{"id":22208,"date":"2024-03-28T15:26:49","date_gmt":"2024-03-28T08:26:49","guid":{"rendered":"https:\/\/gudjob.net\/?p=22208"},"modified":"2026-08-18T08:33:57","modified_gmt":"2026-08-18T01:33:57","slug":"intel-google-va-qualcomm-tim-cach-lat-do-nvidia-bang-nen-tang-lap-trinh-moi","status":"publish","type":"post","link":"https:\/\/gudjob.net\/en\/blog\/intel-google-va-qualcomm-tim-cach-lat-do-nvidia-bang-nen-tang-lap-trinh-moi\/","title":{"rendered":"Intel, Google v\u00e0 Qualcomm t\u00ecm c\u00e1ch ‘l\u1eadt \u0111\u1ed5’ Nvidia b\u1eb1ng n\u1ec1n t\u1ea3ng l\u1eadp tr\u00ecnh m\u1edbi"},"content":{"rendered":"

M\u1ed9t li\u00ean minh c\u00f4ng ngh\u1ec7 g\u1ed3m Intel, Google v\u00e0 Qualcomm \u0111ang t\u00ecm c\u00e1ch ph\u00e1t tri\u1ec3n n\u1ec1n t\u1ea3ng ph\u1ea7n m\u1ec1m m\u1edbi gi\u00fap c\u00e1c nh\u00e0 l\u1eadp tr\u00ecnh tho\u00e1t kh\u1ecfi ph\u1ee5 thu\u1ed9c v\u00e0o n\u1ec1n t\u1ea3ng CUDA do Nvidia ph\u00e1t tri\u1ec3n.<\/h3>\n

Nvidia \u0111\u00e3 tr\u1edf th\u00e0nh h\u00e3ng s\u1ea3n xu\u1ea5t chip \u0111\u1ed3 ho\u1ea1 \u0111\u1ea7u ti\u00ean \u0111\u1ea1t c\u1ed9t m\u1ed1c v\u1ed1n ho\u00e1 2,2 ng\u00e0n t\u1ef7 USD. C\u00e1c GPU c\u1ee7a c\u00f4ng ty n\u00e0y \u0111ang gi\u1eef vai tr\u00f2 quan tr\u1ecdng trong s\u1ef1 b\u00f9ng n\u1ed5 c\u1ee7a nh\u1eefng m\u00f4 h\u00ecnh tr\u00ed tu\u1ec7 nh\u00e2n t\u1ea1o (AI) t\u1ed5ng h\u1ee3p. B\u00ean c\u1ea1nh ph\u1ea7n c\u1ee9ng, s\u1ee9c m\u1ea1nh c\u1ee7a c\u00f4ng ty n\u00e0y c\u00f2n \u0111\u1ebfn t\u1eeb n\u1ec1n t\u1ea3ng ph\u1ea7n m\u1ec1m l\u1eadp tr\u00ecnh CUDA, hi\u1ec7n \u0111ang \u0111\u01b0\u1ee3c s\u1eed d\u1ee5ng b\u1edfi h\u01a1n 3 tri\u1ec7u nh\u00e0 l\u1eadp tr\u00ecnh to\u00e0n c\u1ea7u \u0111\u1ec3 x\u00e2y d\u1ef1ng AI v\u00e0 c\u00e1c \u1ee9ng d\u1ee5ng kh\u00e1c.<\/p>\n

\u201cCh\u00fang t\u00f4i \u0111ang cho nh\u1eefng nh\u00e0 l\u1eadp tr\u00ecnh th\u1ea5y c\u00e1ch th\u1ee9c tho\u00e1t kh\u1ecfi n\u1ec1n t\u1ea3ng c\u1ee7a Nvidia\u201d, Vinesh Sukumar, gi\u00e1m \u0111\u1ed1c b\u1ed9 ph\u1eadn AI v\u00e0 m\u00e1y h\u1ecdc c\u1ee7a Qualcomm, cho hay.<\/p>\n

\"|\"
Nvidia \u0111ang chi ph\u1ed1i l\u0129nh v\u1ef1c AI v\u1edbi c\u1ea3 ph\u1ea7n c\u1ee9ng (chip \u0111\u1ed3 ho\u1ea1) v\u00e0 ph\u1ea7n m\u1ec1m (n\u1ec1n t\u1ea3ng l\u1eadp tr\u00ecnh CUDA). \u1ea2nh: AIM<\/em><\/i><\/figcaption><\/figure>\n

