Performance of Java on FBSD vs. others...

[Nikos Ntarmos]

Hi again.

On Fri, Nov 10, 2006 at 11:49:33PM +0100, Arne H. Juul wrote:
> is it possible to post the program for others to test? (if there's no
> trade secrets or similar in it) I'd like to run it; I've been
> compiling my own version of the Java VM with various tweaks on FreeBSD
> 4 and this sounds like a better test than the programs I've been using
> so far.

Of course. No trade secret there... :) Grab [1]. It's an implementation
of [2] over [3]. Example command line arguments:
	java -classpath <all jars in .../lib>:classes \
		netcins.p2p.dhs.testing.DHSSingleMetricTest \
		1000 100000 64 32 10
where 1000 is the number of nodes in the simulated system, 100000 the
number of items to be added, and the other numbers are parameters of the
simulated application. It should take a while to do the whole hog, but
don't wait for it to finish. After all nodes are online, it prints out
the time it took it to generate them all. If it takes more than
200-220'' (win32 and linux times respectively), we're done. On my 1.7GHz
Pentium-M it's more like 800''...

BTW the given application has two modes of simulated execution: plain
event-driven simulation and simulation with network i/o over virtual
nodes... Well, the latter easily crashes my freebsd box any time i run
it for more than 2000 nodes, no matter how heavily i tweak my kernel
conf, boot.conf, sysctl.conf, et al.

\n\n

[1] http://ntarmos.dyndns.org/Misc/DHS-FreePastry-2.0b2.tgz
[2] N. Ntarmos, P. Triantafillou, and G. Weikum. "Counting at large:
	 Efficient cardinality estimation in Internet-scale data networks."
	 In Proc. ICDE '06.
[3] http://freepastry.org/