c.d.wakelin at reading
Jul 4, 2012, 10:35 AM
Post #1 of 1
(220 views)
Permalink
|
|
pfsend/numademo on Dell R620
|
|
I just a got a couple of brand new Dell R620s (C600 chipset, 2x 8-core
E5-2660s, 32GB 1600MHz RAM, Ubuntu 12.04 64-bit 3.2.0-26 kernel) for my
network monitoring, and was testing out one of them before deployment
next week.
With reference to
http://www.ntop.org/pf_ring/not-all-servers-are-alike-with-dna/ :
numademo is a bit disappointing:
> 2 nodes available
> memory with no policy memcpy Avg 11367.16 MB/s Max
(for contrast I get ~32000 MB/s on an R210)
however the pfsend test seems just about OK with DNA.
I've got a dual-port Intel X520-DA2 adapter with the ports linked
together by a cable for testing (using PF_RING svn 5.4.4)
insmod pf_ring transparent_mode=2
insmod ixgbe RSS=16,16 mtu 1522
ifconfig dna0 up
ifconfig dna1 up
set_irq_affinity.sh dna0
set_irq_affinity.sh dna1
pfcount_multichannel -i dna0
pfsend -i dna1 -l 60 -n 0
gives
> [average 14'618'101.46 pps/9.82 Gbps]
If I use "pfsend -i dna1 -l 60 n 0 -z" for no zero-copy it drops to
> [average 6'586'393.36 pps/4.43 Gbps]
If I don't use multichannel on send (so RSS=16,0)
> [average 14'447'672.84 pps/9.71 Gbps]
If I don't use multichannel on either (RSS=0,0)
> [average 14'445'179.02 pps/9.71 Gbps] and
> [average 7'196'183.28 pps/4.84 Gbps] with -z
Any other interesting tests while I'm at it :) ?
Best Wishes,
Chris
--
--+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+-
Christopher Wakelin, c.d.wakelin [at] reading
IT Services Centre, The University of Reading, Tel: +44 (0)118 378 2908
Whiteknights, Reading, RG6 6AF, UK Fax: +44 (0)118 975 3094
_______________________________________________
Ntop-misc mailing list
Ntop-misc [at] listgateway
http://listgateway.unipi.it/mailman/listinfo/ntop-misc
|
