{"id":67781,"date":"2024-05-14T06:18:58","date_gmt":"2024-05-13T22:18:58","guid":{"rendered":"https:\/\/www.ioiotimes.com\/?p=67781"},"modified":"2024-05-14T06:18:59","modified_gmt":"2024-05-13T22:18:59","slug":"nvidia%e9%80%8f%e9%81%8ecuda-q%e5%b9%b3%e5%8f%b0%e5%8a%a0%e9%80%9f%e5%85%a8%e7%90%83%e9%87%8f%e5%ad%90%e9%81%8b%e7%ae%97%e4%b8%ad%e5%bf%83","status":"publish","type":"post","link":"https:\/\/www.ioiotimes.com\/?p=67781","title":{"rendered":"NVIDIA\u900f\u904eCUDA-Q\u5e73\u53f0\u52a0\u901f\u5168\u7403\u91cf\u5b50\u904b\u7b97\u4e2d\u5fc3"},"content":{"rendered":"\n

\u5fb7\u570b\u3001\u65e5\u672c\u548c\u6ce2\u862d\u7684\u8d85\u7d1a\u96fb\u8166\u63a1\u7528Grace-Hopper\u548c\u91cf\u5b50\u7d93\u5178\u52a0\u901f\u8d85\u7d1a\u904b\u7b97\u5e73\u53f0\u4f86\u63a8\u9032\u91cf\u5b50\u904b\u7b97\u7814\u7a76<\/h3>\n\n\n\n

NVIDIA \u4eca\u5929\u5ba3\u5e03\u5c07\u5229\u7528\u958b\u6e90 NVIDIA CUDA-Q\u2122 \u5e73\u53f0<\/a>\u52a0\u901f\u5168\u7403\u570b\u5bb6\u8d85\u7d1a\u904b\u7b97\u4e2d\u5fc3\u7684\u91cf\u5b50\u904b\u7b97\u5de5\u4f5c\u3002<\/p>\n\n\n\n

\u5fb7\u570b\u3001\u65e5\u672c\u548c\u6ce2\u862d\u7684\u8d85\u7d1a\u904b\u7b97\u7ad9\u9ede\u5c07\u4f7f\u7528\u8a72\u5e73\u53f0\u70ba\u5176 NVIDIA \u52a0\u901f\u7684\u9ad8\u6548\u80fd\u904b\u7b97\u7cfb\u7d71\u5167\u7684\u91cf\u5b50\u8655\u7406\u55ae\u5143<\/a>\uff08QPU\uff09\u63d0\u4f9b\u52d5\u529b\u3002<\/p>\n\n\n\n

QPU \u662f\u91cf\u5b50\u96fb\u8166\u7684\u5927\u8166\uff0c\u5b83\u5229\u7528\u96fb\u5b50\u6216\u5149\u5b50\u7b49\u7c92\u5b50\u7684\u884c\u70ba\u9032\u884c\u8207\u50b3\u7d71\u8655\u7406\u5668\u4e0d\u540c\u7684\u904b\u7b97\uff0c\u6709\u53ef\u80fd\u4f7f\u67d0\u4e9b\u985e\u578b\u7684\u904b\u7b97\u66f4\u5feb\u3002<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

<\/p>\n\n\n\n

\u5fb7\u570b\u65bc\u5229\u5e0c\u7814\u7a76\u4e2d\u5fc3\u7684\u65bc\u5229\u5e0c\u8d85\u7d1a\u904b\u7b97\u4e2d\u5fc3\uff08JSC\uff09\u6b63\u5728\u5b89\u88dd\u7531 IQM Quantum Computers \u6253\u9020\u7684 QPU\uff0c\u88dc\u5145\u5176 JUPITER \u8d85\u7d1a\u96fb\u8166\uff0c\u4e26\u7531 NVIDIA GH200 Grace Hopper\u2122 \u8d85\u7d1a\u6676\u7247<\/a>\u63d0\u4f9b\u589e\u5f37\u529f\u80fd\u3002<\/p>\n\n\n\n

