[kepler-dev] Fwd: Re: Kepler distribution for Solaris
Amanda Waite
Amanda.Waite at Sun.COM
Mon Jul 2 09:04:16 PDT 2007
Hi Gang,
Thanks for the heads up, I'll check them out and let you know how they
build on Solaris.
Regards
Amanda
Gang Zhou wrote:
> Hi,
>
> My advisor Professor Lee forwarded me the following email. I just want to point out that I have developed some generic functionality in Ptolemy to do Java-C co-simulation. There are two actors involved in the $PTII/ptolemy/actor/lib/jni directory (you need to access current cvs):
>
> One is the CompiledCompositeActor which has two modes. When the "invokeJNI" parameter is false, it acts as a normal composite actor. When it is true, the CompiledCompositeActor generates C code for the contained submodel and compiles it to a DLL file. From then on, when the CompiledCompositeActor is fired, the corresponding DLL file is executed.
>
> The other is the EmbeddedCActor which is based on the CompiledCompositeActor and functions as a wrapper around some C code. It provides a user interface (try double-clicking on it) for entering and editing the C code. When the model is simulated, the C code representing the functionality of the associated EmbeddedCActor is compiled to a DLL file and executed. In order to use this actor, the C code needs to use macros to refer to ports and parameters. Note that since the EmbeddedCActor does not specify type constraints, you need to manually specify the types for its ports if the default does not work for you.
>
> There is a very simple demo in the $PTII/ptolemy/actor/lib/jni/demo/scale directory. These actors are still preliminary. I tested them under cygwin. Christopher mentioned making them work under Solaris. They probably won't work under other platforms (I appreciate using Java more since I worked on this). Also right now they only work for transferred tokens with primitive data types...
>
> Gang
>
>
>>> X-ASG-Debug-ID: 1183055264-7bd2002e0000-YtAf27
>>> X-Barracuda-URL: http://bc1.eecs.berkeley.edu:8000/cgi-bin/mark.cgi
>>> X-Original-To: kepler-dev at ecoinformatics.org
>>> Delivered-To: ecoinformatics-kepler-dev at mercury.nceas.ucsb.edu
>>> Date: Thu, 28 Jun 2007 11:27:23 -0700
>>> From: Dan Higgins <higgins at nceas.ucsb.edu>
>>> User-Agent: Thunderbird 2.0.0.4 (Macintosh/20070604)
>>> To: Amanda Waite - SUN ISV-Engineering <Amanda.Waite at Sun.COM>
>>> References: <4682AA57.4040809 at sun.com>
>>> X-NCEAS-Mercury-MailScanner: Found to be clean, Found to be clean
>>> X-NCEAS-MailScanner-MCPCheck: MCP-Clean, MCP-Checker (score=0,
>>>
>> required 1), MCP-Clean, MCP-Checker (score=0, required 1)
>>
>>> X-NCEAS-Mercury-MailScanner-SpamCheck: not spam, SORBS-DNSBL,
>>> SpamAssassin (not cached, score=-1.771, required 2,
>>> autolearn=not spam, AWL 0.69, BAYES_00 -2.60,
>>>
>> FORGED_RCVD_HELO 0.14), not spam, SpamAssassin (cached,
>>
>>> score=-1.777, required 2, AWL 0.69, BAYES_00 -2.60,
>>> FORGED_RCVD_HELO 0.14)
>>> Cc: Kepler-Dev <kepler-dev at ecoinformatics.org>
>>> X-ASG-Orig-Subj: Re: [kepler-dev] Kepler distribution for Solaris
>>> Subject: Re: [kepler-dev] Kepler distribution for Solaris
>>> X-BeenThere: kepler-dev at ecoinformatics.org
>>> X-Mailman-Version: 2.1.6rc4
>>> List-Id: "Development discussions for Kepler." <kepler-
>>>
>> dev.ecoinformatics.org>>List-Unsubscribe:
>> <http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/kepler-dev>,
>>
>>> <mailto:kepler-dev-
>>>
>> request at ecoinformatics.org?subject=unsubscribe>>List-Archive:
>> <http://mercury.nceas.ucsb.edu/ecoinformatics/pipermail/kepler-dev>
>>
>>> List-Post: <mailto:kepler-dev at ecoinformatics.org>
>>> List-Help: <mailto:kepler-dev-
>>>
>> request at ecoinformatics.org?subject=help>>List-Subscribe:
>> <http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/kepler-dev>,
>>
>>> <mailto:kepler-dev-
>>>
>> request at ecoinformatics.org?subject=subscribe>>Sender: kepler-dev-
>> bounces at ecoinformatics.org>X-NCEAS-MailScanner-Information: Please
>> contact the ISP for more information
>>
>>> X-MailScanner-From: kepler-dev-bounces at ecoinformatics.org
>>> X-Barracuda-Connect: mercury.nceas.ucsb.edu[128.111.220.18]
>>> X-Barracuda-Start-Time: 1183055264
>>> X-Barracuda-Virus-Scanned: by EECS Spam Firewall at EECS.Berkeley.EDU
>>> X-Barracuda-Spam-Score: 0.00
>>> X-Barracuda-Spam-Status: No, SCORE=0.00 using global scores of
>>>
>> TAG_LEVEL=2.1 QUARANTINE_LEVEL=1000.0 KILL_LEVEL=1000.0 tests=
>>
>>> X-Barracuda-Spam-Report: Code version 3.1, rules version 3.1.21236
>>> Rule breakdown below
>>> pts rule name description
>>> ---- ---------------------- ------------------------------
>>>
>> --------------------
>>
>>> Hi Amanda,
>>> With regard to your question about actors that call JNI,
>>>
>> there are 3
>>
>>> 'GARP' actors (GARPPresampleLayers, GARPAlgorithm, and
>>>
>> GARPPrediction)
>>
>>> that use a JNI connection to C++ code. A workflow that uses these
>>>
>> actors
>>
>>> is $KEPLER/demos/SEEK/garpModel_ImageJ.xml.
