On Wed, 25 Sep 2002 07:38, [EMAIL PROTECTED] wrote:
> Is anyone looking into the capability of clustering blocks? If phoenix is
> a server of servers, then the next logical step would be the transparent
> distribution of these servers across vm's. If clustering could be
> supported, then this would be a huge feature for phoenix application
> developers.
>
> Thoughts?
It is an interesting idea. The problem arises in the specifics. What does it
mean to cluster an application? Does it just mean that we have the same
applications deployed in each of the phoenix instances? Or perhaps we have
Blocks that are clustered? Some blocks will be easy to cluster (ie Stateless
blocks which have simple or pass-by-value/serializable types in their method
parameters) but some are impossible (ie ConnectionManager which references
native resources).
So I guess I would need to know more about what you are thinking ;)
--
Cheers,
Peter Donald
---------------------------------------------------
"Wise men don't need advice. Fools don't take it."
-Benjamin Franklin
---------------------------------------------------
--
To unsubscribe, e-mail: <mailto:[EMAIL PROTECTED]>
For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