\u4f4d\u65bc\u65e5\u672c\u7522\u696d\u6280\u8853\u7d9c\u5408\u7814\u7a76\u6240\uff08AIST\uff09\u7684 ABCI-Q \u8d85\u7d1a\u96fb\u8166\uff0c\u65e8\u5728\u63a8\u9032\u570b\u5bb6\u7684\u91cf\u5b50\u904b\u7b97\u8a08\u5283\u3002 \u8a72\u7cfb\u7d71\u7531 NVIDIA Hopper\u2122 \u67b6\u69cb\u9a45\u52d5\uff0c\u5c07\u6dfb\u52a0 QuEra \u516c\u53f8\u7684 QPU\u3002<\/p>\n\n\n\n

\u6ce2\u862d\u6ce2\u8332\u5357\u8d85\u7d1a\u904b\u7b97\u548c\u7db2\u8def\u4e2d\u5fc3\uff08PSNC\uff09\u6700\u8fd1\u5b89\u88dd\u4e86\u5169\u500b\u7531 ORCA Computing \u5efa\u69cb\u7684\u5149\u5b50 QPU\uff0c\u9023\u63a5\u5230\u7531 NVIDIA Hopper \u52a0\u901f\u7684\u65b0\u8d85\u7d1a\u96fb\u8166\u5206\u5340\u3002<\/p>\n\n\n\n

NVIDIA \u91cf\u5b50\u548c\u9ad8\u6548\u80fd\u904b\u7b97\u7e3d\u76e3 Tim Costa \u8868\u793a\uff1a\u300c\u6709\u7528\u7684\u91cf\u5b50\u904b\u7b97\u5c07\u900f\u904e\u91cf\u5b50\u8207 GPU \u8d85\u7d1a\u904b\u7b97\u7684\u7dca\u5bc6\u6574\u5408\u4f86\u5be6\u73fe\u3002NVIDIA \u7684\u91cf\u5b50\u904b\u7b97\u5e73\u53f0\u70ba AIST\u3001JSC \u548c PSNC \u7b49\u5148\u9a45\u8005\u63d0\u4f9b\u4e86\u5e6b\u52a9\uff0c\u4ee5\u7a81\u7834\u79d1\u5b78\u63a2\u7d22\u7684\u754c\u9650\uff0c\u4e26\u63a8\u9032\u91cf\u5b50\u6574\u5408\u8d85\u7d1a\u904b\u7b97\u7684\u767c\u5c55\u3002\u300d<\/p>\n\n\n\n

\u8207 ABCI-Q \u6574\u5408\u7684 QPU \u5c07\u4f7f AIST \u7684\u7814\u7a76\u4eba\u54e1\u80fd\u5920\u5229\u7528\u7531\u96f7\u5c04\u5149\u675f\u63a7\u5236\u7684\u92a3\u539f\u5b50\u4f5c\u70ba\u91cf\u5b50\u4f4d\u5143\u4f86\u57f7\u884c\u904b\u7b97\uff0c\u7814\u7a76 AI\u3001\u80fd\u6e90\u548c\u751f\u7269\u5b78\u4e2d\u7684\u91cf\u5b50\u61c9\u7528\u3002\u6b64\u985e\u539f\u5b50\u8207\u7cbe\u5bc6\u539f\u5b50\u9418\u4e2d\u4f7f\u7528\u7684\u539f\u5b50\u985e\u578b\u76f8\u540c\u3002\u6bcf\u500b\u539f\u5b50\u90fd\u662f\u76f8\u540c\u7684\uff0c\u9019\u63d0\u4f9b\u4e86\u4e00\u7a2e\u6709\u6f5b\u529b\u5be6\u73fe\u5927\u898f\u6a21\u3001\u9ad8\u4fdd\u771f\u91cf\u5b50\u8655\u7406\u5668\u7684\u65b9\u6cd5\u3002<\/p>\n\n\n\n

G-QuAT\/AIST \u526f\u7e3d\u76e3\u5800\u90e8\u96c5\u5f18\u8868\u793a\uff1a\u300c\u65e5\u672c\u7814\u7a76\u4eba\u54e1\u5c07\u5229\u7528 ABCI-Q \u91cf\u5b50\u7d93\u5178\u52a0\u901f\u8d85\u7d1a\u96fb\u8166\u5728\u5be6\u969b\u91cf\u5b50\u904b\u7b97\u61c9\u7528\u65b9\u9762\u53d6\u5f97\u9032\u5c55\u3002NVIDIA \u6b63\u5728\u5e6b\u52a9\u9019\u4e9b\u5148\u9a45\u8005\u7a81\u7834\u91cf\u5b50\u904b\u7b97\u7814\u7a76\u7684\u754c\u9650\u3002\u300d<\/p>\n\n\n\n

