Hi Bill, I went back yesterday to collect some numbers for you. These are taken using the cache option so we can see best case scenario.
Replaying a single 440 byte packet: tcpreplay-3.3.2 -i eth1 -q -t -l 10000000 -K single.pcap Actual: 10000000 packets (4400000000 bytes) sent in 43.52 seconds Rated: 101081024.0 bps, 771.19 Mbps/sec, 229729.61 pps CPU was pretty well flat out. Replaying a single 1500 byte packet: tcpreplay-3.3.2 -i eth1 -q -t -l 1000000 -K single.pcap Actual: 1000000 packets (1500000000 bytes) sent in 12.27 seconds Rated: 122217080.0 bps, 932.44 Mbps/sec, 81478.05 pps CPU usage was about between 40 and 50% The hardware that I'm running this on is nothing particularly special. It's about a 2Ghz P4 with about 1gig of ram. It does have a pci-express x4 gigabit ethernet card though. (Syskonnect, from memory) I'll try and get back to you with some figures when not using the cache option. cheers, Andrew Edgecombe William Cobau wrote: > Andrew, > > Average packet size is about 450 bytes with a packet rate or 100k > packets/second. > > So what type of box do you use? How much disk? How long can you > saturate a link with 1k sized packets? > > Bill Cobau > > On May 26, 2008, at 6:29 PM, Andrew Edgecombe wrote: > > >> William Cobau wrote: >> >>> Tcpreplay community, >>> >>> I am already successfully using tcpreplay for my testing activities. >>> Our setup is rather simple with a Linux server (typically Redhat EL5) >>> connected directly via 1000 B/T to the system under test. The linux >>> server uses tcpreplay (presently v3.2.5, but we could/will upgrade to >>> v3.3.x) to send previous captured data streams to the system under >>> test. There is no need for tcprewrite or the other tools in the >>> tcpreplay suite as the tcpreplay box is simulating a mirror port >>> or a tap. >>> >>> We are trying to enhance our testing by increasing amount of data and >>> the data rate. The data was captured at roughly 350 Mbps over 3 hours >>> for 450 Gbytes/1e9 packets. >>> >>> Has anyone built or bought a box that handles this data rate and >>> volume? Any suggestions or comments would be welcome. >>> >>> Bill Cobau >>> >> Hi William, >> What sort of traffic are you trying to replay? In particular, what >> sort >> of packet sizes are you using? >> I've found in testing that the packet rate that I'm able to achieve is >> more or less consistent regardless of the packet size. Meaning that if >> I'm replaying large packets (eg. 1k) I can saturate a gigabit link >> (even >> without the cache option), but if I'm replaying small packets (eg. >> 58bytes) I'll see figures more like 80mb/s. >> >> I second Peter Van Epp's comments about pci-express. If possible >> I'd go >> for a pci-express x4, or better, card. >> >> Andrew Edgecombe >> >> ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Tcpreplay-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/tcpreplay-users Support Information: http://tcpreplay.synfin.net/trac/wiki/Support
