{"title":"Rack Servers","description":"\u003cp\u003e1U and 2U rack machines — HPE ProLiant for general compute, ASRock Rack barebones you populate yourself. Prices ex-VAT, EU stock.\u003c\/p\u003e","products":[{"product_id":"hpe-proliant-dl385","title":"HPE ProLiant DL385","description":"\u003cp\u003e \u003c\/p\u003e\n\u003cdiv class=\"product-description\"\u003e\n\u003ch1\u003eSerwer HPE ProLiant DL385 Generation 10 CTO z GPU NVIDIA A100\u003c\/h1\u003e\n\u003cdiv class=\"product-overview\"\u003e\n\u003cp\u003eSerwer wyposażono w dwa 32-rdzeniowe procesory AMD EPYC 7502 (64 rdzenie, 128 wątków) oraz 512 GB pamięci HPE DDR4. Pamięć masową stanowią 2x dyski 480GB M.2 NVMe i 2x dyski Enterprise SAS SSD 1.6TB. Serwer obsługuje 8-pozycyjny backplane 2.5\" SFF i jest wyposażony w kontroler RAID HPE Smart Array S100i.\u003c\/p\u003e\n\u003cp\u003eW zestawie znajdują się 2x akceleratory GPU NVIDIA A100 80GB, co czyni go idealnym do zadań AI i uczenia maszynowego. Serwer oferuje cztery zintegrowane karty sieciowe GB oraz dwa porty 10GB SFP+. Redundancję zapewniają 2x zasilacze 800W, a zarządzanie upraszcza iLO5.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"specifications\"\u003e\n\u003ch2\u003eSpecyfikacja techniczna\u003c\/h2\u003e\n\u003ctable class=\"specs-table\"\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003cth\u003eProcesory\u003c\/th\u003e\n\u003ctd\u003eDwa 32-rdzeniowe AMD EPYC 7502 (2.5 GHz, Max Turbo 3.4 GHz)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eRdzenie \/ wątki\u003c\/th\u003e\n\u003ctd\u003e64 rdzenie \/ 128 procesorów logicznych\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003ePamięć\u003c\/th\u003e\n\u003ctd\u003e512 GB (8x 64GB) HPE DDR4\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003ePamięć główna\u003c\/th\u003e\n\u003ctd\u003e2x 480GB M.2 NVMe\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003ePamięć dodatkowa\u003c\/th\u003e\n\u003ctd\u003e2x Enterprise 2.5\" SAS SSD 1.6 TB\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eBackplane\u003c\/th\u003e\n\u003ctd\u003e1x 8Bay 2.5\" SFF (8SFF)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eRAID\u003c\/th\u003e\n\u003ctd\u003eHPE Smart Array S100i SR Gen10 SW RAID + kontroler RAID P208i\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eUkłady graficzne (GPU)\u003c\/th\u003e\n\u003ctd\u003e2x akcelerator GPU NVIDIA A100 80 GB SXM4 PCIe\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eZasilanie\u003c\/th\u003e\n\u003ctd\u003e2x redundantny zasilacz 800W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eŁączność sieciowa\u003c\/th\u003e\n\u003ctd\u003eCztery zintegrowane karty GB NIC + dwuportowy 10GB SFP+\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eSzyny rack\u003c\/th\u003e\n\u003ctd\u003eW zestawie\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eZarządzanie\u003c\/th\u003e\n\u003ctd\u003eiLO5 Standard\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eStan urządzenia\u003c\/th\u003e\n\u003ctd\u003eOdnowione\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003eGwarancja\u003c\/th\u003e\n\u003ctd\u003e24 miesiące RTB (Return To Base)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth\u003ePrzetestowano z\u003c\/th\u003e\n\u003ctd\u003eWindows Server 2022\/2019, VMware ESXi 7.0, Windows Server 2016\/2019 HPC, Ubuntu, CentOS 7 64-bit, Proxmox VE\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"key-features\"\u003e\n\u003ch2\u003eKluczowe cechy i korzyści\u003c\/h2\u003e\n\u003cdiv class=\"feature-section\"\u003e\n\u003ch3\u003e1. Potencjał wydajnościowy dzięki GPU NVIDIA A100\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eGPU NVIDIA A100:\u003c\/strong\u003e Zaprojektowany do najbardziej wymagających zadań obliczeniowych, takich jak uczenie maszynowe, uczenie głębokie i przetwarzanie big data. Dzięki 80GB pamięci na akcelerator zapewniają wysoką wydajność i pojemność przy równoległym przetwarzaniu danych.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eOptymalizacja pod AI i ML:\u003c\/strong\u003e Serwer sprawnie obsługuje zadania AI i ML, co ma kluczowe znaczenie dla organizacji pracujących na rozległych zbiorach danych wymagających szybkich obliczeń.