GSoC/GCI Archive
Google Summer of Code 2010


Web Page:

Mailing List:

OMII-UK cultivates, develops and supports a wide range of open source software aimed at the UK research community but available for anyone to use. OMII-UK software spans a wide range such as data access, integration and sharing (OGSA-DAI), workflow enactment and tools (Taverna), video conferencing and collaboration (PAG, RAT and VIC), data transport (DataMINX), portals (RAPID) and computational job execution (GridSAM) and APIs (SAGA) enabling better use and sharing of resources across organisational boundaries to turn ideas into knowledge and promote collaborations between organisations and individuals. OMII-UK has a user-focused approach, encouraging the development of communities that use OMII-UK software components together with other components in novel ways to enable world class research. We work with communities providing information for climate policy, analysing the genetic causes of disease, understanding ancient texts, modelling the interactions of carbohydrates in biofuels, investigating the causes of asthma, simulating social transport scenarios, visualising aerial earthquake images, and creating virtual models of the human physiology.

Our project code repository for GSoC 2010 is here:


  • Alternative presentation layers for OGSA-DAI OGSA-DAI is a middleware bundle for high scale data integration across the network, enabling collaboration by sharing data resources - accessing, transforming, integrating, and delivering them. Its presentation layer is SOAP based and is derived from the outdated Axis 1.4 stack. Benchmarks have shown that CXF outperforms Axis and Axis2 in binary data transfer. Hence a web service presentation layer based on CXF along with an alternative RESTful presentation layer using CXF JAX-RS is preferred.
  • Extension of Rapid to the Apache Hadoop Distributed Computing Framework The project idea is to extend Rapid with code that facilitates the submission, monitoring and execution of jobs through Hadoop Framework. Rapid is a technology to quickly create web portal interfaces to execute applications on large remote clusters. Apache Hadoop is a framework for running applications on large clusters, implementing Map/Reduce and a distributed file system (HDFS). The project goal is to add a module to Rapid, such that it can communicate with the Apache Hadoop Framework.
  • Reimplementing OGSA-DAI DQP query compiler The goal of this project is to reimplement in Clojure the query compiler and query optimisers of the OGSA-DAI Distributed Query Processor while ensuring interoperability with the broader Java framework the query processor is a part of.