>>> Currently, we have the dynamic link libraries for these JNI
>>>
>> actors
>>
>>> in the $KEPELR/lib directory with libraries for Linux, Mac OSX,
>>>
>> and
>>
>>> Windows.
>>> Source code for these and several other JNI actors is in the
>>> $KEPLER/src/cpp directory.
>>>
>>> With regard to putting a precompiled Sun version of Kepler on
>>>
>> the
>>
>>> www.kepler-project.org site, it seems to me that this is just the
>>>
>> Sun
>>
>>> version of the current installers that on the web site now. I
>>>
>> built the
>>
>>> current installers and it would be easy to add a link to a Sun
>>> distribution. Matt Jones is the project manager for this and he
>>>
>> needs to
>>
>>> decide the mechanisms (i.e. you could send me a file, or you
>>>
>> could be
>>
>>> given access to the server). Matt is on vacation now. I'll talk
>>>
>> to him
>>
>>> when he returns.
>>>
>>> Dan Higgins
>>> NCEAS
>>>
>>> ----
>>>
>>> Amanda Waite - SUN ISV-Engineering wrote:
>>>
>>>
>>>> Hi there,
>>>>
>>>> I was originally asked to look at how we could make Kepler
>>>>
>> available on
>>
>>>> the Sun Grid at network.com and that's kind of morphed into,
>>>>
>> "what can
>>
>>>> we do to help the Kepler community with enabling Kepler for the
>>>>
>> Grid
>>
>>>> (Specifically Sun Grid Engine)", something that we are looking
>>>>
>> into. In
>>
>>>> the meantime I've been asked if I could build Kepler on Solaris
>>>>
>> 10 and
>>
>>>> make it available for people to download and install and run on
>>>>
>> Solaris
>>
>>>> without having to make any changes to scripts, etc. This is
>>>>
>> fairly
>>
>>>> straightforward as Kepler builds with no problems on Solaris 10
>>>>
>> and the
>>
>>>> workflow tests and examples all run to completion (with one
>>>>
>> exception
>>
>>>> that I'm still looking into). I feel that the best place for
>>>>
>> hosting a
>>
>>>> Solaris 10 distro would be on kepler-project.org and wanted to
>>>>
>> find out
>>
>>>> if this is possible and what would need to be in place to make
>>>>
>> it
>>
>>>> happen. An alternative option would be to make it available as
>>>>
>> a Soalris
>>
>>>> package from blastwave.org, but | feel that its better if at
>>>>
>> all
>>
>>>> possible, for someone wanting to run Kepler on Solaris to get
>>>>
>> it from
>>
>>>> the community site. Let me know if this is something that would
>>>>
>> be of
>>
>>>> interest to the community.
>>>>
>>>> BTW: Are there any workflows that test the actors that call
>>>>
>> into native
>>
>>>> code via JNI? I should be able to build the libraries on
>>>>
>> Solaris but I'd
>>
>>>> then need to test them.
>>>>
>>>> Kind Regards
>>>>
>>>> Amanda
>>>>
>>>>
>>> _______________________________________________
>>> Kepler-dev mailing list
>>> Kepler-dev at ecoinformatics.org
>>> http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/kepler-dev
>>>
>> ------------
>> Edward A. Lee
>> Chair of EECS and Robert S. Pepper Distinguished Professor
>> 231 Cory Hall, UC Berkeley, Berkeley, CA 94720-1770
>> phone: 510-642-0253, fax: 510-642-2845
>> eal at eecs.Berkeley.EDU, http://ptolemy.eecs.berkeley.edu/~eal
>>
>>
>>
--
Amanda Waite
ISV-Engineering OSS, Sun Microsystems
Tel: +44 (0)1252 420693
Mobile: +44 (0)7802 175732
More information about the Kepler-dev
mailing list