Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [science-iwg] KNIME and Eclipse 4.3M6

Hi,
 
Just to backup Erwin, Dawn has found the Passerelle workbench a very useful feature and I would recommend anyone to using it from an algorithm perspective. We see it as a key part of data analysis along with python and mathematical expressions (using JEXL).
 
We use it as a high level graphical tool for organising algorithms and have created user interface actors which allow people to plot things and pass data along the pipeline using our plotting API. Python actors are also popular and we are working on integration of pydev and Passerelle at the moment to make python actors debuggable inside the workflow. The actors are open source and available at: https://github.com/DawnScience/dawn-workflow/tree/master/org.dawb.passerelle.actors
 
Actors should not have UI dependencies and although ours don’t functionally, they are in the same plugin with some UI. Therefore at some point we have to refactor this plugin to separate ui and actors into separate plugins.
 
Best Regards,
 
Matt
 
 
 
From: science-iwg-bounces@xxxxxxxxxxx [mailto:science-iwg-bounces@xxxxxxxxxxx] On Behalf Of Philip Wenig
Sent: 16 April 2013 11:56
To: science-iwg@xxxxxxxxxxx; Erwin de Ley
Subject: Re: [science-iwg] KNIME and Eclipse 4.3M6
 
Hi Erwin,
 
cool, thanks for your explanations. I'll give it a try in the next days.
 
 
Cheers,
Philip
 
Am 16.04.2013 10:14, schrieb Erwin de Ley:
Hi Philip, Andy, all,
 
Passerelle is a specialization/extension of the Ptolemy II framework of Univ of Berkeley : http://ptolemy.berkeley.edu/ptolemyII/
Another related project, for specific scientific domains  and currently still with a larger users-group than Passerelle (but this will hopefully change ;-) ), is Kepler : https://kepler-project.org/
 
Within this family of workflow-related systems, the basic components are "actors".
There are R & Matlab actors available in Ptolemy and Kepler since a long time.
We have not used them yet in our own work, but they should be straightforward to include in the Passerelle setup or derived/extended tools like DAWN.
 
At iSencia we typically integrate Passerelle in a server-based solution that is integrated with a relational database.
Some basic reporting options are supported via web-UIs then.
And complex reporting needs are then solved via standard SQL and reporting tools like BIRT etc.
 
So we do have some experience with BIRT usage and integration, but more focused on web-apps
And it is not in a same ready-to-use state in the RCP workbench as with KNIME.
 
 
What could be positive drivers to choose for Passerelle, if you're looking for a tool for scientific workflows?
- the underlying Ptolemy framework as a solid base for advanced workflow systems
- Passerelle builds on this to offer extra features, both technical and functional, e.g. :
-- integration with OSGi/Equinox
-- dedicated process/workflow-related features
-- transparent concurrency, support for asynchronous tasks etc
-- the architecture has been defined from the ground up to be open and modular
-- proven on multiple UI environments : RCP, Swing, web. All based on a common modular platform
- the availability of DAWN as a proven RCP-based scientific workbench, integrating Passerelle with a huge features-set and open for further extensions and specializations
- also the active DAWN community is certainly a plus. Depending on your exact needs you may be able to join/reuse DAWN?
- Passerelle has been used since years for different domains: from data acquisition & control for physical experiments to high-throughput server systems for automated diagnostics and repair.
 
What you should be aware of :
- Passerelle itself is not a finished package for specific scientific domains, but a generic actor-based workflow infrastructure.
- We offer a "vanilla" RCP workbench, mainly to illustrate possibilities.
- I.e. Passerelle is more oriented towards developer communities that can then use it to build a specialized end-user package.
- DAWN is the perfect example of this approach, if your're looking at RCP workbenches.
 
Hope this gives you already some useful info for your evaluation. We can certainly go in more detail if you want!
 
Kind regards
erwin
 
 
Op 16/04/2013 07:54, Philip Wenig schreef:
Hi Andy,
 
you're right. The usage of Swing in KNIME is indeed a drawback. But I really admire in KNIME the R- and Matlab-Support as well as the rich reporting functionality.
 
@Erwin:
Is there a R-/Matlab-Support in Passerelle? If no, how hard is it to get it running?
 
 
Cheers,
Philip
 
Am 16.04.2013 07:38, schrieb Andy Gotz:
 
Hi Philippe,
 
we had a look at Knime in the past and yes it looks cool but we finally decided to use Passerelle (http://code.google.com/a/eclipselabs.org/p/passerelle/) for workflows inside Dawn. Dawn integrates Passerelle and provides the visualisation and Python scripting support. It would be great if you could try Passerelle and let us know whether it satisfies your needs. Passerelle is open source so you can modify it.
 
One down side we found with Knime when we looked at it was the weak integration in eclipse. It still used Swing for the UI then. I don't know if this has been ported to SWT since then. And yes you are right Knime does not fit our definition of Open Source either 8-(
 
Regards
 
Andy
 
p.s. I have put the main developer of Passerelle in copy - Erwin
 
On 16/04/2013 07:06, Philip Wenig wrote:
 
Hi folks,
 
has anyone experience with KNIME (http://www.knime.org)?
 
I think it's a cool software and offers it great possibilities to create
customized evaluations and reports. There's a version for Eclipse 3.7.2
and 3.8, but I can't get it work under latest release 4.2.2 or 4.3M6.
I'd really like to use it, and I've tried to get answers in the KNIME
forum, without success:
 
 
 
 
I'm also willing to adapt the source code to the latest Eclipse release
(if possible), but there seems to be no access to the source code. They
advertise it as an open source project. Some people say I'm too tight,
but that's not the meaning of open source. Am I wrong?
 
 
Cheers,
Philip
 
 
 
--
Met vriendelijke groeten - Bien à vous - Kind regards
 
Erwin De Ley
________________________________________
Tel. +32 9 335 22 10
fax: +32 9 335 22 19
erwin.de.ley@isencia.beiSencia Belgium
Voorhavenlaan 31 bus 11
B-9000 Gent
 
 
 
 
 
 
 
 
 
_______________________________________________
science-iwg mailing list
science-iwg@xxxxxxxxxxx
 
 
--
~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Philip Wenig
 
 
~~~~~~~~~~~~~~~~~~~~~~~~
 

 

-- 

This e-mail and any attachments may contain confidential, copyright and or privileged material, and are for the use of the intended addressee only. If you are not the intended addressee or an authorised recipient of the addressee please notify us of receipt by returning the e-mail and do not use, copy, retain, distribute or disclose the information in or attached to the e-mail.
Any opinions expressed within this e-mail are those of the individual and not necessarily of Diamond Light Source Ltd.
Diamond Light Source Ltd. cannot guarantee that this e-mail or any attachments are free from viruses and we cannot accept liability for any damage which you may sustain as a result of software viruses which may be transmitted in or with the message.
Diamond Light Source Limited (company no. 4375679). Registered in England and Wales with its registered office at Diamond House, Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0DE, United Kingdom
 



Back to the top