\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"feature-section\"\u003e\n\u003ch3\u003e2. Wysoka skalowalność i elastyczność\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eWsparcie rozbudowy:\u003c\/strong\u003e Obsługa szerokiego zakresu konfiguracji, w tym rozbudowy pamięci, pamięci masowej i pozostałych komponentów\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMożliwość rozbudowy:\u003c\/strong\u003e Gotowość na przyszłe rozszerzenia o dodatkowe karty graficzne i komponenty\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"feature-section\"\u003e\n\u003ch3\u003e3. Bezpieczeństwo i niezawodność\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eWbudowane zabezpieczenia:\u003c\/strong\u003e Mechanizm Silicon Root of Trust chroniący przed atakami na poziomie sprzętu i firmware\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eSystemy redundantne:\u003c\/strong\u003e Redundantne zasilacze i komponenty klasy enterprise\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"use-cases\"\u003e\n\u003ch2\u003eIdealne zastosowania\u003c\/h2\u003e\n\u003cdiv class=\"use-case\"\u003e\n\u003ch3\u003eAI i uczenie maszynowe (MLaaS)\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003eTrening i inferencja głębokich sieci neuronowych\u003c\/li\u003e\n\u003cli\u003ePrzetwarzanie języka naturalnego\u003c\/li\u003e\n\u003cli\u003eZastosowania w wizji komputerowej\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"revenue-potential\"\u003eSzacowany przychód miesięczny: $2,000 - $10,000\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"use-case\"\u003e\n\u003ch3\u003eBig data i analityka\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003ePrzetwarzanie i analiza dużych zbiorów danych\u003c\/li\u003e\n\u003cli\u003eAnalityka danych w czasie rzeczywistym\u003c\/li\u003e\n\u003cli\u003eZastosowania business intelligence\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"revenue-potential\"\u003eSzacowany przychód miesięczny: $3,000 - $15,000\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"use-case\"\u003e\n\u003ch3\u003eWirtualizacja i chmura obliczeniowa\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003eUruchamianie wielu maszyn wirtualnych\u003c\/li\u003e\n\u003cli\u003eUsługi infrastruktury chmurowej\u003c\/li\u003e\n\u003cli\u003eOrkiestracja kontenerów\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"revenue-potential\"\u003eSzacowany przychód miesięczny: $2,000 - $10,000\u003c\/p\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"use-case\"\u003e\n\u003ch3\u003eAplikacje korporacyjne\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003eWysokowydajne bazy danych\u003c\/li\u003e\n\u003cli\u003ePlanowanie zasobów przedsiębiorstwa (ERP)\u003c\/li\u003e\n\u003cli\u003eZarządzanie relacjami z klientami (CRM)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"revenue-potential\"\u003eSzacowany przychód miesięczny: $1,000 - $7,000\u003c\/p\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv class=\"configuration\"\u003e\n\u003ch2\u003eOpcje konfiguracji (CTO — Configure to Order)\u003c\/h2\u003e\n\u003cp\u003eSerwer można dostosować do konkretnych wymagań organizacji, w tym konfiguracji procesorów, pamięci, pamięci masowej i komponentów sieciowych.\u003c\/p\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e","brand":"HP","offers":[{"title":"Default Title","offer_id":49282689007944,"sku":"","price":43555.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/hpe-proliant-dl385.jpg?v=1778668411"},{"product_id":"barebone-server-gen4-2kw","title":"Serwer barebone Bone64c – G4 – 2kW","description":"\u003ch2\u003e\u003cstrong\u003eProfesjonalny serwer barebone gotowy na AI, na platformie Gen4\u003c\/strong\u003e\u003c\/h2\u003e\n\u003cp\u003eSerwer barebone klasy korporacyjnej zoptymalizowany pod obciążenia AI\/ML\u003c\/p\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003ePrzegląd\u003c\/h3\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003eTITAN-64G4 Base zapewnia wydajność klasy korporacyjnej w elastycznej obudowie 4U. Napędzany procesorem AMD EPYC™ \u003cspan\u003e7763 \u003c\/span\u003ei wyposażony w obsługę PCIe Gen4, serwer ten stanowi doskonałą podstawę dla rozwoju AI, uczenia maszynowego oraz zastosowań obliczeniowych o wysokiej wydajności.