Organisers: Lone Leth Thomsen, Josva Kleist and Bent Thomsen
Lecturers: Brian Vinter (DIKU) and Bent Thomsen (CS AAU)
Welcome to the homepage of the PhD Course
on Programming Super Computers,
All lectures are in room 0.2.12 in the
Maps and other useful information
PC room 0.2.18 has been reserved for the course participants
The schedule is as follows:
Day 1 Lectures by: Brian Vinter |
Day 2 Lectures by Brian Vinter |
Day 3 Lectures by Bent Thomsen |
9.00-9.45 Shared Memory Architectures and SMP programming |
9.00-9.45 MPP Architectures and MPI Programming |
9.00-9.45 Future of (Super) Computer Programming 1 |
9.45-10.00 Break |
9.45-10.00 Break |
9.45-10.00 Break |
10.00-10.45 Shared Memory Architectures and SMP programming |
10.00-10.45 MPP Architectures and MPI Programming |
10.00-10.45 Future of (super) Computer Programming 2 |
10.45-11.00 Break |
10.45-11.00 Break |
10.45-11.00 Break |
11.00-12.00 Introduction to Fyrkat by Josva Kleist |
11.00-12.00 Exercises/Reading |
11.00-12.00 Future of (Super) Computer Programming 3 |
12.00-13.00 Break |
12.00-13.00 Break |
12.00-13.00 Break |
13.00-13.45 NOW architectures and PVM Programming |
13.00-13.45 Remote Memory and Programming with Tuple-spaces and Global Arrays |
|
13.45-14.00 Break |
13.45-14.00 Break |
|
14.00-15.00 Exercises/Reading |
14.00-15.00 Exercises/Reading |
|
Link to Brian Vinter’s slides on Cluster Computing
Day
1 Morning:
Shared Memory Architectures and SMP programming
Andrew D. Birrell, An Introduction to Programming with
Threads, SRC
Research Report 35.
Simultaneous Multithreading Project Academic project with much
information and no 'sales' material.
The Register Story on Intel SMP on a
chip
Getting Started with POSIX Threads
Day
1 Afternoon
NOW architectures and PVM Programming
PVM:Parallel Virtual Machine A Users' Guide and
Tutorial for Networked Parallel Computing
Adam J. Ferrari, JPVM: Network Parallel Computing in Java, ACM 1998 Workshop on Java for
High-Performance Network Computing
Day
2 Morning
MPP Architectures and MPI Programming
The TOP500
project was started in 1993 to provide a reliable basis for tracking and
detecting trends in high-performance computing. Twice a year, a list of the
sites operating the 500 most powerful computer systems is assembled and
released. The best performance on the Linpack benchmark is used as performance
measure for ranking the computer systems. The list contains a variety of
information including the system specifications and its major application
areas.
Appendix G Vector processor from John L. Hennessy and David A.
Patterson: "ComputerArchitecture - A Quantitative Approach", third
edition, Morgan Kaufmann, 2003.
The Earth
Simulator is built
from 640 units of high performance nodes which employ eight vector processors
and 16 Gbytes of high speed memory, interconnected via a high speed crossbar
network.
IBM Blue Gene is described
in IBM Journal of
Research and Development Volume 49, Number 2/3, 2005. The article "Overview of the
Blue Gene/L system architecture" can be recommended.
MPI home page The Message Passing Interface (MPI) standard, contains lots of useful material.
MPI: A Message-Passing Interface
Standard. The
reference manual for MPI. Don't print!
Writing Message Passing Parallel
Programs with MPI.
Course notes for a two day course on MPI usage, N. MacDonald, E. Minty, J.
Malard, T. Harding, S. Brown, M. Antonioletti, Edinburgh Parallel Computing
Centre.
Day
2 Afternoon
Remote Memory and Programming with Tuple-spaces and Global Arrays
Brian Vinter. Chapter 2 and 3 of: PastSet - A Structured Distributed Shared Memory
System. Dr. Scient,
Thesis,
IBM TSpaces Programmer's Guides.
Day
3 Morning
Future of (Super) Computer Programming
Link to Bent Thomsen’s slides on Super Computer Languages
Parallel Programmability and the Chapel Language
X10:An Object-Oriented Approach to Non-Uniform Cluster Computing
A Preview of What's New in C# 3.0
Additional
links and resources
DARPA HPCS - HIGH PRODUCTIVITY COMPUTER SYSTEMS
IBM Research | IBM Research | The X10 Programming Language
Chapel Programming Language Homepage
http://research.microsoft.com/fsharp
Don Syme's WebLog on F# and Other Research Projects : A Taste of F# Interactive in Visual Studio
The Scala Programming Language
Fortress is installed on Fyrkat, but due to some configuration problems you have to installed it in your local directly.
You can run the script ~kleist/bin/installFortress.sh which will do the job.
Scala is installed on Fyrkat in /pack/scala - remember to include java (/pack/jdk) in your PATH and set JAVA_HOME=/pack/jdk