Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out

university_of_vienna

VIEWS: 16 PAGES: 6

									                                 MY SCIENCE
                                 January 30, 2009
                                 Bolzano, Italy


Training Programme at the University of Vienna

Sabri Pllana
Institute of Scientific Computing (ISC)

 Established in 1989                   Head of Institute
    Prof. Hans Zima                        Prof. Siegfried Benkner




                http://www.par.univie.ac.at/

                                                                     2
Organizational Aspects of ISC
 Part of Faculty of Computer Science
    Focus: software and computing
   technologies in Computational Science


 Two groups
    Software Science Group (A)
   (Nordbergstraße 15, 1090 Vienna)
   Data Analysis Group (B)
   (Universitätsstraße 5, 1010 Vienna)

 50 members
    15 faculty
    25 research
    10 admin/support


                                           3
The Research Focus
 The name of Institute changed several times
    Institute for Software Technology and Parallel Systems
    Institute of Software Science
    Institute of Scientific Computing (since 2005)


 The research focus remained unchanged: Scientific Computing
    programming paradigms
    languages and compilers
    software infrastructures


 Application driven research
    material sciences
    financial optimization
    biomedical simulations
                                                     http:// www.wien2k.at
                                                                             4
Timeline

                 Zima et al.             Vienna                   Opus language,           Processor-in-         Workflow                • GridMiner, data      Amadeus,
                 Supercompilers for      HPF Compiler,            high-level task-         Memory (PiM)          specification,          mining on the Grid;    QoS-aware
                 Parallel and Vector     targets regular          parallel interface,      Model,                AGWL (XML),             • Performance          workflow
                 Computers,              applications             heterogeneous            collaboration         UML DSL                 Prophet, performance   environment
                 ACM Press book                                   systems                  with Caltech/JPL                              modeling& prediction


1989                   1992                    1996                      1998                    2002                 2004                     2006                   2008

          1990                     1994                    1997                         2000                  2003                 2005                   2007
                             Vienna Fortran,                                                          Extended VFC
                             data parallel                                                            for hybrid-            • VGE, SO Grid env.,
                             language, with NASA,                                                     parallel               appl.-level QoS;       Autonomic
                             major input for High    HPF+,                                                                   • Chapel, high         computing for           High-level
                                                                                                      programming
     Year of                 Performance Fortran     targets irregular         VFC, the Vienna                               productivity           QoS-aware Grid          multi-core
                                                                                                      on SMP
     inception               (HPF)                   applications              HPF+ Compiler                                 language, with Cray    services                programming
                                                                                                      clusters




    Intel iPSC/860,          Gescher Cluster (80 processors),                       Grid Computing Systems                     Luna Multi-core Cluster,                 STI Cell BE,
 16 Intel 80860 (i860),      a subcluster of single CPU nodes,                                                                Dual-Core AMD Opterons,                IBM QS22 (2xCell)
 RISC chips (40 MHz)         a subcluster of 4xCPU SMP nodes                                                                    (288 processor cores)


                                       Hardware Infrastructure: http://www.par.univie.ac.at/index.php?goto=hardware                                                           5
Thank you for your attention

								
To top