Future Possibilities and Effectiveness of JIT from Elixir Code of Image Processing and Machine Learning into Native Code with SIMD Instructions

• Item type: Symposium(1)
• 公開日: 2022-01-07
• タイトル: Future Possibilities and Effectiveness of JIT from Elixir Code of Image Processing and Machine Learning into Native Code with SIMD Instructions
• タイトル / 言語: en
• タイトル: Future Possibilities and Effectiveness of JIT from Elixir Code of Image Processing and Machine Learning into Native Code with SIMD Instructions
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: Code optimization, image processing, JIT, mechine learning, SIMD
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_5794
• 資源タイプ: conference paper
• 著者所属(英): en
• 著者所属(英): Univ. of Kitakyushu
• 著者名: 山崎, 進
• 著者名(英): Susumu, Yamazaki
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: Nx is a multi-dimensional tensor library for Elixir with multi-staged compilation to the CPU or GPU, similar to NumPy and TensorFlow in Python. Nx is expected to be applied in image processing and machine learning. Code used by image processing and machine learning in C or C++ is often optimized for CPUs into native code with SIMD instructions. In this paper, we will show that native code with SIMD instructions is 1000x+ faster than equivalent Elixir code with Nx, to evaluate future possibilities and effectiveness of such code generation and optimization. Our future works are to implement and evaluate our proposal: a backend of Nx generating SIMD instructions by NIFs and/or BeamAsm using our compiler and/or OpenBLAS or cuBLAS.
• 書誌情報: 第63回プログラミング・シンポジウム予稿集; 巻 2022, p. 73-78, 発行日 2022-01-07
• 出版者 / 言語: ja
• 出版者: 情報処理学会