High Performance Computing on GPU

The slides are available only in the network of the University.


On 28.10.2010 will be lecture in G40.


On 04.11.2010 we will visit the TESLA Hybrid Cluster Bienroderweg 95.


On 10.11.2010 the lecture is cancelled.

Lecturer Dr. Josef Schüe


Schedule Lecture: Wed 09:45-11:15 Uhr in room IZ161



Exercises: Thu 13:15-14:45 Uhr in room G40


Start 27.10.2010


Prerequisites Interest, knowledge in C programming


Target group Students of computer science, mathematics, natural or engineering sciences or guest students in their advanced study period, CSE students


Certificates Homework assignments and active participation in the exercises, test


Office hours Will be announced in the lecture


Assistent Manuel Rodriguez m.rodriguez@tu-bs.de

• Paralleles Rechnen: Performancebetrachtungen zu Gleichungslösern; Josef Schüle, Oldenbourg 2010
• CUDA by Example: An Introduction to General-Purpose GPU Programming; Jason Sanders, Edward Kandrot; Addison Wesley 2010
• Programming Massively Parallel Processors: A Hands-On Approach; David Kirk, Wen-Mei W. Hwu; Morgan Kaufman Publ Inc. 2010

• 00 Introduction
• 01 Preface
• 02 Hardware
• 03 Programming Model
• 04 Scalability
• 05 Memory Management
• 06 Memory Access
• 07 Texture Memory
• 08 Atomic Operations
• 09 Wrap Up
• 10 Interopability
• 11 Optimierung

• Introduction to the Framework
• [Exercise 01] [Framework] [Solution Slides] [Solution Framework]
• [Exercise 02] [Framework] [Solution Slides] [Solution Framework]
• [Exercise 03] [Framework] [Solution Slides] [Solution Framework]
• [Exercise 04] [Framework] [Solution Slides] [Solution Framework]
• [Exercise 05] [Framework] [Solution Slides] [Solution Framework]
• [Exercise 06] [Framework] [Solution Slides] [Solution Framework]
• [Exercise 07] [Framework] [Solution Slides] [Solution Framework]
• [Exercise 08] [Framework] [Solution Slides] [Solution Framework]
• [Exercise 09] [Framework] [Solution Slides] [Solution Framework]
• [Exercise 10] [Solution]

Copyright by Stefan Tobias