Automatically Exploiting Cross-Invocation Parallelism Using Runtime Information

Automatic parallelization is a promising approach to producing scalable multi-threaded programs for multicore architectures. Many existing automatic techniques only parallelize iterations within a loop invocation and synchronize threads at the end of each loop invocation. When parallel code contains many loop invocations, synchronization can easily become a performance bottleneck. Some automatic techniques address this problem by exploiting cross-invocation parallelism. These techniques use static analysis to partition iterations among threads to avoid cross-thread dependences. However, this partitioning is not always achievable at compile-time, because program input determines dependence patterns at run-time. By contrast, this paper proposes DOMORE, the first automatic parallelization technique that uses runtime information to exploit additional cross-invocation parallelism. Instead of partitioning iterations statically, DOMORE dynamically detects cross-thread dependences and synchronizes only when necessary. DOMORE consists of a compiler and a runtime library. At compile time, DOMORE automatically parallelizes loops and inserts a custom runtime engine into programs. At runtime, the engine observes dependences and synchronizes iterations only when necessary. For six programs, DOMORE achieves a geomean loop speedup of 2.1x over parallel execution without cross-invocation parallelization and of 3.2x over sequential execution on eight cores.