[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] Re: [dist-obj] Seven Fallacies Reference
Hi, Umar,
I would like to argue against "complete disconnection from the network ? In
that case it does not
matter as it is not part of the same distributed system anymore .... ".
Physical disconnection doesn't necessarily mean a logical disconnection.
For instance, one robot dispatcher has a list of tasks for a group of robots
to achieve, some task may leads to demise of some robots, a task once may
done, may not be reversed (like an explosion job).
Now the dispatcher deliver a task to one of the robot, the robot got it and
acked and go ahead to do the job, the dispatcher missed the ack, yet he can
no longer "undo" the dispatching, although to him, this deliver transaction
failed.
Does this make the case?
-Qingfeng
> You mean a complete disconnection from the network ? In that case it does
not
> matter as it is not part of the same distributed system anymore .... and
if
> you are only talking about temporary disconnection then obviously that is
like
> a partial failure of a distributed system and transaction will be
performed or
> rolled back when the device connects again.
>
> Cheers,
> ----------------------------------------------------------
> Computo Ergo Sum
> ----------------
>
> UMAR SAIF
>
> Ph.D. Student
> Systems Research Group Office Phone: 01223-(3) 34643
> Computer Laboratory Mobile Phone : 07780 - 687037
> University of Cambridge http://www.cl.cam.ac.uk/~us204
> ----------------------------------------------------------
>
>
>
==========================================================================
To manage your subscription, mailto:dist-obj-help@distributedcoalition.org
Archives, FAQ, etc. http://www.distributedcoalition.org/mailing_lists/