Introduction to High-Performance Computing

Speakers

Lilit Axner

studied and obtained her M.Sc. in Mathematics at NADA (currently CSC), here at KTH. Afterwards she obtained her Ph.D. in computer science--with emphasis on Computational Fluid dynamics--at the University of Amsterdam, the Netherlands. To continue her passion in computer science, she immediately started to work as a consultant at SARA Computing and Networking Services, Amsterdam, the Netherlands. During her two years of employment there, she was actively involved in the DEISA and HPC-Europa2 European Union projects as well as other Dutch national HPC projects. Together with her colleagues at SARA, she was involved in application support for users as well. While working at SARA in the Netherlands, she has studied and obtained her M.B.A. at Blekinge Institute of Technology, Sweden.

Lilit recently joined PDC as an application expert. Her main objectives will continue to be the DEISA and PRACE European Union projects as well as Swedish e-Science projects such as ScalaLife and SeRC. She will also provide overall application support by working closely with users.

David Black-Schaffer

received his PhD in electrical engineering from Stanford University in 2008 focusing on parallel programming systems for many-core processors. After that he worked for at Apple designing and developing the first implementation of the new OpenCL specification for heterogeneous parallel processing on CPUs and GPUs. He is currently an assistant professor at Uppsala University looking at parallel programming systems and runtimes.

Iris Christadler

Iris Christadler is a member of the scientific staff at the Leibniz Computing Centre in Munich, Germany, where she is responsible for HPC user support and performance optimization of user codes on HPC systems. She is interested in debugging and profiling tools, and is trying to "hyperscale" codes within the European Grid Initiative DEISA.

Iris Christadler studied Computer Science at the University of Erlangen-Nuremberg, Germany. Her studies, which she finished in 2004, were already centered on the fields of HPC computing, fluid dynamics, and multigrid methods. She wrote a term paper about patch-adaptive relaxation used as a smoother in a multigrid solver. The subject of her diploma thesis (MSc equivalent) was multigrid algorithms for the computation of optical flow problems using non-standard regularizers. She is currently working on highly optimized multigrid methods for image registration problems in close cooperation with Prof. Ruede at the University of Erlangen.

Björn Engquist

is professor in mathematics at University of Austin, Texas. Formerly, he has had similar positions at Princeton University and the University of California, Los Angeles. Professor Engquist is also a professor of Numerical Analysis and Computing Science at the Royal Institute of Technology in Stockholm (KTH) since 1992. He received his B.S. in 1966 and the Ph.D. in 1975 from Uppsala University, Sweden. Between 1966 and 1985, he held positions at the Courant Institute, the University of California at Berkeley, Stanford University, and Uppsala University. He is the chairman of KTH Center for Computational Science and Engineering and former director of Parallel and Scientific Computing Institute (PSCI) at KTH and Uppsala University and the Center for Parallel Computers (PDC), KTH. His principal research interests are the mathematics and algorithms of scientific computing.

Thomas Ericsson

is working at Chalmers University of Technology. He is an expert in high-performance computing and numerical analysis, in particular in numerical linear algebra.

Erik Hagersten

Uppsala University, teaches courses in computer architecture. His research focus is "Increase data processing speed through adopting architectures and [coherent] data replication." He works in both academia and industry in Sweden and the US, and has initiated a collaborative research program between Uppsala University and Sun's Engineering in the U.S.

Mike Hammill

PDC, has been working as a consultant, teacher, and coordinator in high-performance computing since 1984. Before coming to PDC, he was coordinating the national high-performance consulting program ("Smart Nodes") at the Cornell Theory Center, Cornell University, USA.

Sverker Holmgren

received a Ph.D. in Numerical Analysis from Uppsala University in 1993. Since 2006, Holmgren is the Director of the Swedish National Infrastructure for Computing (SNIC). SNIC is the national computing infrastructure initiative within the Swedish Research Council, with a mandate to provide high performance computing, grid and large-scale storage resources for academia in Sweden and to coordinate development activities in these areas. Holmgren is also the chairman for the Swedish National Graduate School in Scientific Computing (NGSSC). Since 2007, Holmgren is the Head of the Uppsala University program in Computational Science and Engineering, where he leads several research projects in CSE and large-scale computing.

Niclas Jansson

is a Ph.D. student in numerical analysis at CSC, KTH since 2008, and received an M.S. degree in computer science from KTH in 2008. His research is focused on parallel adaptive finite element methods.

Lennart Johnsson

is the Hugh Roy and Lillie Cranz Cullen Distinguished Professor of Computer Science, Mathematics, and Electrical and Computer Engineering at the University of Houston.

He received his M.S. degree in engineering physics in 1967 and the Ph.D. degree in Control Engineering in 1970 from Chalmers Institute of Technology, Gothenburg, Sweden. At ASEA AB (now ABB), from 1971 to 1979, he led the development of real-time supervisory data acquisition and control systems for electric utilities and process industries. He was on the faculty of Caltech from 1979 to 1983, where he initiated a course in parallel scientific computation, taught VLSI design, and did research in parallel algorithms and VLSI design. From 1983 to 1990, he was an associate professor of Computer Science and of Electrical Engineering at Yale University, where he introduced courses in parallel scientific computation and also in computer arithmetic. He also led the research effort that resulted in the acquisition of the first Connection Machine at a university after MIT. He was Gordon McKay Professor at Harvard University from 1990 to 1997.

