A Case Study: Exploiting Neural Machine Translation to Translate CUDA to OpenCL
This work addresses program translation for GPU programming, but it is incremental as it applies an existing method to a new domain.
The authors tackled the problem of translating CUDA programs to OpenCL programs using a neural machine translation model, achieving results demonstrated through a case study on benchmarks like Polybench-gpu-1.0, NVIDIA SDK, and Rodinia.
The sequence-to-sequence (seq2seq) model for neural machine translation has significantly improved the accuracy of language translation. There have been new efforts to use this seq2seq model for program language translation or program comparisons. In this work, we present the detailed steps of using a seq2seq model to translate CUDA programs to OpenCL programs, which both have very similar programming styles. Our work shows (i) a training input set generation method, (ii) pre/post processing, and (iii) a case study using Polybench-gpu-1.0, NVIDIA SDK, and Rodinia benchmarks.