The combination of expert-tuned code expression and aggressive compiler optimizations is known to deliver the best achievable performance for modern multicore processors. The development and maintenance of these optimized code expressions is never trivial. Tedious and error-prone processes greatly decrease the code developer's willingness to adopt manually-tuned optimizations. In this paper, we describe a pre-compilation framework that will take a code expression with a much higher programmability and transform it into an optimized expression that would be much more difficult to produce manually. The user-directed, source-to-source transformations we implement rely heavily on the knowledge of the domain expert. The transformed output, in its optimized format, together with the optimizations provided by an available compiler is intended to deliver exceptionally high performance. Three computational fluid dynamics (CFD) applications are chosen to exemplify our strategy. The performance results show an average of 7.3x speedup over straightforward compilation of the input code expression to our pre-compilation framework. Performance of 7.1 Gflops/s/core on the Intel Nehalem processor, 30% of the peak performance, is achieved using our strategy. The framework is seen to be successful for the CFD domain, and we expect that this approach can be extended to cover more scientific computation domains.
Lin, Pei-Hung; Jayaraj, Jagan; Woodward, Paul; Yew, Pen-Chung.
A Code Transformation Framework for Scientific Applications on Structured Grids.
Retrieved from the University of Minnesota Digital Conservancy,
Content distributed via the University of Minnesota's Digital Conservancy may be subject to additional license and use restrictions applied by the depositor.