[Users] Dispatcher module algorithms

[Daniel-Constantin Mierla]

On 11/24/05 20:52, Douglas Garstang wrote:
> I'm trying to understand a little better the docs for the algorithms in the dispatcher module.
>  
> What exactly does 'hash over call-id', or 'hash over from uri' mean? The emphasis here is on the use of the word 'hash'.
the hash is an integer number computed from a value, based on different 
algorithm. For same value, the hash is the same. In this case a modulo 
to number of destinations is applied to get the entry in the destination 
set. An example of an hashing algorithm for string values is to add the 
ASCII codes of the characters in the string. In openser it is used a 
more complex hashing.
>  For a given call-id or a given from uri, I think I understand that the selected proxy will always be the same. However, is there any way to determine how this will distribute call volume between proxies?
You can do some statistics based on the traffic you have on your 
platform. Watch the dispatching by writing log messages and after 
several days see the results.

>  Is there some magic formula that is used (ie the hash) to select the proxy? I'd like to know what it is so I could run some numbers through it and try and understand how it would break up call placement to proxies for certain input values.
>   
For same value, the hash result is the same. Try with hash over From uri 
and make several calls.

Cheers,
Daniel
>  
> Doug.
>  
>   
> ------------------------------------------------------------------------
>
> _______________________________________________
> Users mailing list
> Users at openser.org
> http://openser.org/cgi-bin/mailman/listinfo/users
>