Combining Compile-Time and Run-Time Parallelization

Final Report: Combining Interprocedural Compile-Time and Run-Time Parallelization

This report describes the final results of a project revolving around an integrated automatic parallelization system that combines high quality interprocedural compile-time analysis with flexible run-time support. This research builds on our previous extensive work on interprocedural compiletime analysis for parallelization, sponsored by DARPA through contracts at Rice University and Stanford U...

A Case for Combining Compile-Time and Run-Time Parallelization

Run Time Compile Time Compile − time Specialization Run − time

Because of its wide applicability, many eecient implementations of the Fast Fourier Transform have been developed. We propose that an eecient implementation can be produced automatically and reliably by partial evaluation. Partial evaluation of an unoptimized implementation produces a speedup of over 9 times. The automatically generated result of partial evaluation has performance comparable to...

Combining Compile Time and Run Time Support for E cient Software Distributed Shared Memory

We describe an integrated compile time and run time system for e cient shared memory parallel computing on distributed memory machines The combined system presents the user with a shared memory programming model with its well known bene ts in terms of ease of use The run time system implements a consistent shared mem ory abstraction using memory access detection and auto matic data caching The ...

Run-Time Parallelization: Its Time Has Come

Current parallelizing compilers cannot identify a significant fraction of parallelizable loops because they have complex or statically insufficiently defined access patterns. This type of loop mostly occurs in irregular, dynamic applications which represent more than 50% of all applications [20]. Making parallel computing succeed has therefore become conditioned by the ability of compilers to a...