\u003c\/p\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eKluczowe cechy\u003c\/h3\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003cstrong\u003eArchitektura gotowa na przyszłość\u003c\/strong\u003e: obsługa PCIe Gen4 dla maksymalnej wydajności GPU\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e\n\u003cstrong\u003ePlatforma klasy korporacyjnej\u003c\/strong\u003e: komponenty klasy serwerowej do pracy 24\/7\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003e\n\u003cstrong\u003eElastyczna obsługa GPU\u003c\/strong\u003e: gotowość na karty konsumenckie i serwerowe\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003e\n\u003cstrong\u003eZdalne zarządzanie\u003c\/strong\u003e: pełna obsługa IPMI z myślą o wdrożeniach w centrum danych\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eSpecyfikacja techniczna\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eCPU: AMD EPYC 7763 (64 rdzenie\/128 wątków, 2.5GHz bazowo, 3.4GHz boost)\n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e64 rdzenie \/ 128 wątków\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eTaktowanie bazowe: 2.45 GHz\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eMaks. boost: 3.5 GHz\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003e256MB pamięci podręcznej L3\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003ePłyta główna: ASRock Rack ROMED8-2T\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eObsługa PCIe Gen4\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eObsługa DDR4-3200\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eRAM: 512GB DDR4-2666 ECC REG \n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eKonfiguracja 8x 64GB DIMM\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eObsługa korekcji błędów\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eModuły ECC Registered\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003ePamięć masowa — konfiguracja bazowa:\u003cbr\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e2 x SSD 2tb Fanxiang NVMe M.2 PCIe 4.0\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eOdczyt sekwencyjny: do 7000 MB\/s\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eZapis sekwencyjny: do 6500 MB\/s    \u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"4\"\u003eFormat: rack 4U\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"5\"\u003eSieć: 2 x RJ45 (10GbE)\n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"5\"\u003e Broadcom BCM57416 - wersja BMC \u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"6\"\u003eChłodzenie: 3x przemysłowy wentylator wlotowy 2.7A bez hot-swap\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"7\"\u003eZarządzanie: IPMI\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eZasilacz: pojedynczy LX2000W 2000W Platinum\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eOpcje obsługi GPU:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eDo 3x karta konsumencka (RTX 4090 \/ przyszłe RTX 5090)\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eDo 5x karta serwerowa (L40\/L4\/A100)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cblockquote\u003e\n\u003cp\u003e\u003cem\u003eUwaga: dla kart serwerowych wymagany jest moduł przepływu powietrza do centrum danych (sprzedawany osobno)\u003c\/em\u003e\u003c\/p\u003e\n\u003c\/blockquote\u003e","brand":"Pcpraha","offers":[{"title":"Default Title","offer_id":49320892957000,"sku":"","price":17891.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/GEN04front.png?v=1737994136"},{"product_id":"barebone-server-bone-64g4-4kw","title":"Serwer barebone Bone64c - G4 4kW","description":"\u003ch2\u003e\u003cstrong\u003eProfesjonalny serwer barebone pod AI na platformie Gen4\u003c\/strong\u003e\u003c\/h2\u003e\n\u003cp\u003eSerwer barebone klasy korporacyjnej zoptymalizowany pod obciążenia AI\/ML\u003c\/p\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003ePrzegląd\u003c\/h3\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003eTITAN-64G4 Base zapewnia wydajność klasy korporacyjnej w elastycznej obudowie 4U. Napędzany procesorem AMD EPYC™ \u003cspan\u003e7763 \u003c\/span\u003e i wyposażony w obsługę PCIe Gen4 serwer ten stanowi doskonałą bazę pod rozwój AI, uczenie maszynowe i zastosowania obliczeń wysokiej wydajności.