 
> duplications of this)..
> Hi, 
> I am working on a clustering initiative that allows
> for processes to be 
> swapped from one machine to another.  The idea is for
> load balancing 
> processes in a NUMA like approach.  
> What I had thought was to bury the cluster ID in the
> high order byte of 
> the process ID so that whenever a system call was made
> any such byte 
> greater than 0 would be proxied to the appropriate
> machine. 
> I would accept suggestions on this.   What would break
> as a result of 
> doing this? What were the historical reasons for
> this,? 
> Thanks! 
> Andy 
> __________________________________________________
> Do You Yahoo!?
> Talk to your friends online with Yahoo! Messenger.