News and Announcements
Jack Receives Ken Kennedy Award
On November 19th, Jack Dongarra received the ACM-IEEE Ken Kennedy Award at an SC13 ceremony in Denver, Colorado. The award recognizes Jack for his leadership in designing and promoting standards for mathematical software used to solve numerical problems common in HPC, and for his work in the development of major software libraries that boost performance and portability in HPC environments.
ACM President Vint Cerf cited Dongarra’s role in anticipating the staggering challenges facing the HPC world: “Jack saw the need to keep pace with the evolution in HPC hardware and software in a world that demands higher speeds and performance levels. His innovations have contributed immensely to the steep growth of high performance computing and its ability to illuminate a wide range of scientific questions facing our society.”
IEEE Computer Society President David Alan Grier said Dongarra’s work remains authoritative: “I’m so pleased to see this award go to Jack Dongarra because he did such foundational work in scientific computing. That work was important in my early career and it remains an influential body of work.”
TOP500 – November 2013
The 42nd TOP500 list was released on November 18th at SC13 in Denver, Colorado. The TOP5 machines are identical to the June 2013 list, with China’s Tianhe-2 maintaining the top spot (33.86 petaflop/s), followed by DOE’s Titan (17.59 petaflop/s).
There was some movement within the TOP10 machines, however, as Piz Daint, a Cray XC30 system installed at the Swiss National Supercomputing Centre (CSCS) in Lugano, Switzerland cracked the TOP10 at No.6. Piz Daint achieved 6.27 petaflop/s on the LINPACK benchmark, making it the fastest machine in Europe. Piz Daint is also the most energy efficient system in the TOP10, consuming a total of 2.33 MW and delivering 2.7 gigaflops/W.
Rounding out the TOP10 are Stampede at the Texas Advanced Computing Center of the University of Texas, Austin, which slipped to No. 7; JUQEEN, a BlueGene/Q system installed at the Forschungszentrum Juelich in Germany is No. 8; No. 9 is taken by Vulcan, another IBM BlueGene/Q system at Lawrence Livermore National Laboratory; and No. 10 is the third system in Europe, the SuperMUC, installed at Leibniz Rechenzentrum in Germany.
| Rank | Site | System | Rmax (TFlop/s) |
|---|---|---|---|
|
1 |
33,862.7 |
||
|
2 |
DOE/SC/Oak Ridge National Laboratory |
Titan – Cray XK7 |
17,590.0 |
|
3 |
DOE/NNSA/LLNL |
17,173.2 |
|
|
4 |
RIKEN Advanced Institute for Computational Science (AICS) |
K computer, SPARC64 VIIIfx |
10,510.0 |
|
5 |
DOE/SC/Argonne National Laboratory |
8,586.6 |
|
| See the full list at TOP500.org. | |||
NICS HPC Seminar Series
The National Institute for Computational Sciences invites you to a Seminar Series on High Performance Computing, every Tuesday and Thursday from 2:10pm to 3:10pm in Claxton 233, starting on Tuesday October 1st. This is a joint effort between different leadership organizations (NICS, JICS, OLCF, ICL) to increase HPC awareness within the academic community.
Different topics will be introduced starting with the most basic and building up to more advanced topics in HPC. No registration is required for the seminar.
Calendar of topics to be covered in December:
| Date | Title |
|---|---|
| 3 | How to run your Program |
| 5 | XSEDE Science Gateways |
Conference Reports
Supercomputing ’13
This year’s Supercomputing Conference was held in Denver, Colorado on November 17th – 22nd. A staple of ICL’s November itinerary, the lab routinely has a significant presence at SC, with faculty, research staff, and students giving talks, presenting papers, and leading BoF sessions; SC13 was no exception.
ICL was once again active in the University of Tennessee’s SC booth. The booth, which was organized and led by the National Institute for Computational Sciences (NICS), was visually designed with the help of ICL/CITR staff, manned with support from ICL researchers attending SC, and featured the lab’s research projects in the booth’s kiosks.
As is tradition, the ICLers both past and present who attended SC13 were invited to the Alumni Dinner. This year, the dinner was held at Tom’s Urban 24, and there were plenty of conversations shared between old friends and colleagues, as the ideas and drinks flowed freely. In the end, everyone had a good time as they capped off the last major conference of the year.
Recent Releases
SC13 Handouts
New project handouts from SC13 are available for download in PDF format.
FT-LA Update Released
An update to FT-LA is now available. FT-LA is an extension to ScaLAPACK that tolerates and recovers from failures. The interface is generally similar to ScaLAPACK, but a number of new operations are added. FT-LA now supports the following operations (in all sdcz precisions):
- QR factorization (protection against fail-stop failures, failure of at most 1 proc at a time, both Q and R factors protected);
- LU factorization (protection against fail-stop failures, failure of at most 1 proc at a time, both L and U factors protected).
Visit the FT-LA software page to download the tarball.
LAPACK 3.5.0 Released
LAPACK 3.5.0 is now available. This new version of LAPACK, released on November 13th, contains the following new features:
- Symmetric/Hermitian LDLT factorization routines with rook pivoting algorithm;
- 2-by-1 CSD to be used for tall and skinny matrix with orthonormal columns;
- New stopping criteria for balancing.
Visit the LAPACK software page for more information.
MAGMA MIC 1.1 Beta Released
MAGMA MIC 1.1 Beta is now available. This release provides implementations for MAGMA’s one-sided (LU, QR, and Cholesky) and two-sided (Hessenberg, bi- and tridiagonal reductions) dense matrix factorizations, as well as a linear and eigenproblem solver for Intel Xeon Phi Coprocessors. More information on the approach is given in this presentation.
The MAGMA MIC 1.1 Beta release adds the following new functionalities:
- LU, QR, and Cholesky factorizations and solvers with CPU interfaces;
- Performance improvements for the two-sided reductions to Hessenberg, bidiagonal, and tridiagonal forms;
- Eigensolvers for symmetric and non-symmetric eigenproblems;
- SVD routine;
- Added orthogonal transformation routines;
- General matrix inversion (routine {z|c|d|s}getri);
- Performance improvements for the one-sided factorizations using single and multiple MICs.
Visit the MAGMA software page to download the tarball.
MAGMA 1.4.1 Beta Released
MAGMA 1.4.1 Beta is now available. This release provides performance improvements and support for the new NVIDIA Kepler GPUs. More information is given in the MAGMA: a New Generation of Linear Algebra Libraries for GPU and Multicore Architectures presentation as well as the MAGMA Quick Reference Guide. The MAGMA 1.4.1 Beta release adds the following new functionalities:
- Improved performance of geev when computing eigenvectors using blocked trevc;
- Added new CMake installation for compiling on Windows.
Visit the MAGMA software page to download the tarball.
clMAGMA 1.1 Beta Released
clMAGMA 1.1 Beta is now available. clMAGMA is an OpenCL port of the MAGMA library. This release adds the following new functionalities:
- MultiGPU implementations for the LU, QR, and Cholesky factorizations;
- LU, QR, and Cholesky factorizations and solvers with CPU interfaces;
- Multi-buffer LU, QR, and Cholesky factorizations that overcome size limitations for single memory allocation, enabling the solution of large problems;
- Performance improvements.
Visit the MAGMA software page to download the tarball.
