The Effects of Distribution and Parallelisation on the Performance of OGSA-DAI Enactments

Grid computing provides technologies for building high performance scientific and business applications. Grid applications need access to distributed and heterogeneous data sources and OGSA-DAI midd leware provides data services to access and integrate data sources. Paral lelism has been used extensively for designing efficient data and computation intensive applications. This project aims to explore paral lelism in OGSA-DAI services to achieve performance improvements for data applications designed using OGSA-DAI services. We build a performance analysis system to study the effect of paral lelism and distribution on processing time of data access queries. We use multiple threads with in a single process to paral lelise the requests. We distribute the requests by connecting multiple workflows submitted to different OGSA-DAI servers running on a single machine or
multiple machines.