COURSE INFORMATION (First Handout)
Designing and Building
Parallel Programs (by Ian Foster): an on-line book
TENTATIVE SCHEDULE
| Week |
|
| WEEK2, Jan 22 |
Introductions, VonNeumann architecture, examples of hard problems, growing performance, Moore's law saturation and need for alternatives (Ch1) |
| WEEK3, Jan 29 |
Parallelism,
speed-up, perfect linear
speed-up, embar.par.prog., obstacles (Ch1): data dep., comm. overhead (*-grained algo's), seq. overhead (Amdahl's law) |
| WEEK4, Feb 5 |
Parallel architectures (CPU's vs. RAM),
shared memory (Ch8), SMP's, threads, Java threads, Example1.java, Example2.java, PRJ1-A, Primes.java |
| WEEK5, Feb 12 |
SMP's (Ch8), threads (8.2.2), Java threads, project1 discussion |
| WEEK6, Feb 19 |
PRJ1-D, Pthreads (C/C++) (8.2.2, App B), PRJ2-A, primes.cc |
| WEEK7, Feb 26 |
Pthreads, critical sections, thread
synchronization (8.3, App B), mutual exclusion, locks, semaphores, deadlocks |
| WEEK8, Mar 5 |
PRJ2-D MIDTERM Distributed memory, clusters, MPI (Ch 2, App A), Galaxy cluster PRJ3-A (prj3.cc) |
| WEEK9, Mar 12 |
SPRING
BREAK |
| WEEK10, Mar 19 |
Midterm
& Project 3 discussion, MPI (Ch 2, App A), Galaxy cluster, parallel lab tour |
| WEEK11, Mar 26 |
PRJ3-D,
Project 3 discussion, PRJ4-A, PlainSpint.tar, Project 4 discussion; MPI (Ch 2, App A), Galaxy cluster, parallel lab tour |
| WEEK12, Apr 2 |
Project 4 discussion |
| WEEK13, Apr 9 |
Project 4 discussion, Beowulf/Supercomputers/Top500, interconnects, switched Ethernet (switch vs. hub vs. router) |
| WEEK14, Apr 16 |
Conf - NO CLASS! |
| WEEK15, Apr 23 |
Some other non-emb.-parallel algo's: parallel prefix, div&conq, sorting networks & merge-sort, matrix mult-n, etc.; Project 4 discussion |
| WEEK16, Apr 30 |
PRJ4-D,
Project 4 Presentations,
Final review |
| FINAL
Mo, May 7, 5:30pm, RI-106 |
|
A = Assigned
D = Due