[Spread-users] Problem with multiple network interfaces

John Lane Schultz jschultz at spreadconcepts.com
Thu Sep 13 10:24:48 EDT 2007


Answers below inline.

Cheers!

---
John Lane Schultz
Spread Concepts LLC
Phn: 443 838 2200 
Fax: 301 560 8875

Thursday, September 13, 2007, 10:06:58 AM, you wrote:

> Just two other questions:
> ----------------------------------

> I get around 70Mb/s, which is good on a 100Mb/s network.
> Nevertheless, what is surprising me is that I Spread achieves the  
> same performance for  reliable, fifo, total, agreed and safe delivery
> semantics.

> Does it make sense for you?

Yes, these different orderings should really only affect latency of delivery
and not throughput.

> 2) Moreover, running the same benchmark on another cluster, I get  
> very bad performance (around 3Mb/s). Did you observe similar  
> behaviors? I suspect these bad results to be the consequence of the  
> fast switch droping multicast packets.

With performance that bad, you are probably correct that your switch
or router is dropping ALL multicast packets.  Spread falls back to
unicast recovery in such a case on a packet by packet basis, but this
causes huge performance impacts.

Once you fix the problem with your switch, or reconfigure your spread
system to match the network that you are given (e.g. - no
broad/multicast), you should see much better performance.






More information about the Spread-users mailing list