Introduction to algorithms
Partitioning of processor arrays: a piecewise regular approach
Integration, the VLSI Journal - Special issue on algorithms and architectures
Automating non-unimodular loop transformations for massive parallelism
Parallel Computing
A singular loop transformation framework based on non-singular matrices
International Journal of Parallel Programming
Beyond unimodular transformations
The Journal of Supercomputing
Generation of Efficient Nested Loops from Polyhedra
International Journal of Parallel Programming - Special issue on instruction-level parallelism and parallelizing compilation, part 2
Scheduling and Automatic Parallelization
Scheduling and Automatic Parallelization
Multirate VLSI Arrays and Their Synthesis
IEEE Transactions on Computers
Loop Transformation Using Nonunimodular Matrices
IEEE Transactions on Parallel and Distributed Systems
Extension Of The Alpha Language To Recurrences On Sparse Periodic Domains
ASAP '96 Proceedings of the IEEE International Conference on Application-Specific Systems, Architectures, and Processors
Code Generation in the Polyhedral Model Is Easier Than You Think
Proceedings of the 13th International Conference on Parallel Architectures and Compilation Techniques
Conference record of the 33rd ACM SIGPLAN-SIGACT symposium on Principles of programming languages
Memory optimization by counting points in integer transformations of parametric polytopes
CASES '06 Proceedings of the 2006 international conference on Compilers, architecture and synthesis for embedded systems
Finding Synchronization-Free Parallelism Represented with Trees of Dependent Operations
ICA3PP '08 Proceedings of the 8th international conference on Algorithms and Architectures for Parallel Processing
Integer affine transformations of parametric ℤ-polytopes and applications to loop nest optimization
ACM Transactions on Architecture and Code Optimization (TACO)
Hi-index | 0.00 |
The polyhedral model is a well developed formalism and has been extensively used in a variety of contexts viz. the automatic parallelization of loop programs, program verification, locality, hardware generationand more recently, in the automatic reduction of asymptotic program complexity. Such analyses and transformations rely on certain closure properties. However, the model is limited in expressivity and the need for a more general class of programs is widely known. We provide the extension to ⁰-polyhedra which are the intersection of polyhedra and lattices. We prove the required closure properties using a novel representation and interpretation of ⁰-polyhedra. In addition, we also prove closure in the ⁰-polyhedral model under images by dependence functions---thereby proving that unions of LBLs, widely assumedto be a richer class of sets, is equal to unions of ⁰-polyhedra. Another corollary of this result is the equivalence of the unions of ⁰-polyhedraand Presburger sets. Our representation and closure properties constitute the foundations of the ⁰-polyhedral model. As an example, we presentthe transformation for automatic reduction of complexity in the ⁰-polyhedral model.