A Block-Oriented, Parallel and Collective Approach to Sparse Indefinite Preconditioning on GPUs
| Autor: | Daniel Thuerck, Michael Goesele, Maxim Naumov, Michael Garland |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Optimization problem
Speedup Iterative method Computer science MathematicsofComputing_NUMERICALANALYSIS 010103 numerical & computational mathematics Parallel computing 01 natural sciences Precondition Parallel processing (DSP implementation) Symmetric matrix 0101 mathematics Block (data storage) Sparse matrix |
| Zdroj: | IA3@SC |
| Popis: | Large sparse symmetric indefinite matrices are notoriously hard to precondition. They often lack diagonal dominance and exhibit Schur-complements that render zero fill-in preconditioning ineffective. Pivoting, a necessity for stable LDLt factorizations, complicates parallel approaches that can take advantage of the latest massively-parallel HPC hardware such as GPUs. We present an approach based on ad-hoc blocking and reordering strategies that allows local, independent collective-oriented processing of small dense blocks. A hybrid block-memory layout compensates for irregular memory access patterns found in sparse matrices. Our method allows restricted fill-in, supernodal pivoting and a dual threshold dropping strategy at little additional cost. It delivers robust preconditioners that in our experiments obtain an average speedup of ~6x even for tough matrices from optimization problems. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|