RE: OpenSSH hpn detection

["Dario Ciccarone (dciccaro)" <dciccaro () cisco com>]

No to be nit-picky, but :)

"for high-bandwidth, high-latency networks."

That isn't 100% true - check
http://www.psc.edu/networking/projects/hpn-ssh/papers/i2mm-hpn-ssh.ppt .
As the BDP is calculated based on BW and RTT, higher RTTs would surely
mean higher BDP values and bigger buffers - but even in high bandwith,
low-latency networks you should see an improvement on the transfer rate.

Dario


> -----Original Message-----
> From: nmap-dev-bounces () insecure org 
> [mailto:nmap-dev-bounces () insecure org] On Behalf Of Matt Selsky
> Sent: Tuesday, October 30, 2007 9:45 AM
> To: Kris Katterjohn
> Cc: nmap-dev () insecure org
> Subject: Re: OpenSSH hpn detection
>
> > Sorry this is more offtopic than not, but I just have to say that
> > HPN-SSH looks mighty interesting!  Is it really as cool as it looks?
>
> Yes, for high-bandwidth, high-latency networks.
>
> For example we were transferring files from Los Angeles to 
> New York over 
> Internet2.  Without HPN, scp was going at 900KB/s.  With the HPN 
> patches, scp sped up to 20MB/s.
>
>
> -- 
> Matt
>
> _______________________________________________
> Sent through the nmap-dev mailing list
> http://cgi.insecure.org/mailman/listinfo/nmap-dev
> Archived at http://SecLists.Org

_______________________________________________
Sent through the nmap-dev mailing list
http://cgi.insecure.org/mailman/listinfo/nmap-dev
Archived at http://SecLists.Org