In Partial Fulfillment of the Requirements for the Degree of
Doctor of Philosophy
Will hold his pre-defense
MPI is the de-facto standard for message passing in parallel scientific applications. MPI-I/O is part of the MPI-2 specification defining file I/O operations in the MPI world. Some of the features that distinguish MPI-I/O from POSIX I/O include the notion of relaxed consistency semantics, file views, collective I/O operations, and shared file pointers. As of today, when it comes to meeting the needs of the HPC community, existing MPI-I/O libraries are limited in terms of performance, modularity, and portability over different hardware architectures and file systems. In this thesis we propose to accomplish the following goals: develop a novel and flexible architecture for an MPI-I/O library, develop new algorithms for collective I/O operations, develop static and dynamic tuning methodologies to tune for the parameters/algorithms that are ideal for a certain environment, and extend the notion of parallel I/O to deal with compressed data files.