{"categories":[],"author_url":"https://blog.hatena.ne.jp/fortran66/","width":"100%","version":"1.0","author_name":"fortran66","url":"https://fortran66.hatenablog.com/entry/2020/08/08/191522","provider_url":"https://hatena.blog","height":"190","image_url":"https://m.media-amazon.com/images/I/51-P+arTLyL._SL160_.jpg","type":"rich","provider_name":"Hatena Blog","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Ffortran66.hatenablog.com%2Fentry%2F2020%2F08%2F08%2F191522\" title=\"\u3010\u30e1\u30e2\u5e33\u3011Fortran news - fortran66\u306e\u30d6\u30ed\u30b0\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","title":"\u3010\u30e1\u30e2\u5e33\u3011Fortran news","published":"2020-08-08 19:15:22","blog_url":"https://fortran66.hatenablog.com/","blog_title":"fortran66\u306e\u30d6\u30ed\u30b0","description":"Tensor Cores developer.nvidia.com \u666e\u901a\u306e Fortran \u3067\u66f8\u3044\u3066\u3001\u884c\u5217\u6f14\u7b97\u547d\u4ee4\u3092\u81ea\u52d5\u3067 GPU \u5074\u3067\u5b9f\u884c\u3057\u3066\u304f\u308c\u308b\u3088\u3046\u3067\u3059\u3002 memo docs.nvidia.com \u30ed\u30b7\u30a2\u8a9e\u3067\u8003\u3048\u308b\u3093\u3060\uff01 \u96fb\u5b50\u7248 \uff14\uff16\uff18\u5186 books.google.co.jp LLNL \u306e Fortran + GPU Closing the FORTRAN gap on next-generation architectures computing.llnl.gov LLNL \u306e\u6b21\u671f\u30a8\u30af\u30b5\u30b9\u30d1\u30b3\u30f3\u306f AMD + Radeon \u3060\u3063\u305f\u304b\u3002\u4eca\u306f IBM + Nvidia \uff1f \u65e5\u672c\u306f\u3001\u3068\u308a\u3042\u3048\u2026"}