From 1986 to 1995, Johnsson also served as the director of Computational Sciences at Thinking Machines Corporation and was responsible for the development of the Connection Machine Scientific Software Library (CMSSL) as well as parts of the Connection Machine Run-Time System (CMRTS). Since 1995, he has been Cullen Distinguished Professor of Computer Science, Mathematics, and Electrical and Computer Engineering at the University of Houston; an adjunct professor of Computer Science at Rice University; and a visiting professor of Computing Sciences and Numerical Analysis at the Royal Institute of Technology, Stockholm. He also serves as the director of both the Texas Learning and Computation Center at the University of Houston and the Houston Area Computational Science Consortium. Johnsson serves on the executive board of the W. M. Keck Center for Computational Biology in Houston and the Los Alamos Computer Science Institute; he is chair of the PDC External Advisory Committee and the Swedish National Allocations Committee for High-Performance Computation; and he is the editor of seven journals.

Pavel A. Korzhavyi

Pavel A. Korzhavyi, docent in theoretical materials science at the Department of Materials Science and Engineering, KTH. He studied and obtained his Candidate and Doctor of Sciences degrees in solid state physics at the Moscow Institute of Steel and Alloys (MISIS). The research interests include electronic structure theory of disordered alloys, statistical mechanics of defects in metals and semiconductors, and multiscale modeling of materials. He is currently involved in the Excellence Center Hierarchic Engineering of Industrial Materials (HERO-M) and Generation-IV Industry-University Swedish consortium (GENIUS).

Erwin Laure

is the director of PDC-HPC. Before joining PDC in 2008 he was the Technical Director of the EU funded project "Enabling Grids for E-Science in Europe (EGEE)" working at the European Organization for Nuclear Research (CERN). After joining CERN in 2002 he worked on data management issues within the EU DataGrid (EDG) project, became the Technical Coordinator of EDG, and coordinated the middleware re-engineering activities in the first phase of EGEE. He holds a PhD in Business Administration and Computer Science from the University of Vienna, Austria. His research interests include grid computing with a focus on data management in grid environments as well as programming environments, languages, compilers and runtime systems for parallel and distributed computing.

Michaela Lechner

has studied Physics and Astronomy at the Institute of Astro- and Particle Physics in Innsbruck, Austria, and was employed several years within the Austrian Grid project. She started at PDC in autumn as a part-time replacement for application expert Ulf Andersson. After finishing her Ph.D. in November she began working full time at PDC on January 1, 2009. During her first months she mainly contributed to maintenance and setup, especially of the next SweGrid cluster, and honed her Grid-operation skills. Her new focus is to provide better overall application support by working more closely with users. In particular, she will be the contact person for the astrophysics community. Her goal is to facilitate and optimize the usage of PDC's HPC systems and the Grid for the users.

Dag Lindbo

is a Ph.D. student in numerical analysis at CSC, KTH since 2007, and received an M.S. degree in engineering physics from KTH in 2006. He has been involved as a tutor in the PDC summer course since 2007.

Elisabet Molin

received a M.Sc. in Applied Physics and Electrical Engineering at Linköping University in 2002. Developing and assessing computational models of physical phenomena has enabled her to participate in research projects at both CERN, KTH and KI. Joining PDC in 2008, she brings a prospective user's perspective in combination with an interest and experience in teaching.

Jesper Oppelstrup

is an assistant professor in the Department of Numerical Analysis and Computing Science (NADA) at KTH. He received an M.S. degree in 1969 in engineering physics, and the Ph.D. degree in numerical analysis in 1976, both at KTH. He was a research project manager at the Swedish Institute of Applied Mathematics from 1977 to 1985, and he has worked in commercial development of finite element software since 1984. He began working at KTH in 1985, and he is currently an assistant professor in numerical analysis. Oppelstrup is engaged in the development of high-performance computing resources and industry/academic collaborative research in high-performance computing and mathematical modeling, especially for fluid dynamics and electromagnetics. He is a member of the national supercomputer resource allocation committee, he is on the Program Board of the Parallel and Scientific Computing Institute at KTH, and he is a member of the Swedish Society of Mathematicians.

Michael Schliephake

Michael Schliephake started working at PDC in January. As a system administrator, he will concentrate on the operation of PDC's HPC computer systems. After spending several years as a software developer creating engineering software, he began working in the area of HPC at the High Performance Computing Center Stuttgart. There, Michael gained experience in the installation and operation of cluster systems as well as in several projects which were associated with the design and implementation of supercomputing and grid infrastructures. These projects were the German D-Grid Initiative and the European projects DEISA and PRACE. Furthermore, he brings with him his past experience educating students in HPC. Michael obtained his qualified engineer degree at the Technical University Leipzig.

Olav Vahtras

received a MS degree in engineering physics in 1988 and PhD in quantum chemistry in 1993, both at Uppsala University. His work at the Minnesota Supercomputer Institute 1993-1994 as a post-doc, and at Linköping University as a research associate 1994-1999 has included research on molecular properties and electronic structure theory. He now works at PDC as an application specialist in the field of computational chemistry.

Anders Ynnerman

Professor Anders Ynnerman received a Ph.D. in physics from Gothenburg University. During the early 90s he was doing research at Oxford University, UK, and Vanderbilt University, USA. In 1996 he started the Swedish National Graduate School in Scientific Computing, which he directed until 1999. From 1997 to 2002 he directed the Swedish National Supercomputer Centre and from 2002 to 2006 he directed the Swedish National Infrastructure for Computing (SNIC). Ynnerman is representing Sweden and the Nordic region in several international collaborations and policy bodies.

Since 1999 he is holding a chair in scientific visualization at Linköping University and In 2000 he founded the Norrköping Visualization and Interaction Studio (NVIS). NVIS currently constitutes one of the main focal points for research and education in computer graphics and visualization in the Nordic region. Ynnerman’s current research interest lies in the area of visualization of large scale and complex data sets with a focus on volume rendering and multi-modal interaction.

Anders Ynnerman’s Specialties: Computer Graphics, Visualization, Multimodal Interaction, Numerical Simulation, High Performance Computing.