taverna-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Stian Soiland-Reyes <st...@apache.org>
Subject Re: Enquiry about the features of Taverna 2.5
Date Mon, 11 Apr 2016 15:30:19 GMT
Hi!

(Folks, Khalid help do lots of work on research object, wfprov and
wfdesc which we use in taverna-language and
taverna-engine/taverna-prov)

On 11 April 2016 at 15:17, Khalid Belhajjame <kbelhajj@googlemail.com> wrote:
> I would like to enquiry about some of the current features of Taverna. I
> have an answer for most of the question, but I would like to know the status
> with respect to the current version (2.5) of Taverna. I will be grateful if
> you could take the time to answer them.
>
> 1- In the current version of Taverna (2.5), I can see that the provenance
> can be saved as PROV, but not as OPM. I am guessing that OPM export was
> deprecated. Let me know if I am wrong. Regarding the PROV export, you are
> using TavernaProv isn’t it? If so, do you have TavernaProv documented
> somewhere? I would like to see how it compares to wfprov.


Yes, you are correct in that the OPM export was deprecated in 2.5 - it
was also less in match with what was happening in the workflow so it
was rather removed.

TavernaProv became part of Taverna from 2.5.  Documentation and example here:

https://github.com/apache/incubator-taverna-engine/tree/master/taverna-prov

(It's not yet fully integrated into Taverna 3)


TavernaProv is basically wfprov+prov+taverna - you will see that
everything is stated both with pure prov and in wfprov - which often
provides shortcuts.  For details such as at which iteration something
happened you have to use the tavernaprov aspects - which also adds a
cnt:chars for embedding smaller values, and checksums like
tavernaprov:sha512.


> 2- Are there any utilities/means that help users to either deploy workflows
> on the cloud or call services/use resource on the cloud? I remember the work
> of Sonja with Unicore. Is there any other work that was conducted on this
> front?

A bit related - we have a GSOC student working on Docker:

https://issues.apache.org/jira/browse/TAVERNA-879

As for actually managing cloud instances or grid jobs I am not aware
of any current initiatives. Perhaps Dmitry or Vadim Surpin knows of
something?

I think the http://www.vph-dare.eu/ project is using VPH Share and
local Grid infrastructure via Taverna Server - but I'm not sure of the
details.


> 3- is the CWL used somehow to allow for the exchange of workflow
> specifications between workflow systems?


In CWL we are discussing how we can use wfprov and some aspects of
tavernaprov. See
https://github.com/common-workflow-language/common-workflow-language/issues/84


We are working on CWL support in Apache Taverna this summer - we might
get three Google Summer Of Code students this year:

https://cwiki.apache.org/confluence/display/TAVERNADEV/2016-03+GSOC+2016

And I will pick up some work here as well for http://bioexcel.eu/ - my
use case is to use CWL and Docker with Open PHACTS and
Taverna/Galaxy/Knime - with deployment on the EGI grid/cloud:

http://slides.com/soilandreyes/2016-02-11-bioexcel-use-case-open-phacts


> 4-Is there anything that has been done in relation to initiatives such as
> iPython?


I think Alan might know about iPython and Jupyter..   Not sure if
anything was done since this:

http://www.taverna.org.uk/2014/05/20/calling-taverna-workflows-from-ipython-notebook/

Does the Taverna Player still support iPython in such a way?


-- 
Stian Soiland-Reyes
Apache Taverna (incubating), Apache Commons RDF (incubating)
http://orcid.org/0000-0001-9842-9718

Mime
View raw message