UXL Foundation, li\u00ean minh c\u00f4ng ngh\u1ec7 ra m\u1eaft v\u00e0o th\u00e1ng 9\/2023, \u0111\u00e3 l\u00ean k\u1ebf ho\u1ea1ch x\u00e2y d\u1ef1ng n\u1ec1n t\u1ea3ng l\u1eadp tr\u00ecnh v\u00e0 b\u1ed9 c\u00f4ng c\u1ee5 m\u1edbi, d\u1ef1a tr\u00ean OneAPI – c\u00f4ng ngh\u1ec7 do Intel ph\u00e1t tri\u1ec3n, h\u01b0\u1edbng \u0111\u1ebfn l\u1eadp tr\u00ecnh nhi\u1ec1u lo\u1ea1i chip AI. D\u1ef1 \u00e1n ngu\u1ed3n m\u1edf n\u00e0y h\u01b0\u1edbng \u0111\u1ebfn \u0111\u01b0a m\u00e3 m\u00e1y t\u00ednh c\u00f3 th\u1ec3 ch\u1ea1y tr\u00ean thi\u1ebft b\u1ecb c\u00f3 chip v\u00e0 ph\u1ea7n c\u1ee9ng b\u1ea5t k\u1ef3.<\/p>\n

\u201cB\u00e0i to\u00e1n \u0111\u1eb7t ra l\u00e0 l\u00e0m sao c\u00f3 th\u1ec3 t\u1ea1o ra m\u1ed9t h\u1ec7 sinh th\u00e1i m\u1edf trong l\u0129nh v\u1ef1c m\u00e1y h\u1ecdc, c\u00f3 th\u1ec3 th\u00fac \u0111\u1ea9y hi\u1ec7u su\u1ea5t v\u00e0 ph\u00f9 h\u1ee3p v\u1edbi \u0111a d\u1ea1ng ph\u1ea7n c\u1ee9ng\u201d, Bill Hugo, gi\u00e1m \u0111\u1ed1c \u0111\u01a1n v\u1ecb \u0111i\u1ec7n to\u00e1n hi\u1ec7u n\u0103ng cao c\u1ee7a Google. G\u00e3 kh\u1ed5ng l\u1ed3 t\u00ecm ki\u1ebfm l\u00e0 m\u1ed9t th\u00e0nh vi\u00ean s\u00e1ng l\u1eadp UXL v\u00e0 h\u1ed7 tr\u1ee3 \u0111\u1ecbnh h\u01b0\u1edbng k\u1ef9 thu\u1eadt cho d\u1ef1 \u00e1n.<\/p>\n

Reuters<\/em> cho bi\u1ebft, theo l\u1ed9 tr\u00ecnh ph\u00e1t tri\u1ec3n, UXL s\u1ebd c\u00f4ng b\u1ed1 nh\u1eefng th\u00f4ng s\u1ed1 k\u1ef9 thu\u1eadt c\u01a1 b\u1ea3n trong n\u1eeda \u0111\u1ea7u n\u0103m nay v\u00e0 s\u1ea3n ph\u1ea9m cu\u1ed1i c\u00f9ng c\u00f3 th\u1ec3 \u0111\u01b0\u1ee3c ho\u00e0n thi\u1ec7n v\u00e0o n\u1eeda cu\u1ed1i n\u0103m.<\/p>\n

B\u00ean c\u1ea1nh nh\u1eefng c\u00f4ng ty s\u00e1ng l\u1eadp ban \u0111\u1ea7u, nh\u00f3m n\u00e0y c\u0169ng \u0111ang k\u00eau g\u1ecdi Amazon, Microsoft v\u00e0 nh\u1eefng nh\u00e0 s\u1ea3n xu\u1ea5t chip kh\u00e1c tham gia “li\u00ean minh”. K\u1ec3 t\u1eeb khi th\u00e0nh l\u1eadp v\u00e0o th\u00e1ng 9\/2023, UXL \u0111\u00e3 nh\u1eadn \u0111\u01b0\u1ee3c s\u1ef1 \u0111\u00f3ng g\u00f3p k\u1ef9 thu\u1eadt t\u1eeb c\u00e1c b\u00ean th\u1ee9 ba, b\u00ean c\u1ea1nh nh\u1eefng th\u00e0nh vi\u00ean t\u1ed5 ch\u1ee9c, v\u00e0 nh\u1eefng ng\u01b0\u1eddi quan t\u00e2m \u0111\u1ebfn s\u1eed d\u1ee5ng c\u00f4ng ngh\u1ec7 m\u1edf. V\u1ec1 l\u00e2u d\u00e0i, n\u1ec1n t\u1ea3ng c\u1ee7a UXL c\u0169ng s\u1ebd h\u1ed7 tr\u1ee3 ph\u1ea7n c\u1ee9ng v\u00e0 m\u00e3 l\u1eadp tr\u00ecnh c\u1ee7a Nvidia.<\/p>\n

