Hi Jeff, The 30 us is for 'small' packets. As packet size increases, latency also increases. For a normal mix of packet sizes, you should see latency around 50 us. However, 'normal' is also a bit subjective based on traffic flows in your network.
Sincerely, David. Jeff Bacon wrote: > > I am, however, left with one mystery. > > > How can the Cisco docs on a FWSM claim a 30-usec latency when clearly > it isn't capable of that, at least not in any configuration that I'm > aware of? Granted that it's all lies, damn lies, and marketing > material, but putting that number in the main data sheet on the main > product page for the card is a pretty ballsy lie even for Cisco. > > > > *From:* David White, Jr. (dwhitejr) [mailto:dwhit...@cisco.com] > *Sent:* Friday, June 03, 2011 12:04 AM > *To:* Jeff Bacon > *Cc:* Pete Templin; Peter Rathlev; cisco-nsp@puck.nether.net > *Subject:* Re: [c-nsp] cat6500/fwsm performance > > > > And here is a great doc TAC wrote up on single flow TCP performance > which should answer all your questions: > > https://supportforums.cisco.com/docs/DOC-12668 > > Sincerely, > > David. > > > _______________________________________________ cisco-nsp mailing list cisco-nsp@puck.nether.net https://puck.nether.net/mailman/listinfo/cisco-nsp archive at http://puck.nether.net/pipermail/cisco-nsp/
