Pure C plus plus Approach to Optimized Parallel Traversal of Regular Data Structures

Many computational problems consider memory throughput a performance bottleneck. The problem becomes even more pronounced in the case of parallel platforms, where the ratio between computing elements and memory bandwidth shifts towards computing. Software needs to be attuned to hardware features like cache architectures or memory banks to reach a decent level of performance efficiency. This can be achieved by selecting the right memory layouts for data structures or changing the order of data structure traversal. In this work, we present an abstraction for traversing a set of regular data structures (e.g., multidimensional arrays) that allows the design of traversal-agnostic algorithms. Such algorithms can be adjusted for particular memory layouts of the data structures, semi-automated parallelization, or auto-tuning without altering their internal code. The proposed solution was implemented as an extension of the Noarr library that simplifies a layout-agnostic design of regular data structures. It is implemented entirely using C++ template meta-programming without any nonstandard dependencies, so it is fully compatible with existing compilers, including CUDA NVCC. We evaluate the performance and expressiveness of our approach on the Polybench-C benchmarks.