High Performance Polar Decomposition on Distributed Memory Systems

Dalal E. Sukkari, Hatem Ltaief, David E. Keyes

Research output: Chapter in Book/Report/Conference proceedingConference contribution

6 Scopus citations


The polar decomposition of a dense matrix is an important operation in linear algebra. It can be directly calculated through the singular value decomposition (SVD) or iteratively using the QR dynamically-weighted Halley algorithm (QDWH). The former is difficult to parallelize due to the preponderant number of memory-bound operations during the bidiagonal reduction. We investigate the latter scenario, which performs more floating-point operations but exposes at the same time more parallelism, and therefore, runs closer to the theoretical peak performance of the system, thanks to more compute-bound matrix operations. Profiling results show the performance scalability of QDWH for calculating the polar decomposition using around 9200 MPI processes on well and ill-conditioned matrices of 100K×100K problem size. We study then the performance impact of the QDWH-based polar decomposition as a pre-processing step toward calculating the SVD itself. The new distributed-memory implementation of the QDWH-SVD solver achieves up to five-fold speedup against current state-of-the-art vendor SVD implementations. © Springer International Publishing Switzerland 2016.
Original languageEnglish (US)
Title of host publicationEuro-Par 2016: Parallel Processing
PublisherSpringer Nature
Number of pages12
ISBN (Print)9783319436586
StatePublished - Aug 9 2016

Bibliographical note

KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: For computer time, this research used the resources from the Swiss National Supercomputing Centre (CSCS) in Lugano, Switzerland.


Dive into the research topics of 'High Performance Polar Decomposition on Distributed Memory Systems'. Together they form a unique fingerprint.

Cite this