{"id":53113,"date":"2018-10-19T12:04:07","date_gmt":"2018-10-19T04:04:07","guid":{"rendered":"http:\/\/www.homejamesfinance.com\/?p=53113"},"modified":"2018-10-19T23:19:35","modified_gmt":"2018-10-19T15:19:35","slug":"ibm%e4%b8%8envidia%e6%90%ba%e6%89%8b%e4%b8%ba%e6%95%b0%e6%8d%ae%e7%a7%91%e5%ad%a6%e5%ae%b6%e6%8b%93%e5%b1%95%e5%bc%80%e6%ba%90%e6%9c%ba%e5%99%a8%e5%ad%a6%e4%b9%a0%e5%b7%a5%e5%85%b7","status":"publish","type":"post","link":"https:\/\/www.homejamesfinance.com\/p\/53113.html","title":{"rendered":"IBM\u4e0eNVIDIA\u643a\u624b\u4e3a\u6570\u636e\u79d1\u5b66\u5bb6\u62d3\u5c55\u5f00\u6e90\u673a\u5668\u5b66\u4e60\u5de5\u5177"},"content":{"rendered":"

\u57fa\u4e8eIBM POWER9\u7684PowerAI\u3001IBM Cloud\u548cIBM Watson Studio\u5c06\u652f\u6301\u5168\u65b0GPU\u52a0\u901f\u8f6f\u4ef6\u5e93RAPIDS<\/p>\n

2018\u5e7410\u6708\uff0cIBM(NYSE: IBM)\u5ba3\u5e03\u8ba1\u5212\u5c06\u5168\u65b0RAPIDS\u5f00\u6e90\u8f6f\u4ef6\u7eb3\u5165\u5230\u5176\u4f01\u4e1a\u7ea7\u6570\u636e\u79d1\u5b66\u5e73\u53f0\u4e2d\uff0c\u6db5\u76d6\u672c\u5730\u9884\u7f6e\u3001\u6df7\u5408\u4e91\u548c\u591a\u4e91\u73af\u5883\u3002\u51ed\u501f\u5176\u5e9e\u5927\u7684\u6df1\u5ea6\u5b66\u4e60\u4e0e\u673a\u5668\u5b66\u4e60\u89e3\u51b3\u65b9\u6848\u7ec4\u5408\uff0cIBM\u80fd\u4e3a\u504f\u597d\u4e0d\u540c\u90e8\u7f72\u6a21\u578b\u7684\u6570\u636e\u79d1\u5b66\u5bb6\u63d0\u4f9b\u8fd9\u4e00\u5f00\u6e90\u6280\u672f\u3002<\/p>\n

IBM\u8ba4\u77e5\u7cfb\u7edf\u65b9\u6848\u9ad8\u7ea7\u526f\u603b\u88c1Bob Picciano\u8868\u793a\uff1a\u201cIBM\u957f\u671f\u4e0eNVIDIA\u5408\u4f5c\uff0c\u5229\u7528IBM POWER9\u5904\u7406\u5668\uff0c\u7ed3\u5408NVIDIA GPU\u7b49\u6280\u672f\uff0c\u5b9e\u73b0\u663e\u8457\u7684\u6027\u80fd\u63d0\u5347\u3002\u6211\u4eec\u5c06RAPIDS\u7eb3\u5165IBM\u4ea7\u54c1\u7ec4\u5408\u7684\u540c\u65f6\uff0c\u4e5f\u671f\u5f85\u80fd\u591f\u7ee7\u7eed\u79ef\u6781\u5730\u5e2e\u52a9\u5ba2\u6237\u7a81\u7834\u4eba\u5de5\u667a\u80fd<\/a>\u7684\u6027\u80fd\u754c\u9650\u3002\u201d<\/p>\n

\u501f\u52a9Apache Arrow\u3001Pandas\u548cScikit-Learn\u7b49\u5f00\u6e90\u673a\u5668\u5b66\u4e60\u8f6f\u4ef6\u7684\u4f18\u52bf\uff0cRAPIDS\u5c06GPU\u7684\u52a0\u901f\u80fd\u529b\u878d\u5165IBM\u4ea7\u54c1\u4e2d\u3002\u5305\u62ecAnaconda\u3001BlazingDB\u3001Graphistry\u3001NERSC\u3001PyData\u3001INRIA\u548cUrsa Labs\u5728\u5185\u7684\u4e3b\u8981\u5f00\u6e90\u8d21\u732e\u8005\uff0c\u90fd\u5728RAPIDS\u63a8\u51fa\u540e\u7acb\u5373\u7ed9\u4e88\u4e86\u5176\u5e7f\u6cdb\u7684\u751f\u6001\u7cfb\u7edf\u652f\u6301\u3002<\/p>\n

