{"id":173047,"date":"2026-01-30T16:42:15","date_gmt":"2026-01-30T16:42:15","guid":{"rendered":"https:\/\/kamucalisani.net\/?p=173047"},"modified":"2026-01-30T16:42:15","modified_gmt":"2026-01-30T16:42:15","slug":"nvidia-vera-cpu-ile-intel-xeon-ve-amd-epyc-islemcilere-meydan-okuyor","status":"publish","type":"post","link":"https:\/\/kamucalisani.net\/index.php\/2026\/01\/30\/nvidia-vera-cpu-ile-intel-xeon-ve-amd-epyc-islemcilere-meydan-okuyor\/","title":{"rendered":"Nvidia, Vera CPU ile Intel Xeon ve AMD EPYC i\u015flemcilere meydan okuyor"},"content":{"rendered":"

<\/span> Nvidia CEO\u2019su Jensen Huang, \u015firketin yeni Vera CPU<\/strong> i\u015flemcisinin veri merkezlerinde AMD EPYC<\/strong> ve Intel Xeon<\/strong> ailesine do\u011frudan rakip olaca\u011f\u0131n\u0131 duyurdu. Arm tabanl\u0131 olarak geli\u015ftirilen Vera, Nvidia\u2019n\u0131n tarihinde ilk kez GPU\u2019lardan ba\u011f\u0131ms\u0131z \u015fekilde, tek ba\u015f\u0131na tam bir bili\u015fim altyap\u0131s\u0131n\u0131 \u00e7al\u0131\u015ft\u0131rabilecek bir i\u015flemci olarak konumland\u0131r\u0131l\u0131yor. <\/figure>\n

Huang, Vera\u2019y\u0131 \u201cdevrim niteli\u011finde<\/strong>\u201d olarak tan\u0131mlarken baz\u0131 erken d\u00f6nem i\u015f ortaklar\u0131n\u0131n resmi duyurular yap\u0131lmadan \u00f6nce bile da\u011f\u0131t\u0131m planlar\u0131 \u00fczerinde \u00e7al\u0131\u015ft\u0131\u011f\u0131n\u0131 ifade etti. Nvidia, bu \u00fcr\u00fcnle birlikte yapay zeka ve y\u00fcksek performansl\u0131 hesaplama alan\u0131nda u\u00e7tan uca donan\u0131m sa\u011flay\u0131c\u0131s\u0131 olma hedefini net bi\u00e7imde ortaya koyuyor.<\/p>\n

Arm tabanl\u0131 yeni rakip<\/b><\/p>\n

Vera CPU, Nvidia\u2019n\u0131n \u00f6zel tasarlad\u0131\u011f\u0131 88 Armv9.2 \u00e7ekirde\u011fi<\/strong> \u00fczerine kurulu. Her \u00e7ekirde\u011fin iki i\u015f par\u00e7ac\u0131\u011f\u0131 \u00e7al\u0131\u015ft\u0131rabilmesi sayesinde i\u015flemci toplamda 176 i\u015f par\u00e7ac\u0131\u011f\u0131<\/strong> sunuyor. Bu yap\u0131, geleneksel olarak x86 mimarisinin hakim oldu\u011fu veri merkezi pazar\u0131nda Arm tabanl\u0131 g\u00fc\u00e7l\u00fc bir alternatif olu\u015fturmay\u0131 ama\u00e7l\u0131yor. Nvidia, Arm mimarisi sayesinde performans ve g\u00fc\u00e7 verimlili\u011fi dengesini daha esnek bi\u00e7imde y\u00f6netmeyi hedefliyor.<\/p>\n

<\/span> \u0130\u015flemci, 1,5 TB LPDDR5X<\/strong> bellek ve 1,2 TB\/s bant geni\u015fli\u011fi<\/strong> ile bellek yo\u011fun i\u015f y\u00fcklerine odaklan\u0131yor. Vera\u2019n\u0131n mimarisindeki en belirleyici unsurlardan birisi de ikinci nesil Scalable Coherency Fabric yap\u0131s\u0131. Bu y\u00fcksek h\u0131zl\u0131 ara ba\u011flant\u0131 sistemi, 88 \u00e7ekirde\u011fin tamam\u0131n\u0131 tek par\u00e7a (monolitik) bir yonga \u00fczerinde birbirine ba\u011fl\u0131yor. Ortaya \u00e7\u0131kan 3,4 terabayt\/saniye \u00e7ift y\u00f6nl\u00fc bant geni\u015fli\u011fi, \u00e7ekirdekler aras\u0131 veri aktar\u0131m\u0131n\u0131 d\u00fc\u015f\u00fck gecikmeyle m\u00fcmk\u00fcn k\u0131l\u0131yor. <\/figure>\n

Ayr\u0131ca Vera, Nvidia\u2019n\u0131n ikinci nesil NVLink Chip-to-Chip<\/strong> teknolojisi \u00fczerinden GPU\u2019larla do\u011frudan ve y\u00fcksek h\u0131zl\u0131 ba\u011flant\u0131 kurabiliyor. Bu ba\u011flant\u0131, yakla\u015fan Rubin GPU gibi harici bile\u015fenlere 1,8 terabayt\/saniyeye kadar tutarl\u0131 bant geni\u015fli\u011fi sunuyor.<\/p>\n

Vera\u2019n\u0131n \u00e7ekirdekleri, yapay zeka ve vekt\u00f6r i\u015flemleri i\u00e7in ek donan\u0131m yeteneklerine de sahip. Bu yakla\u015f\u0131m, baz\u0131 yapay zeka \u00e7\u0131kar\u0131m<\/strong> i\u015flemlerinin do\u011frudan CPU \u00fczerinde yap\u0131lmas\u0131na olanak tan\u0131yarak gecikme ve enerji maliyetlerini d\u00fc\u015f\u00fcrmeyi ama\u00e7l\u0131yor.<\/p>\n\n

Kaynak\u00a0 :\u00a0https:\/\/www.donanimhaber.com\/nvidia-vera-cpu-ile-intel-xeon-ve-amd-epyc-e-meydan-okuyor–201499<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

Nvidia CEO\u2019su Jensen Huang, \u015firketin yeni Vera CPU i\u015flemcisinin veri merkezlerinde AMD EPYC ve Intel Xeon ailesine do\u011frudan rakip olaca\u011f\u0131n\u0131 duyurdu. Arm tabanl\u0131 olarak geli\u015ftirilen Vera, Nvidia\u2019n\u0131n tarihinde ilk kez GPU\u2019lardan ba\u011f\u0131ms\u0131z \u015fekilde …<\/p>\n","protected":false},"author":1,"featured_media":173048,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[8],"tags":[501,211,164,9056],"class_list":["post-173047","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-teknoloji","tag-arm","tag-is","tag-nvidia","tag-vera"],"_links":{"self":[{"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/posts\/173047","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/comments?post=173047"}],"version-history":[{"count":1,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/posts\/173047\/revisions"}],"predecessor-version":[{"id":173051,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/posts\/173047\/revisions\/173051"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/media\/173048"}],"wp:attachment":[{"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/media?parent=173047"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/categories?post=173047"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/kamucalisani.net\/index.php\/wp-json\/wp\/v2\/tags?post=173047"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}