A study on GPGPU acceleration of simultaneous processing heterogeneous tasks with mutual dependence relation

点击翻译按钮获取中文摘要

A general biophysical model here is a set of mathematical expressions (functions) and ordinary differential equations with variables that represent ion concentration in cells, electrical current of neurons and so on. The biophysical simulation calculates time variation of these variables. Flint is a simulator that numerically integrates heterogeneous biophysical models.We propose two acceleration methods for Flint using a graphics processing unit (GPU). The first method interprets multiple bytecodes in parallel on the GPU. It automatically parallelizes the simulation using a level scheduling algorithm. The second method translates a model into a source code through the internal bytecode, which speeds up the compilation of the generated source codes, because the code size is diminished because of bytecode unification. The latter achieves a performance of 2.4 times higher than that of the former.