Implementation Issues of Loop { level SpeculativeRun { time

Current parallelizing compilers cannot identify a signiicant fraction of parallelizable loops because they have complex or statically insuuciently deened access patterns. We advocate a novel framework for the identiication of parallel loops. It speculatively executes a loop as a doall and applies a fully parallel data dependence test to check for any unsatissed data dependencies; if the test fails, then the loop is re{executed serially. We will present the principles of the design and implementation of a compiler that employs both run-time and static techniques to parallelize dynamic applications. Run-time optimizations always represent a tradeoo between a speculated potential beneet and a certain (sure) overhead that must be paid. We will introduce techniques that take advantage of classic compiler methods to reduce the cost of run-time optimization thus tilting the outcome of speculation in favor of signiicant performance gains. Experimental results from the PERFECT, SPEC and NCSA Benchmark suites show that these techniques yield speedups not obtainable by any other known method. To achieve a high level of performance for a particular program on today's super-computers, software developers are often forced to tediously hand{code optimizations tailored to a speciic machine. Such hand{coding is diicult, increases the possibility of error over sequential programming, and the resulting code may not be portable to other machines. Restructuring, or parallelizing, compilers address these problems by detecting and exploiting parallelism in sequential programs written in conventional languages. Although compiler techniques for the automatic detection of parallelism have been studied extensively over the last two decades, current parallelizing compilers cannot extract a signiicant fraction of the available parallelism in a loop if it has a complex and/or statically insuu-ciently deened access pattern. Typical examples are complex simulations such