Introduction to Parallel and Distributed Computing (SS 2018)

Section outline

• Select section Introduction to Parallel and Distributed Computing (SS 2018)

  Introduction to Parallel and Distributed Computing (SS 2018)

  Collapse all Expand all

  326.081/326.0AD, Monday 8:30-10:00, S2 219, Start: March 5, 2018
  

  The efficient application of parallel and distributed systems (multi-processors and computer networks) is nowadays an important task for computer scientists and mathematicians. The goal of this course is to provide an integrated view of the various facets of software development on such systems: hardware architectures, programming languages and models, software development tools, software engineering concepts and design patterns, performance modeling and analysis, experimenting and measuring.

  Class presentation will be accompanied by hands-on experience on an SGI Altix UltraViolet 1000 distributed shared memory multiprocessor using the following programming models:

  • Auto-parallelization and OpenMP and Cilk Plus (Intel C/C++),
  • Java multi-threading and networking API,
  • Message Programming Interface (MPI-C/C++).

  Some background in programming in C/C++ and/or Java is assumed. Course grades will be based on a couple of programming exercises.
  

  To take part in the course, you have to enrol in the KUSSS system. Since the exercises will be submitted via Moodle, you also have to login in Moodle and register as a course participant. You will then also receive per email all messages posted in the News forum.
• Select section Introduction and Organization

  Introduction and Organization

  • Select activity Introduction to Parallel and Distributed Computing

    

    Introduction to Parallel and Distributed Computing File
  • Select activity RISC Course Software

    

    RISC Course Software URL
  • Select activity How to Use the Parallel Computing Server

    

    How to Use the Parallel Computing Server Page
  • Select activity How to Use the Server From Other Networks (restricted)

    

    How to Use the Server From Other Networks (restricted) URL
  • Select activity Some More Information (in German)

    

    Some More Information (in German) Page
  • Select activity Restricted Area (password handed out in class)

    

    Restricted Area (password handed out in class) URL
• Select section Parallel Architectures

  Parallel Architectures

  • Select activity Parallel Architectures

    

    Parallel Architectures File
  • Select activity Technical Advances in the SGI UV Architecture

    

    Technical Advances in the SGI UV Architecture File
  • Select activity Performance and Productivity Breakthroughs with Very Large Coherent Shared Memory: The SGI ® UV Architecture

    

    Performance and Productivity Breakthroughs with Very Large Coherent Shared Memory: The SGI ® UV Architecture File
  • Select activity SGI UV Datasheet

    

    SGI UV Datasheet File
  • Select activity JKU Datasheet

    

    JKU Datasheet File
  • Select activity The Supercomputer "MACH-2"

    

    The Supercomputer "MACH-2" URL
• Select section Auto-Parallelization and OpenMP and Cilk Plus

  Auto-Parallelization and OpenMP and Cilk Plus

  • Select activity Automatic Parallelism

    

    Automatic Parallelism File
  • Select activity Example

    

    Example File
  • Select activity User and Reference Guide for the Intel C++ Compiler

    

    User and Reference Guide for the Intel C++ Compiler URL
  • Select activity Automatic Parallelization

    

    Automatic Parallelization URL
  • Select activity Guided Auto-Parallelism

    

    Guided Auto-Parallelism URL
  • Select activity OpenMP

    

    OpenMP File
  • Select activity Sources

    

    Sources File
  • Select activity Example

    

    Example File
  • Select activity OpenMP

    

    OpenMP URL
  • Select activity OpenMP Specification

    

    OpenMP Specification URL
  • Select activity OpenMP Tutorial

    

    OpenMP Tutorial URL
  • Select activity Intel OpenMP* Support

    

    Intel OpenMP* Support URL
  • Select activity OpenMP* Pragmas and Clauses Summary

    

    OpenMP* Pragmas and Clauses Summary URL
  • Select activity Work Sharing Using OpenMP*

    

    Work Sharing Using OpenMP* URL
  • Select activity Thread Affinity Interface

    

    Thread Affinity Interface URL
  • Select activity GNU libgomp

    

    GNU libgomp URL
  • Select activity Enabling OpenMP

    

    Enabling OpenMP URL
  • Select activity GOMP_CPU_AFFINITY

    

    GOMP_CPU_AFFINITY URL
  • Select activity Cilk Plus

    

    Cilk Plus URL
  • Select activity Example

    

    Example File
  • Select activity The Cilk Project

    

    The Cilk Project URL
  • Select activity Cilk Plus Tutorial

    

    Cilk Plus Tutorial URL
  • Select activity Intel Cilk Plus

    

    Intel Cilk Plus URL
  • Select activity Set Worker Count

    

    Set Worker Count URL
  • Select activity GCC Cilk Plus

    

    GCC Cilk Plus URL
• Select section Performance Analysis

  Performance Analysis

  Modeling and analyzing the performance of parallel programs, some material from Chapter 3 of Foster's Designing and Building Parallel Programs.
  

  • Select activity Performance Analysis

    

    Performance Analysis File
• Select section Multi-Threaded Client/Server Programming

  Multi-Threaded Client/Server Programming

  • Select activity Concurrency in Java

    

    Concurrency in Java File
  • Select activity Sources

    

    Sources File
  • Select activity Lesson Concurrency

    

    Lesson Concurrency URL
  • Select activity Trail: Custom Networking

    

    Trail: Custom Networking URL
  • Select activity Posix Threads Programming

    

    Posix Threads Programming URL
  • Select activity C Sockets

    

    C Sockets URL
  • Select activity Client Example

    

    Client Example URL
  • Select activity Server Example

    

    Server Example URL
• Select section Parallel Program Design

  Parallel Program Design

  The engineering of parallel programs based on Chapter 2 of Foster's Designing and Building Parallel Programs.
  

  • Select activity Parallel Program Design

    

    Parallel Program Design File
• Select section The Message Passing Interface MPI

  The Message Passing Interface MPI

  Programming in MPI, partially based on Chapter 8 of Foster's Designing and Building Parallel Programs.
  

  • Select activity MPI

    

    MPI File
  • Select activity Sources

    

    Sources File
  • Select activity MPI Forum

    

    MPI Forum URL
  • Select activity MPI Specification

    

    MPI Specification URL
  • Select activity MPI Tutorial

    

    MPI Tutorial URL
• Select section Distributed Memory Algorithms

  Distributed Memory Algorithms

  • Select activity Presentation

    

    Presentation File
  • Select activity Figures (restricted)

    

    Figures (restricted) URL
• Select section Exercises

  Exercises

  Four programming/benchmarking exercises will be handed out . The best three results will be used for grading.