IBM\u8ba1\u5212\u5c06RAPIDS\u5f15\u5165\u672c\u5730\u9884\u7f6e\u3001\u516c\u6709\u4e91\u3001\u6df7\u5408\u4e91\u4e0e\u591a\u4e91\u73af\u5883\u7b49\u4e3b\u8981\u9886\u57df\uff0c\u540c\u65f6\u57fa\u4e8eIBM POWER9\u7684PowerAI\u5c06\u501f\u52a9RAPIDS\u4e3a\u6570\u636e\u79d1\u5b66\u5bb6\u6269\u5c55\u66f4\u591a\u5f00\u6e90\u673a\u5668\u5b66\u4e60\u4e0e\u5206\u6790\u5e93\u7684\u53ef\u7528\u9009\u9879\u3002\u4e8b\u5b9e\u8bc1\u660e\uff0cNVIDIA\u4e0eIBM\u56f4\u7ed5POWER9\u5904\u7406\u5668\u6240\u505a\u7684\u4e13\u9879\u5de5\u7a0b\uff0c\u5305\u62ec\u5c06NVIDIA NVLink \u548cNVIDIA Tesla TensorCore GPU\u8fdb\u884c\u96c6\u6210\u6240\u505a\u7684\u52aa\u529b\uff0c\u76f4\u63a5\u6709\u5229\u4e8e\u52a0\u901f\u5de5\u4f5c\u8d1f\u8f7d\u3002\u4f5c\u4e3aIBM\u7684\u8f6f\u4ef6\u5c42\uff0c\u5982\u4ecaPowerAI\u53ef\u4ee5\u4f18\u5316\u8fd0\u884c\u5728\u5f02\u6784\u8ba1\u7b97\u7cfb\u7edf\u4e0a\u7684\u6570\u636e\u79d1\u5b66\u4e0eAI<\/a>\u5de5\u4f5c\u8d1f\u8f7d\uff0c\u540c\u65f6\u501f\u52a9RAPIDS\u6301\u7eed\u63d0\u5347POWER9\u4e0a\u7ecfGPU\u52a0\u901f\u540e\u7684\u5de5\u4f5c\u8d1f\u8f7d\u7684\u6027\u80fd\u3002<\/p>\n

IBM Watson Studio\u4e0eWatson\u673a\u5668\u5b66\u4e60(Watson Machine Learning)\u5229\u7528NVIDIA GPU\u7684\u5f3a\u5927\u529f\u80fd\uff0c\u4f7f\u6570\u636e\u79d1\u5b66\u5bb6\u548cAI\u5f00\u53d1\u4eba\u5458\u80fd\u591f\u900f\u8fc7IBM Cloud Private for Data \u548cIBM Cloud\uff0c\u5728\u591a\u4e91\u73af\u5883\u4e2d\u6784\u5efa\u3001\u90e8\u7f72\u548c\u8fd0\u884c\u901f\u5ea6\u66f4\u5feb\u7684\u6a21\u578b\uff0c\u8fdc\u8d85\u4ec5\u91c7\u7528CPU\u7684\u90e8\u7f72\u65b9\u5f0f\u3002<\/p>\n

IBM Cloud\u80fd\u591f\u5e2e\u52a9\u91c7\u7528GPU\u8bbe\u5907\u7684\u7528\u6237\uff0c\u5c06RAPIDS\u4e2d\u7684\u52a0\u901f\u673a\u5668\u5b66\u4e60\u4e0e\u5206\u6790\u5e93\u5e94\u7528\u4e8e\u5176\u4e91\u7aef\u5e94\u7528\uff0c\u4ece\u800c\u53d7\u76ca\u4e8e\u673a\u5668\u5b66\u4e60\u3002<\/p>\n

NVIDIA\u526f\u603b\u88c1\u3001\u52a0\u901f\u8ba1\u7b97\u90e8\u603b\u7ecf\u7406Ian Buck\u8868\u793a\uff1a\u201cIBM\u4e0eNVIDIA\u591a\u5e74\u6765\u5bc6\u5207\u5408\u4f5c\uff0c\u5df2\u7ecf\u5e2e\u52a9\u5168\u7403\u9886\u5148\u7684\u4f01\u4e1a\u548c\u673a\u6784\u89e3\u51b3\u4e86\u8bf8\u591a\u5982\u4eca\u4e16\u754c\u4e0a\u6700\u5927\u7684\u96be\u9898\u3002\u968f\u7740IBM\u5bf9NVIDIA RAPIDS\u5f00\u6e90\u5e93\u7684\u4f7f\u7528\uff0c\u6570\u636e\u79d1\u5b66\u5bb6\u5c06\u501f\u52a9GPU\u52a0\u901f\u673a\u5668\u5b66\u4e60\u80fd\u529b\uff0c\u4ee5\u524d\u6240\u672a\u6709\u7684\u901f\u5ea6\u66f4\u5feb\u5730\u5206\u6790\u5927\u6570\u636e\u3002\u201d<\/p>\n