PSNC \u7684 QPU \u5c07\u4f7f\u7814\u7a76\u4eba\u54e1\u80fd\u5920\u5229\u7528\u5169\u500b PT-1 \u91cf\u5b50\u5149\u5b50\u7cfb\u7d71\u63a2\u7d22\u751f\u7269\u5b78\u3001\u5316\u5b78\u548c\u6a5f\u5668\u5b78\u7fd2\u3002\u8a72\u7cfb\u7d71\u4f7f\u7528\u96fb\u4fe1\u983b\u7387\u7684\u55ae\u5149\u5b50\uff0c\u4e5f\u5c31\u662f\u5149\u7684\u6700\u5c0f\u55ae\u4f4d\uff0c\u4f5c\u70ba\u91cf\u5b50\u4f4d\u5143\u3002\u9019\u5141\u8a31\u4f7f\u7528\u6a19\u6e96\u7684\u3001\u73fe\u6210\u7684\u96fb\u4fe1\u96f6\u4ef6\u4f86\u5be6\u73fe\u5206\u6563\u5f0f\u3001\u53ef\u64f4\u5c55\u548c\u6a21\u7d44\u5316\u7684\u91cf\u5b50\u67b6\u69cb\u3002<\/p>\n\n\n\n

PSNC \u6280\u8853\u9577\u66a8\u526f\u7e3d\u76e3 Krzysztof Kurowski \u8868\u793a\uff1a\u300c\u6211\u5011\u8207 ORCA \u548c NVIDIA \u7684\u5408\u4f5c\u4f7f\u6211\u5011\u80fd\u5920\u5275\u9020\u4e00\u500b\u7368\u7279\u7684\u74b0\u5883\uff0c\u4e26\u5728 PSNC \u69cb\u5efa\u4e00\u500b\u65b0\u7684\u91cf\u5b50\u7d93\u5178\u6df7\u5408\u7cfb\u7d71\u3002\u7531\u4ee5\u4f7f\u7528\u8005\u70ba\u4e2d\u5fc3\u7684\u670d\u52d9\u6709\u6548\u7ba1\u7406\u591a\u500b QPU \u548c GPU \u7684\u958b\u653e\u3001\u8f15\u9b06\u6574\u5408\u548c\u7a0b\u5f0f\u958b\u767c\uff0c\u5c0d\u65bc\u958b\u767c\u4eba\u54e1\u548c\u4f7f\u7528\u8005\u81f3\u95dc\u91cd\u8981\u3002\u9019\u7a2e\u5bc6\u5207\u7684\u5408\u4f5c\u8b93\u6211\u5011\u5728\u73fe\u5728\uff0c\u800c\u4e0d\u662f\u660e\u5929\uff0c\u70ba\u65b0\u4e00\u4ee3\u91cf\u5b50\u52a0\u901f\u8d85\u7d1a\u96fb\u8166\u958b\u555f\u65bc\u8a31\u591a\u5275\u65b0\u61c9\u7528\u9818\u57df\u3002\u300d<\/p>\n\n\n\n

\u8207 JUPITER \u6574\u5408\u7684 QPU \u5c07\u4f7f JSC \u7814\u7a76\u4eba\u54e1\u80fd\u5920\u958b\u767c\u5316\u5b78\u6a21\u64ec\u548c\u6700\u4f73\u5316\u554f\u984c\u7684\u91cf\u5b50\u61c9\u7528\uff0c\u4e26\u6f14\u793a\u5982\u4f55\u900f\u904e\u91cf\u5b50\u96fb\u8166\u52a0\u901f\u7d93\u5178\u8d85\u7d1a\u96fb\u8166\u3002\u5b83\u662f\u7528\u8d85\u5c0e\u91cf\u5b50\u4f4d\u5143\u6216\u96fb\u5b50\u8ae7\u632f\u96fb\u8def\u5efa\u69cb\u7684\uff0c\u4e26\u53ef\u4ee5\u64cd\u63a7\u5728\u4f4e\u6eab\u4e0b\u8868\u73fe\u5f97\u50cf\u4eba\u9020\u539f\u5b50\u3002<\/p>\n\n\n\n

