Re: Performances are not scale with multiple ports

[Thomas Monjalon <thomas.monjalon-pdR9zngts4EAvxtiuMwx3w@public.gmane.org>]

Hello,

28/05/2013 04:26, Shinae Woo :
> ./build/l2fwd -cf -n3 -- -p3
> ./build/l2fwd -cf -n3 -- -pf
> 
> But both cases, the aggregated performance are not scale.
[..]
> The question is that
> 1. How I can achieve each port receiving full 14.88Mpps ?
>     What might be the bottleneck in current environment?
> 2. Why the performance using multiple ports is not scale?
>     I guess doubling ports shows the doubling the receiving performance,
>     but it shows not. I am curious about what is limiting the packet
> receivng performance.

In order to scale with line-rate performance, you probably have to adjust the 
configuration to your hardware. Several parameters should be considered:
- PCIe bus (is it fast enough ?)
- NUMA sockets (copy between NUMA nodes could slow forwarding)
- CPU threads (classic SMP is better than hyperthreading)

To make it short, you could check your PCI bus and your CPU configuration.
For the latter, /proc/cpuinfo should provide enough information with fields 
"physical id" and "core id".

-- 
Thomas