Identification and Elimination of the Overhead of Accelerate with a Super-resolution Application

• Item type: Trans(1)
• 公開日: 2017-12-12
• タイトル: Identification and Elimination of the Overhead of Accelerate with a Super-resolution Application
• タイトル / 言語: en
• タイトル: Identification and Elimination of the Overhead of Accelerate with a Super-resolution Application
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: [通常論文] GPGPU, Accelerate
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_6501
• 資源タイプ: journal article
• 著者所属: School of Computing, Tokyo Institute of Technology
• 著者所属: School of Computing, Tokyo Institute of Technology
• 著者所属: Graduate School of Engineering, Kochi University of Technology
• 著者所属: School of Information, Kochi University of Technology
• 著者所属(英): en
• 著者所属(英): School of Computing, Tokyo Institute of Technology
• 著者所属(英): en
• 著者所属(英): School of Computing, Tokyo Institute of Technology
• 著者所属(英): en
• 著者所属(英): Graduate School of Engineering, Kochi University of Technology
• 著者所属(英): en
• 著者所属(英): School of Information, Kochi University of Technology
• 著者名: Izumi, Asakura; Hidehiko, Masuhara; Takuya, Matsumoto; Kiminori, Matsuzaki
• 著者名(英): Izumi, Asakura; Hidehiko, Masuhara; Takuya, Matsumoto; Kiminori, Matsuzaki
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: This paper supplements the evaluation of libraries for parallel computing in Haskell by Matsumoto and Matsuzaki. The previous work implemented an image processing program by using two different libraries and compared expressiveness and parallel execution performance. In particular, it found that a parallel execution of the Accelerate program on a GPU is slower than the execution of the handwritten CUDA-C program by a factor of more than 70. We analyzed the execution of the Accelerate program and identified that the major sources of the overhead are in the process of the syntax tree before execution of a GPU kernel. Since the program realizes repeated execution by constructing a large syntax tree whose size is proportional to the number of repetitions, we rewrote this part by using a while-construct, which significantly reduces the overhead as it runs half as fast as the CUDA-C program. We also discuss language design alternatives for avoiding this type of overhead.; ------------------------------; This is a preprint of an article intended for publication Journal of; Information Processing(JIP). This preprint should not be cited. This; article should be cited as: Journal of Information Processing Vol.26(2018) (online); ------------------------------
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: This paper supplements the evaluation of libraries for parallel computing in Haskell by Matsumoto and Matsuzaki. The previous work implemented an image processing program by using two different libraries and compared expressiveness and parallel execution performance. In particular, it found that a parallel execution of the Accelerate program on a GPU is slower than the execution of the handwritten CUDA-C program by a factor of more than 70. We analyzed the execution of the Accelerate program and identified that the major sources of the overhead are in the process of the syntax tree before execution of a GPU kernel. Since the program realizes repeated execution by constructing a large syntax tree whose size is proportional to the number of repetitions, we rewrote this part by using a while-construct, which significantly reduces the overhead as it runs half as fast as the CUDA-C program. We also discuss language design alternatives for avoiding this type of overhead.; ------------------------------; This is a preprint of an article intended for publication Journal of; Information Processing(JIP). This preprint should not be cited. This; article should be cited as: Journal of Information Processing Vol.26(2018) (online); ------------------------------
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AA11464814
• 書誌情報: 情報処理学会論文誌プログラミング（PRO）; 巻 10, 号 6, 発行日 2017-12-12
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 1882-7802
• 出版者 / 言語: ja
• 出版者: 情報処理学会