activemq-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From egetchell <eric.getch...@distributedlogic.com>
Subject Re: Defining Multiple Network Connectors in a Network of Brokers
Date Mon, 14 Jan 2008 15:42:19 GMT

Reading the Network of Brokers page, the following is what sent us down this
path:

"It is possible to have more than one network connector between two brokers.
Each network connector uses one underlying transport connection, so you may
wish to do this to increase throughput, or have a more flexible
configuration."

I interpreted the above to a design similar to, say telnet, FTP, or SSH. 
You can have multiple connections established from a single client.  Would
it be possible to provide clarification on the above?

During testing we seem to be saturating broker A at around 2000 messages a
second, but the CPU of the box was only at 30%.  Our solution was to add in
another "A" broker, yielding the following:

A1 --->
           B
A2 --->

And this seemed to increase our throughput.  Mind you this is a very
simplified version of our topology.

Thanks,

Eric


We don't support multiple  directional connections between two brokers  
- i.e.
We support:

    --->
A        B
    <---

but not

  ---->
A      B
   --->


-- 
View this message in context: http://www.nabble.com/Defining-Multiple-Network-Connectors-in-a-Network-of-Brokers-tp14744240s2354p14804190.html
Sent from the ActiveMQ - User mailing list archive at Nabble.com.


Mime
View raw message