D\u1ef1 \u00e1n c\u1ee7a UXL Foundation l\u00e0 m\u1ed9t trong nhi\u1ec1u n\u1ed7 l\u1ef1c nh\u1eb1m lo\u1ea1i b\u1ecf s\u1ef1 th\u1ed1ng tr\u1ecb c\u1ee7a Nvidia \u0111\u1ed1i v\u1edbi ph\u1ea7n m\u1ec1m l\u1eadp tr\u00ecnh h\u1ed7 tr\u1ee3 AI. Theo h\u00e3ng th\u1ed1ng k\u00ea d\u1eef li\u1ec7u PitchBook, c\u00e1c nh\u00e0 \u0111\u1ea7u t\u01b0 m\u1ea1o hi\u1ec3m v\u00e0 doanh nghi\u1ec7p \u0111\u00e3 r\u00f3t h\u01a1n 4 t\u1ef7 USD cho 93 d\u1ef1 \u00e1n ri\u00eang bi\u1ec7t.<\/p>\n

M\u1ed1i quan t\u00e2m \u0111\u1ebfn vi\u1ec7c \u0111\u00e1nh b\u1ea1i Nvidia th\u00f4ng qua m\u1eaft x\u00edch ph\u1ea7n m\u1ec1m ng\u00e0y c\u00e0ng nh\u1eadn \u0111\u01b0\u1ee3c s\u1ef1 ch\u00fa \u00fd k\u1ec3 t\u1eeb n\u0103m ngo\u00e1i. C\u00f4ng ty \u0111\u1ed3 ho\u1ea1 n\u00e0y \u0111\u00e3 ph\u1ee7 b\u00f3ng qu\u00e1 l\u1edbn khi\u1ebfn r\u1ea5t \u00edt c\u00f4ng ty kh\u1edfi nghi\u1ec7p c\u00f3 th\u1ec3 t\u1ea1o ra \u0111\u1ed9t ph\u00e1, khi CUDA tr\u00ean l\u00fd thuy\u1ebft l\u00e0 m\u1ed9t ph\u1ea7n m\u1ec1m h\u1ea5p d\u1eabn, c\u00f3 \u0111\u1ea7y \u0111\u1ee7 t\u00ednh n\u0103ng v\u00e0 li\u00ean t\u1ee5c \u0111\u01b0\u1ee3c ph\u00e1t tri\u1ec3n b\u1edfi Nvidia v\u00e0 c\u1ed9ng \u0111\u1ed3ng l\u1eadp tr\u00ecnh.<\/p>\n

\u201c\u0110i\u1ec1u quan tr\u1ecdng l\u00e0 m\u1ecdi ng\u01b0\u1eddi \u0111\u00e3 s\u1eed d\u1ee5ng CUDA trong h\u01a1n 15 n\u0103m qua v\u00e0 s\u1eed d\u1ee5ng n\u1ec1n t\u1ea3ng n\u00e0y \u0111\u1ec3 x\u00e2y d\u1ef1ng c\u00e1c m\u00e3 l\u1eadp tr\u00ecnh kh\u00e1c\u201d, Jay Goldberg, CEO c\u00f4ng ty t\u01b0 v\u1ea5n t\u00e0i ch\u00ednh v\u00e0 chi\u1ebfn l\u1ef1c D2D Advisory nh\u1eadn x\u00e9t.<\/p>\n

Th\u1ebf Vinh_ Vietnamnet<\/em><\/p>","protected":false},"excerpt":{"rendered":"

M\u1ed9t li\u00ean minh c\u00f4ng ngh\u1ec7 g\u1ed3m Intel, Google v\u00e0 Qualcomm \u0111ang t\u00ecm c\u00e1ch ph\u00e1t tri\u1ec3n n\u1ec1n t\u1ea3ng ph\u1ea7n m\u1ec1m m\u1edbi gi\u00fap c\u00e1c nh\u00e0 l\u1eadp tr\u00ecnh tho\u00e1t kh\u1ecfi ph\u1ee5 thu\u1ed9c v\u00e0o n\u1ec1n t\u1ea3ng CUDA do Nvidia ph\u00e1t tri\u1ec3n. Nvidia \u0111\u00e3 tr\u1edf th\u00e0nh h\u00e3ng s\u1ea3n xu\u1ea5t chip \u0111\u1ed3 ho\u1ea1 \u0111\u1ea7u ti\u00ean \u0111\u1ea1t c\u1ed9t m\u1ed1c v\u1ed1n ho\u00e1 2,2 […]<\/p>\n","protected":false},"author":1,"featured_media":22325,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"wpai_generated_summary":"","footnotes":""},"categories":[174],"tags":[182],"class_list":["post-22208","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-cong-nghe","tag-cong-nghe"],"acf":[],"_links":{"self":[{"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/posts\/22208","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/comments?post=22208"}],"version-history":[{"count":1,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/posts\/22208\/revisions"}],"predecessor-version":[{"id":22927,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/posts\/22208\/revisions\/22927"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/media\/22325"}],"wp:attachment":[{"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/media?parent=22208"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/categories?post=22208"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/tags?post=22208"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}