Thanks a lot Ian and Gary for your answers. You give me useful
information which could help me.
I will try 1000MT cards to see if performance is improved.
Thanks,
Jorge
Ian McDonald escribió:
We have a testing environment, with a Intel(R) Pentium(R) D CPU 3.00GHz,
1 GB memory, a Realtek RTL-8169 NIC, where we want to reproduce the
improvement of PF_RING functionality. It also has a separated NIC for
administration. There is a Debian etch 32bits running. We will use
tcpreplay for injecting traffic (using a real capture from the real
environment)
In our testing environment, following the howtos about PF_RING, we have
downloaded the code, edited and run mkpatch... We have a comment here...
this script does the patch itself?? You don't need to execute "patch"
afterwards as the instructions say. Just a point.
We execute make menuconfig and select the options (rx and tx polling for
r8169 as a module, PF_RING sockets as a module) and generate the new
kernel which we boot (Debian like). No problem
I once tried to capture traffic with an r8169. I gave up very quickly,
and fitted a Pro-1000 card. I think at first I jammed a server card in
the 32-bit slot in the machine I had, and it worked, until a few 1000-MT
desktop cards arrived. It made a huge difference on the test hardware we
had.
I'm successfully capturing 5 minute averages in excess of 500Mb @~
80kpps on a PE1850, in 32 bit mode, on an 82541 Intel GE, using ring 3.7.7.
# cat /proc/net/pf_ring/info
Version : 3.7.7
Bucket length : 1514 bytes
Ring slots : 8192
Slot version : 9
Capture TX : No [RX only]
IP Defragment : No
Transparent mode : No
Total rings : 2
Total plugins : 0
--
Jorge Cuevas González <[EMAIL PROTECTED]>
Director de Proyectos www.nesys-st.com
Tfno: 94 406 0546 Móvil: 617165161
GPG: 3C1C CE59 3258 07C6 BB44 76B2 7861 9F93 9710 0C6F
_______________________________________________
Ntop mailing list
[email protected]
http://listgateway.unipi.it/mailman/listinfo/ntop
