Image courtesy Jay Lopez, stock.xchng

The recent trend towards Service Oriented Architecture — collections of services which operate according to a request/reply model — has stimulated the development of Workflow Management Systems (WMSs) which target the composition of services such as Taverna and Triana.

A new, grid-enabled scientific workflow management system, WS-VLAM, developed in the context of the Virtual Laboratory for e-Science, provides a basic set of tools for building workflows by connecting components to each other based on data dependencies.

The WS-VLAM workflow management system is designed to provide and support the coordinated execution of geographically distributed grid-enabled software components, which can be combined into a workflow. The system takes advantage of the underlying grid infrastructure and unites it with a flexible, high-level, rapid prototyping environment for scientific applications. WS-VLAM breaks down an application into a collection of services and software components, which communicate with one another only when the input data becomes available.

SigWin is an application developed using WS-VLAM.

WS-VLAM

The main features of WS-VLAM are:
• An interface to develop and port existing applications in C, C++, JAVA, and Python languages
• Streaming facilities between applications that are executed in geographically distributed resources
• Support for legacy application, and the execution of hierarchical workflows
• Support for remote graphical output
• Tools to define, annotate and search for workflow components
• The capability to detach/attach for long running workflows
• Monitoring facilities based on the WS-notification
• Workflow farming possibilities

The traditional batch processing of grid jobs and workflow execution based on file exchange between the components is not suitable in some scenarios. In contrast, WS-VLAM supports the simultaneous execution of co-allocated processes on the grid, which enables direct data streaming between distributed components, making it highly useful for near real-time distributed applications, such as bio-medical research and online video processing and analysis.
 
Performance tests indicate that the overhead of WS-VLAM is negligible compared to the standard Globus tool ‘globus-url-copy,’ which uses the GridFTP protocol. It is currently deployed on the Dutch Distributed Supercomputer 3, but WS-VLAM could also be used for any grid (Globus) enabled system.

—Adam Belloum, for iSGTW. Edited by Seth Bell