Dear LAPACK and ScaLAPACK users,
LAPACK 3.4.0 and ScaLAPACK 2.0.0 were released on November 11, 2011.
What's new in LAPACK 3.4.0:
• xGEQRT: QR factorization (improved interface). Contribution by Rodney James (UC Denver). xGEQRT is analogous to xGEQRF with a modified interface which enables better performance when the blocked reflectors need to be reused. The companion subroutines xGEMQRT apply the reflectors.
• xGEQRT3: recursive QR factorization. Contribution by Rodney James (UC Denver). The recursive QR factorization enables cache-oblivious and enables high performance on tall and skinny matrices. See reference [1] below.
• xTPQRT: Communication-Avoiding QR sequential kernels. Contribution by Rodney James (UC Denver). These subroutines are useful for updating a QR factorization and are used in sequential and parallel Communication Avoiding QR. These subroutines support the general case Triangle on top of Pentagon which includes as special cases so-called Triangle on top of Triangle and Triangle on top of Square. This is the right-looking version of the subroutines and the subroutines are blocked. The T matrices and the block size are part of the interface. The companion subroutines xTPMQRT apply the reflectors.
• Doxygen documentation. LAPACK routine documentation has never been more accessible. See http://www.netlib.org/lapack/explore-html/.
What's new in ScaLAPACK 2.0.0:
• PxHSEQR: Nonsymmetric Eigenvalue Problem. Contribution by Robert Granat, Bo Kågström, Meiyue Shao (Umeå University and HPC2N), and Daniel Kressner (École Polytechnique Fédérale de Lausanne). Compute the eigenvalues of a nonsymmetric real matrix. Implement the parallel distributed Hessenberg QR algorithm with the small bulge multi-shift QR algorithm together with aggressive early deflation. See reference [1] below.
• PxSYEVR/PxHEEVR: MRRR (Multiple Relatively Robust Representations) algorithm. Contribution by Christof Voemel. Compute the eigenvalues of a symmetric/Hermitian matrix. Implement the parallel distributed MRRR algorithm. See reference [2] below.
• BLACS revamping. Contribution by Rodney James (UC Denver).
The LAPACK and ScaLAPACK team would like to thank all the users who contributed to this release.
(And bigger thank to those who not only provided bug report but also bug patches!)
For details, please see
- http://netlib.org/scalapack/scalapack-2.0.0.html
- http://www.netlib.org/lapack/lapack-3.4.0.html
To download ScaLAPACK 2.2.0: http://netlib.org/scalapack/scalapack.tgz
To download LAPACK 3.4.0: http://www.netlib.org/lapack/lapack.tgz
For support, you can use either the LAPACK/SCALAPACK forum at http://icl.cs.utk.edu/lapack-forum/ or the LAPACK mailing list lapack@cs.utk.edu
Regards,
The LAPACK and ScaLAPACK team