[openamq-dev] Testing Throughput Vs Latency

[Martin Sustrik]

Malinga,

The high peaks in the latency are mostly caused by the operating system. 
So, OS may choose to interrupt openAMQ at any moment and do some 
different job. It's very hard to avoid this behaviour. If you need 
guaranteed worse-case latencies, you should use some kind of real-time 
operating system, but even then, it's not a plug-and-play thing. 
Ensuring real-time behaviour is highly untrivial matter, actually it is 
a science in itself.

Martin

Malinga wrote:
> Hi Pieter,
> 
>     No we are not sending one message to many subscribers, our test 
> application, just send 150 bytes message to a queue in a continues while 
> loop at a high rate, and listening to that  messages  from the other end 
> (do nothing bu listening on that queue), and calculate the latency.
> 
>     Do you have any ideas how this happening, because  if one message 
> get delayed all of the other message will get delay.
> 
> -Malinga
> 
> ------------------------------------------------------------------------
> Looking for a deal? Find great prices on flights and hotels 
> <http://us.rd.yahoo.com/evt=47094/*http://farechase.yahoo.com/;_ylc=X3oDMTFicDJoNDllBF9TAzk3NDA3NTg5BHBvcwMxMwRzZWMDZ3JvdXBzBHNsawNlbWFpbC1uY20-> 
> with Yahoo! FareChase.
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> openamq-dev mailing list
> openamq-dev at lists.openamq.org
> http://lists.openamq.org/mailman/listinfo/openamq-dev