Basic Information

News

Course overview

This course exposes students to the principal issues involved in software development for parallel computing and discusses a number of approaches to handle the problems and opportunities caused by the increased availability of parallel platforms.

The course includes lectures, assignments, self-study, and a project. 50% of your grade is determined by project work and 50% is determined by a written exam; the exam is given during the official examination period, and there is no makeup exam. Students must be able to program using Java and C/C++.

The course may cover: memory coherence and consistency models, implications for language-specific memory models, Java memory model, models of parallel programming and parallel program execution, performance models for parallel systems, transactional memory, compiler extraction of parallelism, language and compiler support for parallel programming, threads and their execution environment, synchronization, and implementation issues of these topics.

Course schedule

Lectures are given Mondays 13:15 - 16:00 in CAB G 61.

Recitation sessions take place Thursdays 13:15 -- 15:00 in LEE D 101 and take place when announced. Some of the recitation session hours will be devoted to other activities (tutorials, reviews, etc) or will be devoted to group meetings. Please watch this page for updates and announcements.

Tentative schedule of lectures

Week Monday Thursday
1 09/17: no lecture 09/20: no recitation
2 09/24: Organization - Introduction 09/27: MPI Tutorial - Code
3 10/01: Caches and Cache Coherence (1pp) (6pp) (unrolled animations) 10/04: Caches recap - Euler cluster - Euler sample job script
4 10/08: Memory models (1pp) (6pp) (unrolled animations) 10/11: Teams and initial projects presentations - OpenMP intro
5 10/15: Languages, Fast Locks, and Lock-Free (1pp) (6pp) (unrolled animations) 10/18: Memory Models - MPI Part 2
6 10/22: Fast Locks and Lock-Free (1pp) (6pp) (unrolled animations) 10/25: SPIN Tutorial - Continuing MPI Part 2
7 10/29: Intermediate Project Presentations 11/01: No recitation session
8 11/05: Amdahl's Law, PRAM, Alpha-Beta Model, Little's Law, Operational Intensity, Roofline Model I 11/08: Amdahl's Law, PRAM
9 11/12: Notes - Roofline Model II - Balance Principles - Scheduling 11/15: Roofline Model - Balance Principles
10 11/19: SIMD Vector Extensions 11/22: Balance Principles, SIMD - Vandermonde matrix determinant vectorized
11 11/26: Finishing consensus, scalable lock study, and oblivious algorithms (1pp) (6pp) (unrolled animations) 11/29: Work-Depth Model - MS thesis proposal
12 12/03: I/O complexity, Red-Blue Pebble Game, and Recomputation in Neural Networks 12/06: Red-Blue Pebble Game
13 12/10: Oblivious and Non-Oblivious Algorithms (1pp) (6pp) 12/13: Prefix-Sum, Network Models
14 12/17: Final Presentations

Mailinglist

If you want to discuss issues related to the class, please subscribe to the class mailinglist. This list is read by the instructors as well as the TA, and any student of the class who voluntarily subscribed to it. So it is a great place to discuss questions about homework or the lecture slides.

Assignments

Assignments are an important part of the course. You will not learn this material from listening to a lecture alone -- you have to do the assignments.

Note: Do not hesitate to write an email to your TA if you have trouble with the assignments!

Number Assignment Description Solution
1 Parallel PI See last slide of the first recitation session. PI-seq: download download
2 Caches and Cache Coherence Assignment 2 Solution - False sharing benchmark
3 Sequential Consistency, Locks Assignment 3 Solution
4 Locks and Measuring Cache Misses Assignment 4 Solution - Code
5 Roofline model, Balance Principles Assignment 5 Solution
6 SIMD Assignment 5 - Code template Solution

Teams

Number Members Project Description Final Presentation
1 Jack Clark, Genming Bai, Liaowang Huang, Zhifei Yang Collective Communications for Distributed Machine Learning
2 Lea Fritschi, Roman Haag, Michael Bernasconi, Giovanni Balduzzi Connected Components
3 Fabian Wolff, David Rohr, Pascal Oberholzer, Elias Stalder Single-Source Shortest Paths
5 Daniel Maag, Julian Dunskus, Silvan Läubli, Cédric Neukom Distributed MST
6 Andrey Oblupin, Gabriela Evrova, Clara Brimnes Gardner, Erik Träff Solving the Single Source Shortest Path (SSSP) problem on distributed graphs
7 Samuel Petitjean, Momchil Peychev, Daniel Zvara, David Ittah Parallel Strongly Connected Components
8 Anastasios Papageorgiou, Diego Renner, Ramy Tanios, Konstantinos Triantafyllou Parallel N-Body Simulations
9 Tommaso Bonato, Francesco Forcher, Anouk Paradis, Stefano d'Apolito Parallel Suffix Tree Construction
10 Thomas Cambier, Raphaël Dang-Nhu, Thibault Dardinier, Clément Trassoudaine Minimum Spanning Tree: A Parallel Approach
12 Kevin Ni, Fredrik Strupe, Samuele Decarli, Ankur Magdum Concurrent Skip List
Teams are listed in the order they have been announced.

Template for project report

Grading

50% of your grade is determined by the project, and the other 50% of the grade is determined by a written 2 hr exam. You are not allowed to use any electronic devices or books, notes, etc. to the exam.

Resources

There exist a large number of books on programming multi-processors, multi-core system, or threads. These books may explore some topics in more depth than the lectures, or may provide background information. None of these books is mandatory. Copies of the first two books are "on reserve" the CS library.