DD2451 Parallel and Distributed Computing, pardis11
News
20 Dec: Course evaluation form is now available.
16 Dec: All handins are now graded. Answers and reference answers to all three handins are available.
Content
Advanced course in computer science on the theory, algorithms, and techniques of parallel and distributed computing systems. Focus is on principles and algorithms rather than practical systems construction. The companion course FDD3008 Distributed Algorithms is given for postgraduate students with some extra assignments.
Objectives
- To understand and account for models, limitations, concepts and algorithms in the area of message passing and shared memory concurrency, and apply this understanding to example systems and algorithms
- To analyze parallel and distributed algorithms for correctness, performance, reliability, and security.
Schedule
Lectures and student presentations:
Lecture 1 |
24 Oct |
10-12 |
D35 |
Introduction, mutual exclusion, locks |
Slides, 6 per page |
Lecture 2 |
27 Oct |
13-15 |
B23 |
Concurrent objects, linearization, sequential consistency |
Slides, 6 per page |
Lecture 3 |
31 Oct |
10-12 |
D42 |
Registers, snapshots |
Slides, 6 per page |
Lecture 4 |
2 Nov |
13-15 |
E34 |
Consensus, I |
Slides, 6 per page |
Exercise 1 |
7 Nov |
16-17 |
1537 |
Exercises lecture 1 and 2 |
|
Lecture 5 |
8 Nov |
13-15 |
E31 |
Spin locks |
Slides, 6 per page |
Lecture 6 |
10 Nov |
15-17 |
Q17 |
Concurrent data structures |
Slides, 6 per page |
Lecture 7 |
14 Nov |
13-15 |
D35 |
Consensus, II |
Slides, 6 per page |
Exercise 2 |
15 Nov |
17-18 |
1537 |
Exercises lecture 2 and 3 |
|
Lecture 8 |
16 Nov |
15-17 |
D42 |
Consensus, III |
- |
Presentations |
18 Nov |
10-12 |
V33 |
|
|
Lecture 9 |
21 Nov |
10-12 |
D35 |
Leader Election |
Slides, 6 per page |
Exercises 3 |
22 nov |
17-18 |
1537 |
Exercises lecture 4, 5, 6 |
|
Presentations |
23 Nov |
15-17 |
V23 |
|
|
Lecture 10 |
25 Nov |
10-12 |
E36 |
Fault tolerance |
Slides, 6 per page |
Lecture 11 |
28 Nov |
10-12 |
D35 |
Spanning trees, aggregation. Guest lecture Rolf Stadler, EES |
Slides |
Exercises 4 |
29 Nov |
17-18 |
1537 |
Exercises lecture 7, 8, 9 (?) |
|
Presentations |
1 Dec |
10-12 |
D34 |
|
|
Exercises 5 |
2 Dec? |
TBA |
TBA |
Exercises, CANCELLED |
|
Lecture 12 |
5 Dec |
10-12 |
D35 |
P2P systems, DHT's |
Slides, 6 per page |
Presentations |
6 Dec |
10-12 |
Q31 |
|
|
The subjects indicated above are subject to change as we move along
Student Presentations
Each student has the option to present a research paper one of the presentations sessions.
Course Material
Main textbook is
- Maurice Herlihy, Nir Shavit: The art of multiprocessor programming, Morgan Kaufmann 2008.
Other useful references - among many:
- Hagit Attiya, Jennifer Welch: Distributed Computing: Fundamentals, Simulations and Advanced Topics, McGraw-Hill Publishing, 2004
- David Peleg: Distributed Computing: A Locality-Sensitive Approach. Society for Industrial and Applied Mathematics (SIAM), 2000
- George Coulouris, Jean Dollimore, Tim Kindberg, Gordon Blair: Distributed Systems, Concepts and Design. Pearson 2012
- Christian Cachin, Rachid Guerraoui, Luis Rodrigues: Reliable and secure Distributed Programming, 2n ed., Springer 2011
Other material will be made availalbe here as the course moves on.
Examination and Grading
Probably examination will be by hand-ins. The details, including grading, will be announced at the first lecture. If we decide to include labs or other activities, they will count towards the final grade as well. Additional requirements will apply to FDD3008.
Prerequisites
The course does not have formal prerequisites. Target audience if D4, F4, and the MD line of SU. Some basic familiarity with algorithms, basic mathematical discourse in cs, and programming will be very useful.
Professor
Mads Dam, mfd@kth.se, 790 6229, room 4416. No specific office hours. Preferred contact by email.
Code of Honour
It goes without saying that the Nada code of honour applies.
|