I want to make a benchmark for the statespace generator (similar to
dining philosophers). So I need the NACs...
Am 30.11.2010 11:55, schrieb Stefan Jurack:
Good as well! This is the approach with the least effort.
Anyhow, the benchmark result shall be compared to equal
transformation setting (same rules, same start graph, same
machine) but different EObject implementations.
greetx, Stefan
Am 30.11.2010 11:44, schrieb Enrico Biermann:
Hello,
For limiting the iteration depth, I suggest just changing the do
... while(true)-loop to a for-loop in the
SierpinskiBenchmark.java with the desired depth. Adding
application conditions or additional attributes might distort
the benchmark results.
Bye,
Enrico
On 30.11.2010 09:20, Christian Krause wrote:
Hi,
I believe Enrico tested that already once and he did not
manage to get a 'level' further. Anyway: I'm using the minimal
implementation also for the state space model, and it should
indeed save quite some memory.
About Sierpinski: I wanna do another benchmark but for that I
need to limit the maximum iteration depth. I figured the
easiest way to do that is to add an integer attribute to the
nodes that I increase on every step. Then I can just add a NAC
so that it will stop at a certain depth. I guess I rather make
a copy of the Sierpinski model and trafo for that, right?
Ciao,
Christian
Am 30.11.2010 10:05, schrieb Stefan Jurack:
Hi guys,
if there is anyone bored of sharpen pencils and stuff, he
might want to take a look at "MinimalEObjectImpl" (e.g. see
http://ed-merks.blogspot.com/2009/01/emf-ultra-slim-diet.html).
According to the reduction in size of EObjects the "Worst
case kinds of numbers are 124 bytes reduced to 16 bytes"
(citation of Ed M.).
If would be interesting to know, on the one hand,
how much is the decrease in performance and, on the
other hand, how much nodes can be handled by the
Sierpinski transformation example in such a case.
Best regards,
Stefan
_______________________________________________
henshin-dev mailing list
henshin-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/henshin-dev
_______________________________________________
henshin-dev mailing list
henshin-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/henshin-dev
_______________________________________________
henshin-dev mailing list
henshin-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/henshin-dev
_______________________________________________
henshin-dev mailing list
henshin-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/henshin-dev
|