Section: New Results

Automatic Generation of Adaptive Simulation Codes

Participants : Cédric Bastoul, Maxime Schmitt.

Compiler automatic optimization and parallelization techniques are well suited for some classes of simulation or signal processing applications, however they usually don't take into account neither domain-specific knowledge nor the possibility to change or to remove some computations to achieve “good enough” results. Quite differently, production simulation and signal processing codes have adaptive capabilities: they are designed to compute precise results only where it matters if the complete problem is not tractable or if the computation time must be short. In this research, we design a new way to provide adaptive capabilities to compute-intensive codes automatically, inspired by Adaptive Mesh Refinement a classical numerical analysis technique to achieve precise computation only in pertinent areas. It relies on domain-specific knowledge provided through special pragmas by the programmer in the input code and on polyhedral compilation techniques, to continuously regenerate at runtime a code that performs heavy computations only where it matters at every moment. A case study on a fluid simulation application shows that our strategy enables dramatic computation savings in the optimized portion of the application while maintaining good precision, with a minimal effort from the programmer.

This research direction started in 2015 and complements our other efforts on dynamic optimization. In 2016, we started a collaboration on this topic with Inria Nancy - Grand Est team TONUS, specialized on applied mathematics (contact: Philippe Helluy), to bring models and techniques from this field to compilers. This collaboration received the support from the excellence laboratory (LabEx) IRMIA through the funding of the thesis of Maxime Schmitt on this topic. Two papers on this new research direction has been accepted this year on this topic (IMPACT 2017 workshop, HiPC 2017 conference [20]).