Domain-specific web services for scientific application developers

Many scientists today routinely conduct simulations and run models using high performance computing (HPC) resources provided by national, regional and campus grid infrastructures, either directly logging into such resources or indirectly via web portals and other application client software, i.e. science gateways. Science gateways have proven to be an effective way of bringing HPC resources to a much larger user base. However, developing a new gateway demands substantial effort, requiring both cyberinfrastructure and domain expertise. This paper describes an approach by core software developers to provide domain specific web services (DomWS) to science gateway and application developers. The core developers deliver a set of web services that understand domain specific semantics and translate domain application requests into computational and other tasks that are dispatched onto the distributed high end resources. This approach allows gateway and application developers to easily create customized application services and focus on the gateway functions and interfaces, such as gadgets and desktop clients, that serve their user community. The DomWS approach is not a generic solution for all simulation services. Our aim is to help application developers apply this framework with ease to their specific models and simulations with a small degree of code modification. DomWS also provides reusable modules and templates for common tasks needed by the targeted user community. This paper describes a set of web services that have been developed for Community Climate System Model (CCSM) version 4 using this approach.

[1]  T. Oinn,et al.  Soaplab - a unified Sesame door to analysis tools , 2003 .

[2]  K. Taylor,et al.  The Community Climate System Model , 2001 .

[3]  Michael E. Papka,et al.  A Web 2.0-Based Scientific Application Framework , 2010, 2010 IEEE International Conference on Web Services.

[4]  Jae-Woo Lee,et al.  Puffin: a fast and efficient web service-based condor job manager , 2009, GCE '09.

[5]  Shantenu Jha,et al.  Grid Interoperability at the Application Level Using SAGA , 2007, Third IEEE International Conference on e-Science and Grid Computing (e-Science 2007).

[6]  Thierry Delaitre,et al.  High-level grid application environment to use legacy codes as OGSA grid services , 2004, Fifth IEEE/ACM International Workshop on Grid Computing.

[7]  Yi Huang,et al.  Building web services for scientific grid applications , 2006, IBM J. Res. Dev..

[8]  Herman Lam,et al.  A service-oriented, scalable approach to grid-enabling of legacy scientific applications , 2005, IEEE International Conference on Web Services (ICWS'05).

[9]  Kim K. Baldridge,et al.  Opal: SimpleWeb Services Wrappers for Scientific Applications , 2006, 2006 IEEE International Conference on Web Services (ICWS'06).

[10]  Rob van Nieuwpoort,et al.  The Grid Application Toolkit: Toward Generic and Easy Application Programming Interfaces for the Grid , 2005, Proceedings of the IEEE.

[11]  Donald F. Ferguson,et al.  The WS-Resource Framework , 2004 .

[12]  Nancy Wilkins-Diehr,et al.  Science gateway, portal and other community interfaces to high end resources , 2006, SC.

[13]  Bertram Ludäscher,et al.  Kepler: an extensible system for design and execution of scientific workflows , 2004 .

[14]  Nancy Wilkins-Diehr,et al.  TeraGrid Science Gateways and Their Impact on Science , 2008, Computer.