@article{oai:osu.repo.nii.ac.jp:00002157,
 author = {伊藤, 誠 and ITOH, Makoto},
 issue = {2},
 journal = {大阪産業大学経済論集, OSAKA SANGYO UNIVERSITY JOURNAL OF ECONOMICS},
 month = {Mar},
 note = {We constructed the personal parallel calculation environment  using  GPU  and  CUDA Toolkit. It is very important to recognize the concept of Warp of NVIDIA GPU  in CUDA programming. We showed that Warp size(32) had the  decisive influence on  calculation  time.  We also showed the relation between number of threads and calculation time, and in the CUDA  program  the number of the threads in each block must be a multiple of 32(Warp size).},
 pages = {13--25},
 title = {GPU並列計算環境の構築とスレッド階層構造},
 volume = {20},
 year = {2019},
 yomi = {イトウ, マコト}
}