On 09/20/2010 11:05 AM, Kerry Bonin wrote:
On the C++ Broker, just remember that the technology choice made for
clustering (OpenAIS/Corosync) is a Linux only solution, and that
codebase (~100k SLOC) was not written for portability.
Our project needs a Windows option, and this is already hurting us. I
(and my team) would be willing to spend some cycles supporting any
reasonable effort to address this, even if its not C++...
On portability: I think it would not be difficult to abstract the C++ cluster to
run over a different virtual synchrony library. If there is such a library on
windows it would probably pay to make the existing code portable rather than
fork a completely new cluster solution. It would be beneficial to have more
people testing the same core cluster code rather than spreading the effort over
separate code bases, and very beneficial to have more eyes on that code.
I'm getting together some thoughts on how the C++ cluster architecture code
could be restructured for greater performance and maintainability which might be
of interest. The general idea is to replicate a smaller and more clearly defined
subset of broker state and to remove the need for identical ordering of messages
on replicated queues. I'll try to get something written up this week.
---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project: http://qpid.apache.org
Use/Interact: mailto:dev-subscr...@qpid.apache.org
