Return to Article Details An explicit parallelism study based on thread-level speculation Download Download PDF