Universität Wien FIND
Achtung! Das Lehrangebot ist noch nicht vollständig und wird bis Semesterbeginn laufend ergänzt.

050115 SE Seminar Parallel Computing (2009S)

Prüfungsimmanente Lehrveranstaltung

Vorbesprechung: Mittwoch 4.3., 11:00, C316, Nordbergstrasse 15/C3

Details

max. 15 Teilnehmer*innen
Sprache: Deutsch

Lehrende

Termine

Zur Zeit sind keine Termine bekannt.

Information

Ziele, Inhalte und Methode der Lehrveranstaltung

High performance computing has become the third pillar of science and
technology, providing the superior computational capability required for
dramatic advances in fields such as DNA analysis, drug design and astrophysics.
However, in today¿s dominating programming paradigm users wanting to fully
exploit the capabilities of parallel machines are forced to adopt a low-level
programming style that leads to high-cost software production and error-prone
programs that are difficult to write, reuse, and maintain. Emerging petascale
architectures built with many-core components providing millions of threads, and
applications of growing size and complexity will further aggravate this problem.

This seminar will discuss the state-of-the-art in programming paradigms for high
performance computing and identify the challenges posed by petascale
architectures characterized by massive parallelism across multiple levels of
computational hierarchies, deep memory structures, and sophisticated
interconnection networks. We will outline the requirements for high-productivity
programming languages that provide a viable compromise between the dual goals of
human productivity and target code efficiency, followed by an overview of the
major ideas underlying a number of key developments in this field.

Specifically, the topics discussed in the seminar will include:
- an overview of the major trends in language design and implementation since the
emergence of distributed-memory architectures in the 1990s,
- the High Performance Fortran (HPF) language family [6, 5]
- the high productivity languages Chapel [1, 7] and X10 [2] developed in the
DARPA-funded High-Productivity-Computing-Systems (HPCS) program in the USA
- recent developments focused on multi-core and many-core systems [3, 4, 8]

Art der Leistungskontrolle und erlaubte Hilfsmittel

Presentation and term paper.

Mindestanforderungen und Beurteilungsmaßstab

The goal of the seminar is to create an understanding of the opportunities and challenges related to some important trends in high-level parallel programming for future machines.

Prüfungsstoff

This seminar will discuss the state-of-the-art in programming paradigms for high
performance computing and identify the challenges posed by petascale
architectures characterized by massive parallelism across multiple levels of
computational hierarchies, deep memory structures, and sophisticated
interconnection networks. We will outline the requirements for high-productivity
programming languages that provide a viable compromise between the dual goals of
human productivity and target code efficiency, followed by an overview of the
major ideas underlying a number of key developments in this field.

Literatur

[1] Bradford L. Chamberlain, David Callahan, and Hans P. Zima. Parallel
Programmability and the Chapel Language. International Journal of HPC
Applications, 21(3):291¿312, 2007. Special Issue on High Productivity Languages
and Models.

[2] Philippe Charles, Christopher Donawa, Kemal Ebcioglu, Christian Grothoff,
Allan Kielstra, Christoph von Praun, Vijay Saraswat, and Vivek Sarkar. X10: An
Object-Oriented Approach to Non-Uniform Cluster Computing. In Proceedings of the
2005 Conference on Object-Oriented Programming Systems, Languages and
Applications (OOPSLA 05), pages 519-538, October 2005.

[3] Khronos OpenCL Working Group. The OpenCL Specification. Version 1.0,
Document Revision: 33.

[4] C. R. Johns and D. A. Brokenshire. Introduction to the Cell Broadband Engine
Architecture. IBM J. Res. & Dev., 51(5):503-519, 2007.

[5] Ken Kennedy, Chuck Koelbel, and Hans P. Zima. The Rise and Fall of High
Performance Fortran: An Historical Object Lesson. In HOPL III: Proceedings of
the Third ACM SIGPLAN Conference on History of Programming Languages Conference
(HOPL-III), San Diego, California, June 2007.

[6] Piyush Mehrotra, John Van Rosendale, and Hans P. Zima. High Performance
Fortran: History, status and future. Parallel Computing, 24(3):325-354, May
1998.

[7] R.L.Diaconescu and H.P.Zima. An approach to data distributions in chapel.
The International Journal of High Performance Computing Applications,
21(3):313-335, 2007. Special Issue on High Productivity Programming Languages
and Models.

[8] Samuel Williams, John Shalf, Leonid Oliker, Shoaib Kamil, Parry Husbands,
and Katherine Yelick. The Potential of the Cell Processor for Scientific
Computing. In Proceedings of CF¿06, May 2006.

Zuordnung im Vorlesungsverzeichnis

Letzte Änderung: Fr 31.08.2018 08:48