JSC \u91cf\u5b50\u8cc7\u8a0a\u8655\u7406\u5c0f\u7d44\u8ca0\u8cac\u4eba Kristel Michielsen \u8868\u793a\uff1a\u300c\u6df7\u5408\u91cf\u5b50\u7d93\u5178\u52a0\u901f\u8d85\u7d1a\u904b\u7b97\u6b63\u5728\u4f7f\u91cf\u5b50\u904b\u7b97\u66f4\u52a0\u63a5\u8fd1\u3002\u900f\u904e\u6211\u5011\u8207 NVIDIA \u7684\u6301\u7e8c\u5408\u4f5c\uff0cJSC \u7684\u7814\u7a76\u4eba\u54e1\u5c07\u63a8\u52d5\u91cf\u5b50\u904b\u7b97\u4ee5\u53ca\u5316\u5b78\u548c\u6750\u6599\u79d1\u5b78\u9818\u57df\u7684\u767c\u5c55\u3002\u300d<\/p>\n\n\n\n

\u900f\u904e\u5c07\u91cf\u5b50\u96fb\u8166\u8207\u8d85\u7d1a\u96fb\u8166\u7dca\u5bc6\u6574\u5408\uff0cCUDA-Q \u9084\u652f\u63f4\u91cf\u5b50\u904b\u7b97\u8207 AI \u7d50\u5408<\/a>\uff0c\u89e3\u6c7a\u96dc\u8a0a\u91cf\u5b50\u4f4d\u5143\u7b49\u554f\u984c\u4e26\u958b\u767c\u9ad8\u6548\u6f14\u7b97\u6cd5\u3002<\/p>\n\n\n\n

CUDA-Q \u662f\u4e00\u500b\u958b\u6e90\u4e14\u8207\u652f\u63f4\u5404\u5f0fQPU\u7684\u91cf\u5b50\u7d93\u5178\u52a0\u901f\u8d85\u7d1a\u904b\u7b97\u5e73\u53f0\u3002CUDA-Q \u5df2\u88ab\u5927\u591a\u6578\u90e8\u7f72 QPU \u7684\u516c\u53f8\u4f7f\u7528\uff0c\u4e26\u63d0\u4f9b\u4e00\u6d41\u7684\u6548\u80fd<\/a>\u3002<\/p>\n\n\n\n

<\/p>\n\n\n\n

<\/p>\n\n\n\n


\ud83d\udfe6\u73fe\u5728\u5c31\u52a0\u5165 ioioTIMES \u81c9\u66f8\u7c89\u7d72\u5718<\/a> \u66f4\u591a\u4e92\u52d5\u3001\u66f4\u591a\u597d\u5eb7\u650f\u62b5\u52a0!!<\/strong>
\ud83d\udfe6\u6211\u5011\u6709LINE TODAY<\/a>\u983b\u9053\u4e86\uff0c\u5feb\u4f86\u8ffd\u8e2a\u6211\u5011\u5427!!–\u6700\u65b0\u79d1\u6280\u65b0\u805e \u76e1\u5728\u4f60\u624b<\/strong><\/h4>\n","protected":false},"excerpt":{"rendered":"

\u5fb7\u570b\u3001\u65e5\u672c\u548c\u6ce2\u862d\u7684\u8d85\u7d1a\u96fb\u8166\u63a1\u7528Gra<\/p>\n","protected":false},"author":3,"featured_media":67825,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"rank_math_lock_modified_date":false,"footnotes":""},"categories":[13],"tags":[9604,580,71,9605,5987],"class_list":["post-67781","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news","tag-cuda-q","tag-focus","tag-nvidia","tag-qpu","tag-5987"],"_links":{"self":[{"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/posts\/67781"}],"collection":[{"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=67781"}],"version-history":[{"count":3,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/posts\/67781\/revisions"}],"predecessor-version":[{"id":67827,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/posts\/67781\/revisions\/67827"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/media\/67825"}],"wp:attachment":[{"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=67781"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=67781"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=67781"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}