stOMP: A Specializing Threads Library for OpenMP Runtime Compilation and Optimizations for OpenMP

While runtime compilation has in practice been largely restricted to programming languages that execute on virtual machines, such as Java and C#, parallel OpenMP programs show many promising traits for efficient and effective runtime optimization. This work introduces stOMP: a specializing thread-library for OpenMP. Using a combined compile-time and run-time system, stOMP specializes OpenMP parallel regions for frequently-seen values and the configuration of the runtime system. We present a detailed description of the system, focusing on the optimizations implemented and techniques to minimize the runtime overhead: a context-based hot-spot detector; a pruning mechanism that eliminates poorly behaved variables as specialization targets; several runtime optimization policies; and several code optimizations and transformations that further allow performance improvement. We evaluate our work on the SPEC OMP benchmark suite, showing a performance increase of up to 7.8%.