TTC:一种适用于多体系结构的张量转置算法
TTC: A Tensor Transposition Compiler for Multiple Architectures
摘要:
我们考虑转置张量的任意维的问题,并描述TTC,一个开源的特定领域的并行编译器。TTC生成优化的并行C++/CUDA C代码,实现系统峰值内存带宽的显著部分。TTC在多种架构中表现出高性能,包括现代基于AVX的系统(例如,Intel Haswell、AMD Steamroller)、Intel的Knights Corner以及不同的基于CUDA的GPU,如NVIDIA的Kepler和Maxwell架构。我们报告了TTC的加速比外部C++编译器生成的有意义的基线实现;结果表明,特定领域的编译器可以大大优于其通用的同行:例如,与英特尔最新的C++编译器的Haswell和骑士角架构相比,TTC产生的加速比高达$8\times$和$32\times$,分别。我们还展示了TTC对多个领先维度的支持,使其成为生成无处不在的BLAS 3例程核心的性能关键打包函数的合适候选者。
英文摘要:
We consider the problem of transposing tensors of arbitrary dimension and describe TTC, an open source domain-specific parallel compiler. TTC generates optimized parallel C++/CUDA C code that achieves a significant fraction of the system's peak memory bandwidth. TTC exhibits high performance across multiple architectures, including modern AVX-based systems (e.g.,~Intel Haswell, AMD Steamroller), Intel's Knights Corner as well as different CUDA-based GPUs such as NVIDIA's Kepler and Maxwell architectures. We report speedups of TTC over a meaningful baseline implementation generated by external C++ compilers; the results suggest that a domain-specific compiler can outperform its general purpose counterpart significantly: For instance, comparing with Intel's latest C++ compiler on the Haswell and Knights Corner architecture, TTC yields speedups of up to $8\times$ and $32\times$, respectively. We also showcase TTC's support for multiple leading dimensions, making it a suitable candidate for the generation of performance-critical packing functions that are at the core of the ubiquitous BLAS 3 routines.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。