A. SHIWA Concepts
A. SHIWA concepts #
1. What is SHIWA? #
SHIWA is a European FP7 project which aims at developing workflow systems interoperability technologies. SHIWA develop a Workflows Repository enabling sharing of executable workflow artifacts among user communities. It set up an execution environment (SHIWA Simulation Platform) to execute these workflows on various Distributed Computing Infrastructures (DCIs). It also enables the creation of meta-workflows composed of workflows from different workflow management systems and it develops techniques for workflow languages translation. More information about the project, the project partners and subcontractors can be found under the PROJECT menu.
2. What does SHIWA mean to me? #
The SHIWA platform is of interest for any Distributed Computing Infrastructure (DCI) user who need to share workflows among his/her community, compose workflows reusing existing sub-workflow components and execute workflow-based scientific experiments.
Still not clear? Read the SHIWA Comics to learn more in an easier way.
3. What are the main goals of SHIWA? #
SHIWA develop technologies to (1) develop workflows and expertise sharing among eScience research communities; (2) support eScientists in their in silico experiments; (3) develop new workflow interoperability solutions; (4) improve interoperability among distributed computing infrastructures; (5) simplify access to multiple computing infrastructures; and (6) promote the use of the European e-Infrastructure.
4. What is workflow interoperability? #
There exists many different scientific workflow management systems taylored towards different user communities and exeuction infrastructure targets. SHIWA enables the execution of workflow designed from different workflow environments from a single platform. Furthermore, it develops technologies to help relocation of workflow activities execution (migration over Distributed Computing Infrastructures), to assemble meta-workflow composed of heterogeneous sub-workflow components and to translate workflow programs into a common representation.
5. What is a Distributed Computing Infrastructure (DCI)? #
A Distributed Computing Infrastructure (DCI) is a high-end computing infrastructure composed of multiple computing resources communicating through a local or a wide area network. DCIs are increasingly used to support large-scale scientific experiments for their capability to deliver high throughput computing and to process large scientific data-sets efficiently. Among many DCI user communities, workflow management systems have been adopted as a standard and efficient mean of interfacing with the DCI. Workflow systems provide a language to formalize scientific experiments computational flow and an execution back-end interface to one or more DCI(s) to enact workflows efficiently exploiting the DCI resources.
6. What is Coarse-Grained Interoperability (CGI)? #
Coarse-Grained workflow Interoperability (CGI) is a workflow interoperability technique that makes it possible to assemble several sub-workflows into a master workflow in a hierarchical manner, thus allowing the reuse and the repurposing of existing workflows. CGI can accomodate sub-workflows from different workflow management systems and languages. The meta-workflow is written using one particular workflow system/language that interfaces to other, embedded workflow systems.
Read more on Gabor Terstyanszky, Tamas Kukla, Tamas Kiss, Stephen Winter, Peter Kacsuk, Akos Balasko Sharing Workflows through Coarse-Grained Workflow Interoperability In GRID'2012, 5th International Conference on Distributed Computing and Grid Technologies in Science and Education, Dubna, Russia, 17-21 July 2012
7. What is Fine-Grained Interoperability (FGI)? #
Fine-Grained workflow interoperability (FGI) is a workflow interoperability technique that consists in translating worflows from their native language to a common representation, namely IWIR. FGI is based on language converters for each supported workflow system. The challenge in FGI is to preserve the semantic of source workflows while transforming them into IWIR workflows.
Read more on: Kassian Plankensteiner, Radu Prodan, Thomas Fahringer, Johan Montagnat, David Rogers, Ian Harvey, Ian Taylor, Akos Balasko and Péter Kacsuk Fine-Grain Interoperability of Scientific Workflows in Distributed Computing Infrastructures ''Journal of Grid Computing''. September 2013, Volume 11, Issue 3, pp 429-455
8. What is the Interoperable Workflow Intermediate Representation (IWIR)? #
The Interoperable Workflow Intermediate Representation (IWIR) is a low-level workflow representation language designed to express as many constructs from different workflow languages as possible while remaining simple enough. It is designed as a machine-readable layer between the human-readable workflow language and the workflow enactment engine. Its primary focus was the support of the 5 workflow languages from the ASKALON MOTEUR, Pegasus, P-GRADE, and Triana workflow systems. IWIR workflows can then be enacted through IWIR-compliant workflow systems.
Read more on: K. Plankensteiner, J. Montagnat, R. Prodan IWIR: A Language Enabling Portability Across Grid Workflow Systems In ''Proceedings of 6th Workshop on Workflows in Support of Large-Scale Science (WORKS'11)'' as a part of Supercomputing'11 conference, Seattle, USA
9. What are Open Reuse and Exchange (ORE) bundles? #
The Open Reuse and Exchange (ORE) is a standard defined by the Open Archives Intiative for the description of Web Resources. In SHIWA ORE is used to create self-contained bundles describing executable workflow artifacts that can be exchanged among the SHIWA platform components such as Desktop clients and the SHIWA Simulation Platform.
10. Where can I find more support? #
The SHIWA services are described here. In particular, the documentation for the SHIWA Repository and the SHIWA Simulation Platform are linked from here. The SHIWA web-site also proposes a where different topics related to the exploitation of the SHIWA platform are discussed. In particular, a Troubleshooting category is available for reporting problems in the use of the paltform and getting assistance. Please read this FAQ carefuly before posting to the User Forum.
SHIWA (project number 261585) was supported by a Grant from the European Commission's FP7 INFRASTRUCTURES-2010-2 call under grant agreement n°261585
Provided by the MTA SZTAKI Computer and Automation Research Institute - Maintained by Laboratory of Parallel and Distributed Computing with Free and Open Source Software (2010)