Parallel Programming Languages Spring 2011

Homework 1

• Download and install an X10 implementation; choose between the Java-based 1.5 or the Scala-based 2.0.
• Modify this MapReduce implementation such that reduce runs in logaritmic time in the number of processors.
• Implement Parallel Prefix Computation in X10; it is specified by the Haskell function scanl1 on the first page of the linked paper. Make your implementation as parallel and efficient as you can.

Homework 2

• Implement the algorithm for May-Happen-in-Parallel Analysis of X10 in Featherweight X10: a core calculus for async-finish parallelism, Jonathan K. Lee and Jens Palsberg, in Proceedings of PPOPP'10, 15th ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, Bangalore, India, January 2010; or take the constraints from Complexity Results for May-Happen-in-Parallel Analysis, Jonathan K. Lee, Jens Palsberg, and Rupak Majumdar. Manuscript, 2010.
• Grammar (source) for a subset of X10 called miniX10.
• Benchmarks written in miniX10: MapReduce.x10 (prints 9045050) and Series.x10 (prints nothing)
• Here is Main.java for working with jtb and javacc.
• Here is my script runall.sh for running x10 on both MapReduce.x10 and Series.x10.
• Here is my script testall.sh for running Main.java on MapReduce.x10 and Series.x10.
• Here is the X10 1.5 release page.