Physis
Naoya Maruyama, Tatsuo Nomura, Kento Sato et al. · 2011
This paper proposes a compiler-based programming framework that automatically translates user-written structured grid code into scalable parallel implementation code for GPU-equipped clusters. To enable such automatic translations, we design a small set of declarative constructs that allow the user to express stencil computations in a portable and implicitly parallel manner. Our framework translates the user-written code into actual implementation code in CUDA for GPU acceleration and MPI for node-level parallelization with automatic optimizations such as computation and communication overlapping. We demonstrate the feasibility of such automatic translations by implementing several structured grid applications in our framework. Experimental results on the TSUBAME2.0 GPU-based supercomputer show that the performance is comparable as hand-written code and good strong and weak scalability up to 256 GPUs.
Cited by 172 works. doi:10.1145/2063384.2063398
Authors
References
- Basic Linear Algebra Subprograms for Fortran Usage
- The Design and Implementation of FFTW3
- Efficient Management of Parallelism in Object-Oriented Numerical Software Libraries
- Optimization principles and application performance evaluation of a multithreaded GPU using CUDA
- X10
- More iteration space tiling
- Co-array Fortran for parallel programming
- Stencil computation optimization and auto-tuning on state-of-the-art multicore architectures
- Parallel Programmability and the Chapel Language
- Introduction to UPC and Language Specification