Litcius/Paper detail

FlexTensor

Size Zheng, Yun Liang, Shuo Wang, Renze Chen, Kaiwen Sheng

2020164 citationsDOI

Abstract

Tensor computation plays a paramount role in a broad range of domains, including machine learning, data analytics, and scientific computing. The wide adoption of tensor computation and its huge computation cost has led to high demand for flexible, portable, and high-performance library implementation on heterogeneous hardware accelerators such as GPUs and FPGAs. However, the current tensor library implementation mainly requires programmers to manually design low-level implementation and optimize from the algorithm, architecture, and compilation perspectives. Such a manual development process often takes months or even years, which falls far behind the rapid evolution of the application algorithms.

Topics & Concepts

Computer scienceComputationField-programmable gate arrayProcess (computing)Tensor (intrinsic definition)Computer architectureAnalyticsRange (aeronautics)ArchitectureParallel computingComputer engineeringComputational scienceEmbedded systemData scienceProgramming languageComposite materialArtMathematicsPure mathematicsMaterials scienceVisual artsParallel Computing and Optimization TechniquesComputational Physics and Python ApplicationsTensor decomposition and applications