@techreport {, title = {Prospectus for the Next LAPACK and ScaLAPACK Libraries: Basic ALgebra LIbraries for Sustainable Technology with Interdisciplinary Collaboration (BALLISTIC)}, journal = {LAPACK Working Notes}, number = {297, ICL-UT-20-07}, year = {2020}, month = {2020/07}, publisher = {University of Tennessee}, abstract = {The convergence of several unprecedented changes, including formidable new system design constraints and revolutionary levels of heterogeneity, has made it clear that much of the essential software infrastructure of computational science and engineering is, or will soon be, obsolete. Math libraries have historically been in the vanguard of software that must be adapted first to such changes, both because these low-level workhorses are so critical to the accuracy and performance of so many different types of applications, and because they have proved to be outstanding vehicles for finding and implementing solutions to the problems that novel architectures pose. Under the Basic ALgebra LIbraries for Sustainable Technology with Interdisciplinary Collaboration (BALLISTIC) project, the principal designers of the Linear Algebra PACKage (LAPACK) and the Scalable Linear Algebra PACKage (ScaLAPACK), the combination of which is abbreviated Sca/LAPACK, aim to enhance and update these libraries for the ongoing revolution in processor architecture, system design, and application requirements by incorporating them into a layered package of software components{\textemdash}the BALLISTIC ecosystem{\textemdash}that provides users seamless access to state-of-the-art solver implementations through familiar and improved Sca/LAPACK interfaces.}, author = {James Demmel and Jack Dongarra and Julie Langou and Julien Langou and Piotr Luszczek and Michael Mahoney} } @techreport {1174, title = {2016 Dense Linear Algebra Software Packages Survey}, journal = {University of Tennessee Computer Science Technical Report}, number = {UT-EECS-16-744 / LAWN 290}, year = {2016}, month = {2016-09}, publisher = {University of Tennessee}, abstract = {The 2016 Dense Linear Algebra Software Packages Survey was administered from January 1st 2016 to April 12 2016. 234 respondents answered the survey. The survey was advertised directly to the Linear Algebra community via our LAPACK/ScaLAPACK forum, NA Digest and we also directly contacted vendors and linear algebra experts. The breakdown of respondents was: 74\% researchers or scientists, 25\% were Principal Investigators and 25\% Software maintainers or System administrators. The goal of the survey was to get the Linear Algebra community opinion and provide input on dense linear algebra software packages, in particular LAPACK, ScaLAPACK, PLASMA and MAGMA. The ultimate purpose of the survey was to improve these libraries to benefit our user community. The survey would allow the team to prioritize the many possible improvements that could be done. We also asked input from users accessing these libraries via 3rd party interfaces, for example MATLAB, Intel{\textquoteright}s MKL, Python{\textquoteright}s NumPy, AMD{\textquoteright}s ACML, and many others.}, author = {Jack Dongarra and Jim Demmel and Julien Langou and Julie Langou} } @article {, title = {Accelerating Scientific Computations with Mixed Precision Algorithms}, journal = {Computer Physics Communications}, volume = {180}, year = {2009}, month = {2009-12}, pages = {2526-2533}, abstract = {On modern architectures, the performance of 32-bit operations is often at least twice as fast as the performance of 64-bit operations. By using a combination of 32-bit and 64-bit floating point arithmetic, the performance of many dense and sparse linear algebra algorithms can be significantly enhanced while maintaining the 64-bit accuracy of the resulting solution. The approach presented here can apply not only to conventional processors but also to other technologies such as Field Programmable Gate Arrays (FPGA), Graphical Processing Units (GPU), and the STI Cell BE processor. Results on modern processor architectures and the STI Cell BE are presented.}, doi = {https://doi.org/10.1016/j.cpc.2008.11.005}, author = {Marc Baboulin and Alfredo Buttari and Jack Dongarra and Jakub Kurzak and Julie Langou and Julien Langou and Piotr Luszczek and Stanimire Tomov} } @article {icl:392, title = {Exploiting Mixed Precision Floating Point Hardware in Scientific Computations}, journal = {In High Performance Computing and Grids in Action (to appear)}, year = {2007}, month = {2007-00}, publisher = {IOS Press}, address = {Amsterdam}, author = {Alfredo Buttari and Jack Dongarra and Jakub Kurzak and Julien Langou and Julie Langou and Piotr Luszczek and Stanimire Tomov}, editor = {Lucio Grandinetti} } @article {icl:395, title = {Mixed Precision Iterative Refinement Techniques for the Solution of Dense Linear Systems}, journal = {International Journal of High Performance Computer Applications (to appear)}, year = {2007}, month = {2007-08}, author = {Alfredo Buttari and Jack Dongarra and Julien Langou and Julie Langou and Piotr Luszczek and Jakub Kurzak} }