Out-of-Core Computation of the QR Factorization on Multi-core Processors

Authors:
Mercedes Marqués;Gregorio Quintana-Ortí;Enrique S. Quintana-Ortí;Robert Geijn
Affiliations:
Depto. de Ingeniería y Ciencia de Computadores, Universidad Jaume I (UJI), Castellón, Spain 12.071;Depto. de Ingeniería y Ciencia de Computadores, Universidad Jaume I (UJI), Castellón, Spain 12.071;Depto. de Ingeniería y Ciencia de Computadores, Universidad Jaume I (UJI), Castellón, Spain 12.071;Department of Computer Sciences, The University of Texas at Austin, Austin 78712
Venue:
Euro-Par '09 Proceedings of the 15th International Euro-Par Conference on Parallel Processing
Year:
2009

Citing 9
Cited 4

A set of level 3 basic linear algebra subprograms

ACM Transactions on Mathematical Software (TOMS)
LAPACK's user's guide

LAPACK's user's guide
A survey of out-of-core algorithms in numerical linear algebra

External memory algorithms
The Design and Implementation of the Parallel Out-of-coreScaLAPACK LU, QR, and Cholesky Factorization Routines

The Design and Implementation of the Parallel Out-of-coreScaLAPACK LU, QR, and Cholesky Factorization Routines
POOCLAPACK: Parallel Out-of-Core Linear Algebra Package

POOCLAPACK: Parallel Out-of-Core Linear Algebra Package
Parallel out-of-core computation and updating of the QR factorization

ACM Transactions on Mathematical Software (TOMS)
Computational methods and processing strategies for estimating earth's gravity field

Computational methods and processing strategies for estimating earth's gravity field
Programming matrix algorithms-by-blocks for thread-level parallelism

ACM Transactions on Mathematical Software (TOMS)
Solving "large dense matrix problems on multi-core processors

IPDPS '09 Proceedings of the 2009 IEEE International Symposium on Parallel&Distributed Processing

Parallel direct methods for solving the system of linear equations with pipelining on a multicore using OpenMP

Journal of Computational and Applied Mathematics
Using desktop computers to solve large-scale dense linear algebra problems

The Journal of Supercomputing
Performance study of matrix computations using multi-core programming tools

Proceedings of the Fifth Balkan Conference in Informatics
Scaling LAPACK panel operations using parallel cache assignment

ACM Transactions on Mathematical Software (TOMS)

Quantified Score

Hi-index	0.01

Visualization

Abstract

We target the development of high-performance algorithms for dense matrix operations where data resides on disk and has to be explicitly moved in and out of the main memory. We provide strong evidence that, even for a complex operation like the QR factorization, the use of a run-time system creates a separation of concerns between the matrix computations and I/O operations with the result that no significant changes need to be introduced to existing in-core algorithms. The library developer can thus focus on the design of algorithms-by-blocks, addressing disk memory as just another level of the memory hierarchy. Experimental results for the out-of-core computation of the QR factorization on a multi-core processor reveal the potential of this approach.