Robust Workflow Systems Flexible Geoprocessing Services Geo-enabled Model Web? - Geographical Information Systems: Trends and Technologies - page 190

perform parallel and scalable computations for data-intensive business
and scientifi c applications. The authors proposed the implementation of
specifi c MapReduce actors to support MapReduce applications into Kepler
scientifi c workfl ows.
Taverna-based Geo-enabled workfl ows
Taverna provides a processor of type arbitrary WSDL type , which is able to
call a single operation of a WSDL-based web service (Section “Taverna”).
Indeed, a simple way to integrate external services into Taverna is just
by providing a WSDL interface for these services. In the case of services
in the life sciences domain, Taverna provides a specifi c mechanism via
the BioCatalogue 27 tool, a registry for life sciences web services that is
accessible and discoverable from Taverna. One registers a web service
in the BioCatalogue and immediately it is made reachable from Taverna
through the corresponding processor type. Bhagat et al. (2010) described
the advantages of using BioCatalogue against WSDL-based services, as for
example the ability to annotate a service at several dimensions (functional,
profi le, provenance, etc.) and the use of lightweight response formats such
as JSON. 28
Regarding the support of OGC services, de Jesus et al. (2012) have
recently presented a WPS server implementation (based on pyWPS) 29 that is
able to automatically generate WSDL descriptions from WPS-based services.
In their tool, such WSDL descriptions are then used in conjunction with
a Taverna WSDL processor type to invoke the counterpart geoprocesing
web service from Taverna workfl ows. Rather than extending Taverna by
developing a new kind of OGC WPS processor type , de Jesus et al. (2012) took
the approach to augment WPS services with WSDL descriptions which are
natively supported in Taverna.
It is worthwhile to note that Taverna workfl ows are mostly addressed
to bioinformatics activities. 30 Indeed, one of Taverna's key values is the
availability of bioinformatics services already integrated in the platform
(Deelman et al. 2009). For example, Hull et al. (2006) described how
bioinformatics can use hundreds of web services without prior programing
knowledge. Taverna provides a built-in mechanism for discovery and access
to a wide range of web service repositories (Hull et al. 2006) such as the
aforementioned BioCatalogue. Kawas et al. (2006) exploited the extension
capacity of Taverna by implementing a plug-in named BioMobu, whose
27 http://www.biocatalogue.org/
28 JavaScript Object Notation: http://www.json.org
29 https://github.com/geopython/PyWPS
30 http://www.taverna.org.uk/introduction/taverna-in-use/