Seema Hiranandani

Learn More
Fortran D, a version of Fortran extended with data decomposition speciications, is designed to provide a machine-independent data-parallel programming model. This paper describes analysis, optimization , and code generation algorithms employed in the Fortran D compiler. The compiler rst partitions programs using the owner computes rule. It then performs(More)
The Fortran D compiler uses data decomposition specifications to automatically translate Fortran programs for execution on MIMD distributed-memory machines. This paper introduces and classifies a number of advanced optimizations needed to achieve acceptable performance; they are analyzed and empirically evaluated for stencil computations. Profitability(More)
An effective data-parallel prog ra m m ing environment will use a variety of tools that support the development of efficient data-parallel p rog ra ms w h ile insulating the programmer from the intricacies of the explicitly parallel code. ver the past decade, researchers have sought to provide an efficient machine-independent programming system that(More)
26] Max Lemke and Daniel Quinlan. P++, a C++ virtual shared grids based programming environment for architecture-independent development of structured grid applications.grid representation of emission source clusters in regional air quality modeling. 31] R. Meakin. Moving body overset grid methods for complete aircraft tiltrotor simulations, AIAA-93-3350. A(More)
\EEcient implementation of distributed barrier synchronization in hypercube multicomputers. \EEcient implementation of barrier synchronization in wormhole-routed hypercube multicomputers. 17 Current research projects address the implementation of higher-level SPMD operations and parallel eigensolvers based on the algorithms presented in this paper.(More)
This paper presents Fortran D, a version of Fortran enhanced with data decomposition spec-iications. It is designed to support two fundamental stages of writing a data-parallel program: problem mapping using sophisticated array alignments, and machine mapping through a rich set of data distribution functions. We believe that Fortran D provides a simple(More)
Because of the complexity and variety of parallel archi-tectures, an eecient machine-independent parallel programming model is needed to make parallel computing truly usable for scientiic programmers. We believe that Fortran D, a version of Fortran enhanced with data decomposition speciications, can provide such a programming model. This paper presents the(More)
Massively parallel MIMD distributed-memory machines can provide enormous computation power. However, the difficulty of developing parallel programs for these machines has limited their accessibility. This paper presents compiler algorithms to automatically derive efficient message-passing programs based on data decompositions. Optimizations are presented to(More)