[kepler-users] Kepler API access

[Matt Jones]

Hi,

There are two mechanisms to get data out of the SRB in Kepler.  One is
to use the native SRB actors that communicate with the SRB directly.

The other is to use the EML Data Source actor to retrieve a metadata
document that has pointers to objects in the srb.  The EML Data Source
actor will then take care of retrieving the data for you from the SRB,
and it is cache aware so it will place any retrieved objects in the
cache.  This makes sure that you only download a given object once, and
the workflow won't execute until the objects have been cached.

The native SRB actors are not cache aware but could probably be modified
to utilize the cache.

I'm not sure what else you might mean by "stage" other than download the
data.  You also mention "notification mechanisms to enable execution" --
I'm not quite sure what you mean but if you are saying that you want to
be notified when the download is complete then I believe that is taken
care of by the ptolemy engine -- the data is available when you find a
token on the output port of the SRB actor.

Hope this helps.  Maybe Efrat can clarify some of the SRB actor
functionality a bit better than I.

Matt


Vijayakumar, Nithya N. wrote:
> hi Matt,
> 
> I am interested in using Kepler to execute a data reduction workflow where the original raw data goes through a series of 
> applications before producing the final output. Our workflow is a data flow model and we are using SRB for data management. A part of our need is to stage the required data for different parts of the workflow and notification mechanisms to enable execution. We will probably need the Kepler-SRB API to stage data. It will be very helpful if you can advise me on what other modules of Kepler will be useful for this purpose and what would be an efficient mechanism to implement the data reduction workflow using Kepler.
> 
> thank you,
> Nithya
> 
> 
> -----Original Message-----
> From: Matt Jones [mailto:jones at nceas.ucsb.edu]
> Sent: Thu 7/7/2005 11:31 AM
> To: Vijayakumar, Nithya N.
> Cc: kepler-users at ecoinformatics.org
> Subject: Re: [kepler-users] Kepler API access
>  
> Hi Nithya,
> 
> We certainly do have an API, in fact many, depending on which part of
> Kepler you wish to programatically control.  For example, the
> CacheManager has an API for manipulating objects in the cache.  Is there
> a something specific you are trying to do?  If so, we might be able to
> be more specific than a simple "yes".
> 
> Glad to hear you liked Kepler's functionality.
> 
> Matt
> 
> Vijayakumar, Nithya N. wrote:
> 
>>Hi,
>>
>>I am new user to Kepler and I am impressed with its functionality. I am curious to know if you have an API access to Kepler? 
>>
>>thank you,
>>Nithya
>>
>>
>>Nithya Vijayakumar
>>Oak Ridge National Laboratory
>>_______________________________________________
>>Kepler-users mailing list
>>Kepler-users at ecoinformatics.org
>>http://mercury.nceas.ucsb.edu/ecoinformatics/mailman/listinfo/kepler-users
> 
> 

-- 
-------------------------------------------------------------------
Matt Jones                                     jones at nceas.ucsb.edu
http://www.nceas.ucsb.edu/    Fax: 425-920-2439    Ph: 907-789-0496
National Center for Ecological Analysis and Synthesis (NCEAS)
University of California Santa Barbara
Interested in ecological informatics? http://www.ecoinformatics.org
-------------------------------------------------------------------