\u673a\u5668\u5b66\u4e60\u662fAI\u7684\u4e00\u79cd\u5f62\u5f0f\uff0c\u53ef\u4ee5\u8ba9\u7cfb\u7edf\u900f\u8fc7\u6570\u636e\u800c\u975e\u4e13\u7528\u7a0b\u5e8f\u8fdb\u884c\u5b66\u4e60\u3002\u96f6\u552e\u3001\u91d1\u878d\u548c\u7535\u4fe1\u7b49\u9886\u57df\u7684\u4f17\u591a\u4f01\u4e1a\u8981\u4e48\u6b63\u5728\u4e3b\u52a8\u4f7f\u7528\u673a\u5668\u5b66\u4e60\uff0c\u8981\u4e48\u6b63\u5728\u63a2\u7d22\u673a\u5668\u5b66\u4e60\uff0c\u90fd\u8bd5\u56fe\u5229\u7528\u5927\u6570\u636e\u66f4\u597d\u5730\u4e86\u89e3\u6d88\u8d39\u8005\u7684\u884c\u4e3a\u3001\u504f\u597d\u6216\u5ba2\u6237\u6ee1\u610f\u5ea6\u4e2d\u5fae\u5999\u7684\u53d8\u5316\uff0c\u4ece\u800c\u4e3a\u4f01\u4e1a\u5e26\u6765\u6f5c\u5728\u7684\u4ef7\u503c\u3002<\/p>\n

\u4eca\u5e74\u65e9\u4e9b\u65f6\u5019\uff0cIBM\u4ee5\u8d85\u51fa\u4e4b\u524d\u7eaa\u5f55\u4fdd\u6301\u800546\u500d\u7684\u901f\u5ea6\uff0c\u5728\u5146\u7ea7\u673a\u5668\u5b66\u4e60\u57fa\u51c6\u4e0a\u7f14\u9020\u4e86\u65b0\u7684\u7eaa\u5f55\u3002IBM\u7814\u7a76\u4eba\u5458\u5229\u7528IBM Research\u5f00\u53d1\u7684IBM Snap Machine Learning(Snap ML)\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\uff0c\u5728\u642d\u8f7dNVIDIA Tesla V100 Tensor Core GPU\u7684IBM Power Systems AC922\u670d\u52a1\u5668\u4e0a\u8fd0\u884c\uff0c\u4ec5\u82b1\u8d3991.5\u79d2\u5c31\u57fa\u4e8e\u7531Criteo Labs\u53d1\u5e03\u7684\u8d8540\u4ebf\u4e2a\u5728\u7ebf\u5e7f\u544a\u6570\u636e\u6848\u4f8b\uff0c\u5b8c\u6210\u4e86\u5bf9\u903b\u8f91\u56de\u5f52\u5206\u7c7b\u5668\u7684\u8bad\u7ec3\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"

\u57fa\u4e8eIBM POWER9\u7684PowerAI\u3001IBM Cloud\u548cIBM Watson Studio\u5c06\u652f\u6301\u5168\u65b0GPU\u52a0\u901f\u8f6f\u4ef6\u5e93RAPIDS 2018\u5e7410\u6708\uff0cIBM(NYSE: IBM)\u5ba3\u5e03\u8ba1\u5212\u5c06\u5168\u65b0RAPIDS\u5f00\u6e90\u8f6f\u4ef6\u7eb3\u5165\u5230\u5176\u4f01\u4e1a\u7ea7\u6570\u636e\u79d1\u5b66\u5e73…<\/p>\n","protected":false},"author":164,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[4],"tags":[3661,4845,3894,16641],"topic":[],"class_list":["post-53113","post","type-post","status-publish","format-standard","hentry","category-tech","tag-IBM","tag-NVIDIA","tag-kaiyuan","tag-16641"],"_links":{"self":[{"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/posts\/53113"}],"collection":[{"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/users\/164"}],"replies":[{"embeddable":true,"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/comments?post=53113"}],"version-history":[{"count":1,"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/posts\/53113\/revisions"}],"predecessor-version":[{"id":53114,"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/posts\/53113\/revisions\/53114"}],"wp:attachment":[{"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/media?parent=53113"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/categories?post=53113"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/tags?post=53113"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/www.homejamesfinance.com\/wp-json\/wp\/v2\/topic?post=53113"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}精品国产午夜肉伦伦影院,双性老师灌满浓jing上课h,天天做天天爱夜夜爽,攵女乱h边做边走