GSI Forum: FutureDAQ - Simulation » Ptolemy II perfomance

Home » FutureDAQ » FutureDAQ - Simulation » Ptolemy II perfomance

Show: Today's Messages :: Polls :: Message Navigator

Further tests of Ptolemy II Perfomance [message #374 is a reply to message #369]

Mon, 26 April 2004 18:13

Sergey Linev
Messages: 13
Registered: April 2004
Location: GSI Darmstadt

occasional visitor

From: depc150.gsi.de

I made further tests with Ptolemy II.

I create several models, which are just chain of standard TimedDelay actors. In the beginning I put Clock actor, which generate sequence of 0 and 1 with interval 1 sec. All these tokens just transferred further by each TimedDelay actors with delay of 1 sec. In the end all tokens are eaten by Discard actor. It can be replaced by Monitor actor to see token, which are arrived at the end.

There are four files:

File-----------actors---tokens---transf----exec---rate--memory
chain_10.xml-------10---100000--~1000000----14s--~70000---15M
chain_100.xml-----100----10000--~1000000----25s--~40000---16M
chain_1000.xml---1000-----1000---~500000----88s---~5000---28M-
chain_10000.xml-10000------100-----~5000-->1000s--~5?----168M

In this table:
actors - number of TimedDelay actors in chain
tokens - number of generated tokens by Clock actor
transf - total number of data transfers during run
exec - execution time on Athlon 1800M+, 512Mb RAM
rate - transfers per second
memory - used memory size as shown by top

First two files can be viewed and run with vergil, with other can be some problems.

I run these examples with ptexecute routine like:
> $PTII/bin/ptexecute chain_100.xml

Variable JAVAFLAGS was set to use 400 Mbytes memory for heap.
JAVAFLAGS=-Xmx400m

How I can explain and improve these results?

Attachment: chain_10.xml
(Size: 10.30KB, Downloaded 872 times)
Attachment: chain_100.xml
(Size: 76.04KB, Downloaded 846 times)
Attachment: chain_1000.xml
(Size: 739.52KB, Downloaded 914 times)

Report message to a moderator

[Message index]

		Ptolemy II perfomance By: Sergey Linev on Mon, 26 April 2004 10:13
		Re: Ptolemy II perfomance By: Elena Litvinenko on Mon, 26 April 2004 13:36
		Re: Ptolemy II perfomance By: Sergey Linev on Mon, 26 April 2004 15:00
		Further tests of Ptolemy II Perfomance By: Sergey Linev on Mon, 26 April 2004 18:13
		SystemC versus Ptolemy II perfomance By: Sergey Linev on Wed, 28 April 2004 11:17
		Re: SystemC versus Ptolemy II perfomance By: Ivan Kisel on Wed, 28 April 2004 16:33
		Re: SystemC versus Ptolemy II perfomance By: Sergey Linev on Wed, 28 April 2004 19:44
		Re: SystemC versus Ptolemy II perfomance By: Walter F.J. Müller on Wed, 28 April 2004 20:01
		Re: SystemC versus Ptolemy II perfomance By: Sergey Linev on Thu, 29 April 2004 08:37
		Re: Ptolemy II perfomance By: Walter F.J. Müller on Fri, 07 May 2004 21:45
		Re: Ptolemy II perfomance By: Krzysztof Korcyl on Thu, 13 May 2004 14:29
		Re: Ptolemy II perfomance By: Sergey Linev on Thu, 13 May 2004 18:14
		Re: Ptolemy II perfomance By: Krzysztof Korcyl on Fri, 14 May 2004 09:45
		Re: Ptolemy II perfomance By: Sergey Linev on Fri, 14 May 2004 10:09
		Re: Ptolemy II perfomance By: Elena Litvinenko on Fri, 14 May 2004 11:42
		Re: Ptolemy II perfomance By: Sergey Linev on Fri, 14 May 2004 12:46
		Re: Ptolemy II perfomance By: Elena Litvinenko on Fri, 14 May 2004 14:05
		Re: Ptolemy II perfomance By: Sergey Linev on Fri, 14 May 2004 14:51
		Re: Ptolemy II perfomance By: Walter F.J. Müller on Fri, 14 May 2004 12:52
		Re: Ptolemy II perfomance By: Krzysztof Korcyl on Tue, 25 May 2004 12:18
		Re: Ptolemy II perfomance By: Sergey Linev on Tue, 25 May 2004 12:47
		Re: Ptolemy II perfomance By: Krzysztof Korcyl on Tue, 25 May 2004 18:32
		Re: Ptolemy II perfomance By: Sergey Linev on Wed, 26 May 2004 08:56
		Re: Ptolemy II perfomance By: Krzysztof Korcyl on Tue, 08 June 2004 11:35

Previous Topic:

Ptolemy Classic

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Thu Dec 26 15:41:21 CET 2024

Total time taken to generate the page: 0.48935 seconds