Logo des Repositoriums

Auflistung nach Schlagwort "High Performance Computing"

1 - 4 von 4
  • Konferenzbeitrag
    Anpassung von Stencil-Codes zur Laufzeit mit Hilfe von Umschreiben auf Binärebene für dynamisch bestimmtes Speicherlayout
    (SKILL 2018 - Studierendenkonferenz Informatik, 2018) Pröll, Konrad M.
    Die Optimierung von Stencil-Codes ist eine zentrale Herausforderung im Bereich des Hochleistungsrechnen. Die meisten Ansätze fokussieren sich entweder darauf, diese zur möglichst effizient zur Berechnung auf groSSen Parallelrechenstrukturen zu parallelisieren oder die möglichst effiziente Ausnutzung des Caches, um die Leistung zu steigern. Viele Stencil-Codes nutzen einfache Arrays zur Speicherung der Matrix. Komplexere Datenstrukturen erhöhen die Rechenzeit des Stencil-Codes dadurch, dass die Berechnung der Speicheradresse einer Zelle deutlich komplizierter wird und sogar bedingte Sprünge enthält. In dieser Arbeit wird ein Ansatz vorgeschlagen, wie aus bereits kompilierten Stencil-Codes zur Laufzeit das Speicherlayout analysiert werden kann und das Programm durch partielle Evaluation optimiert wird. Im Gegensatz zu konventioneller partieller Evaluation wird hierbei nicht für konstante Argumente, sondern für Wertebereiche, in denen sich ein Argument befindet, spezialisiert. Durch diese Methode können die Leistungseinbußen merklich reduziert werden.
  • Konferenzbeitrag
    A Comparison of HPC based Quantum Computing Simulators using Quantum Volume
    (INFORMATIK 2024, 2024) Boehme, Christian; van Niekerk, Lourens; Kumar, Dhiraj; Sharma, Aasish Kumar; Meisel, Tino; Paleico, Martin Leandro
    This paper compares quantum computing simulators running on a single CPU or GPU-based HPC node using the Quantum Volume benchmark commonly proposed for comparing NISQ systems. As simulators do not suffer from noise, the metric used in the comparison is the time required to simulate a set Quantum Volume. The results are important to estimate the feasibility of proof of concept studies and debugging of quantum algorithms on HPC systems. Besides benchmarks of some commonly used simulators, this paper also offers an overview of their main features, a review of the state of quantum computing simulation and quantum computing benchmarking, and some insight into the theory of Quantum Volume.
  • Zeitschriftenartikel
    Fooling the masses with performance results: Old classics and some new ideas
    (PARS: Parallel-Algorithmen, -Rechnerstrukturen und -Systemsoftware: Vol. 29, No. 1, 2012) Wellein, G.; Hager, G.
    G. Wellein and G. Hager Department for Computer Science and Erlangen Regional Computing Center Friedrich-Alexander-Universität Erlangen-Nürnberg In 1991, David H. Bailey published his insightful 'Twelve Ways to Fool the Masses When Giving Performance Results on Parallel Computers.' In that humorous article, Bailey pinpointed typical 'evade and disguise' techniques for presenting mediocre performance results in the best possible light. At that time, the supercomputing landscape was governed by the 'chicken vs. oxen' debate: Could strong vector CPUs survive against the new massively parallel systems? In the past two decades, hybrid, hierarchical systems, multi-core processors, accelerator technology, and the dominating presence of commodity hardware have reshaped the landscape of High Performance Computing. It's also not so much oxen vs. chickens anymore; billions of ants have entered the battlefield. This talk gives an update of the 'Twelve Ways.' Old classics are presented alongside new 'stunts' that reflect today's technological boundary conditions. DISCLAIMER: Although these musings are certainly inspired by experience with many publications and talks in HPC, I wish to point out that (i) no offense is intended, (ii) I am not immune to the inherent temptations myself and (iii) this all still just meant to be fun.
  • Konferenzbeitrag
    Geographic Grid-Computing and HPC empowering Dynamical Visualisation for Geoscientific Information Systems
    (Grid service engineering and management – the 4th international conference on grid service engineering and management – GSEM 2007, 2007) Rückemann, Claus-Peter
    This paper gives an overview of the potential of the cur- rent implementation of portable components for Geoscientific Information Systems (GIS) within the GISIG actmap-project. The computing problems addressed are multifold and for the first time presented here: With Active Source having extended the framework for conventional GIS, new features have been enabled like the use of Grid Computing and clus- ter resources, dynamical visualisation, and High Performance Computing (HPC) in order to be used for Geographic Grid Computing. Base of scientific content can for example be geophysical information like environmental or seismological data, geographical and spatial information using Geographic Data Infrastructures (GDI), as well as data from industrial, economic, cultural, and social sources. An integrated solution for moni- toring, accounting, billing supporting the geo-information market can be incorporated into this context. An outlook is given for Geographic Grid Computing e.g. for the extended use of Web Services and GDI in the future.