Wim Bohm
Office: 470 CS Building
Office Hours: MW 10:00-11:00
Louis Rabiet
Office: CSB 120
Office Hours: Tue 09:00 - 11:00 am
Lecture Time and Place:
9:00-9:50, MWF, CSB Room 130

Parallel programming is becoming increasingly critical for a computer scientist/engineer. All new machines are now parallel. There is already a movement that suggests that all students be taught parallel programming in the introductory courses, with sequential programming viewed as a special case. However, parallel programming is not easy. For current CPUs it requires awareness of the multicore architecture, the cache hierarchy, and in some cases the vector floating point capabilitites .
This course will teach you the basic concepts of designing, writing, debugging, and analyzing parallel programs. It covers the two main paradigms: shared memory, and message passing. We will also examine a number of parallel algorithms for a range of problems.

Wednesday Dec 10, class cancelled.
As the 1060s are out of order, the CUDA1 assignment should be run on the smaller machines, and can be submitted until the late date without late penalty. See the editted PA5 description.
Final exam.
CS475 on campus exam: Monday Dec 15 9:40 am - 11:40 am
On line: One 2 hour online test with all the questions, which can be taken from Monday Dec 15, 10:00 am until Wednesday Dec 17, 23:59.
Final test material: from week 8 (Matrix Multiply) to week 15 (CUDA coalescing).

Please grade your TA here .
As many students have not submitted PA4, we have moved the late date for PA4 to Wednesday Nov 19.
Jan 17: here are the echo lectures for this course.