CS 6204: Ubiquitous Parallelism 
Spring 2015 
Dongyoon Lee

With the advent of parallel systems, it is now the role of programmers to
explicitly expose parallelism in software and take advantage of parallel
computing resources. However, parallel programming is inherently complex
as programmers have to reason about all possible interleavings. The
course will include recent advances and new directions that are being
pursued to address the challenges in developing easy-to-write, efficient
and reliable parallel programs. We will discuss solutions that cut across
the computing stack from languages to compilers to operating systems to
processor design, including topics such as active testing, deterministic
parallel computing, GPGPU, MapReduce, OS support for multi-core, memory
consistency models, etc.