Re: [NSE] DB2 library and scripts

[David Fifield <david () bamsoftware com>]

On Tue, May 11, 2010 at 07:57:34PM +0200, Patrik Karlsson wrote:
> Hi again,
>
> I'm attaching yet another version of the DB2-brute script.
> In this version, I've used the excellent code provided by Patrick (thanks!) and changed the behavior/design according 
> to Fyodor's suggestions [1].
> I'm also providing some *very* simple benchmarking of this new script and the old single threaded one running against 
> DB2 on Linux.
>
> The benchmarking was performed against a Linux DB2 installation running in Virtualbox. 
> The default usernames and passwords files were used, which results in 50840 username and password combinations.
>
> These are the results of the old single threaded script:
> abuse:nmap-dev patrik$ ./nmap -p 60000 192.168.56.5 --script db2-brute-st --script-args db2-brute.dbname=haxxoree 
> Nmap done: 1 IP address (1 host up) scanned in 142.75 seconds
> Nmap done: 1 IP address (1 host up) scanned in 135.33 seconds
> Nmap done: 1 IP address (1 host up) scanned in 125.51 seconds
> Average: 134,53
> TPS: 377,91
>
> These are the results of the new attached script (running 10 threads):
> abuse:nmap-dev patrik$ ./nmap -p 60000 192.168.56.5 --script db2-brute --script-args db2-brute.dbname=haxxoree 
> Nmap done: 1 IP address (1 host up) scanned in 107.44 seconds
> Nmap done: 1 IP address (1 host up) scanned in 105.29 seconds
> Nmap done: 1 IP address (1 host up) scanned in 110.32 seconds
> Average: 107,68
> TPS: 472,14
>
> So this rudimentary benchmark suggests a 25% performance increase when
> running with 10 threads.

I have an idea to sanity-check these results. Make 10 copies of the
unparallelized one-thread db2-brute script, and run all 10 copies
against the same server. If the process is 100% parallelizable on the
server, then it will take the same time to run all 10 in parallel as it
takes to run 1. If it is 0% parallelizable, it will take 10 times as
long. What we expect to see is a time around 10 times what you measured
with 10 thread, or about 1100 seconds. If it's much different than that,
then you're probably hitting some different bottleneck on the client.

Does that make sense?

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