\u003c\/p\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eKluczowe cechy\u003c\/h3\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003cstrong\u003eArchitektura gotowa na przyszłość\u003c\/strong\u003e: obsługa PCIe Gen4 dla maksymalnej wydajności GPU\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e\n\u003cstrong\u003ePlatforma klasy korporacyjnej\u003c\/strong\u003e: podzespoły serwerowe do pracy 24\/7\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003e\n\u003cstrong\u003eElastyczna obsługa GPU\u003c\/strong\u003e: gotowość na karty konsumenckie i serwerowe\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003e\n\u003cstrong\u003eZarządzanie zdalne\u003c\/strong\u003e: pełna obsługa IPMI pod wdrożenia w centrum danych\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eSpecyfikacja techniczna\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eCPU: AMD EPYC 7763 (64 rdzenie\/128 wątków, 2.5GHz bazowo, 3.4GHz boost)\n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e64 rdzenie \/ 128 wątków\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eZegar bazowy: 2.45 GHz\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eMaks. boost: 3.5 GHz\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003e256MB pamięci cache L3\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003ePłyta główna: ASRock Rack ROMED8-2T\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eObsługa PCIe Gen4\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eObsługa DDR4-3200\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eRAM: 512GB DDR4-2666 ECC REG\n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eKonfiguracja 8x 64GB DIMM\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eObsługa korekcji błędów\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eModuły Registered ECC\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003eBazowa konfiguracja dysków:\u003cbr\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e2 x SSD 2tb Fanxiang NVMe M.2 PCIe 4.0\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eOdczyt sekwencyjny: do 7000 MB\/s\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eZapis sekwencyjny: do 6500 MB\/s    \u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"4\"\u003eFormat: montaż w racku 4U\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"5\"\u003eSieć: 2 x RJ45 (10GbE)\n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"5\"\u003e Broadcom BCM57416 - wersja BMC \u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"6\"\u003eChłodzenie: 3x przemysłowe wentylatory nawiewne 2.7A bez hot-swap\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"7\"\u003eZarządzanie: IPMI\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cul\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eZasilacz: podwójny zasilacz LX2000W 2000W Platinum\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eOpcje obsługi GPU:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eDo 4x GPU konsumenckich (RTX 4090\/przyszłe RTX 5090)\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eDo 6x GPU serwerowych (L40\/L4\/A100)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cblockquote\u003e\n\u003cp\u003e\u003cem\u003eUwaga: do kart serwerowych wymagany jest moduł przepływu powietrza do centrum danych (sprzedawany osobno)\u003c\/em\u003e\u003c\/p\u003e\n\u003c\/blockquote\u003e\n","brand":"Pcpraha","offers":[{"title":"Default Title","offer_id":49321796370760,"sku":null,"price":18521.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/GEN04front_02.png?v=1737994136"},{"product_id":"barebone-server-bone-64-g5-2kw","title":"Serwer barebone Bone64c - G5 - 2kW","description":"\u003ch2 class=\"font-600 text-xl font-bold\" level=\"2\"\u003eOpis produktu\u003c\/h2\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003eBone - 64 - G5 to kolejna generacja infrastruktury obliczeniowej AI, wyposażona w najnowszy procesor AMD EPYC™ 9554P oraz technologię PCIe Gen5. Ten serwer barebone klasy korporacyjnej zapewnia wyjątkową wydajność w pracach nad AI, uczeniu maszynowym i obliczeniach wysokiej wydajności.\u003c\/p\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eNajważniejsze cechy\u003c\/h3\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003cstrong\u003eArchitektura nowej generacji\u003c\/strong\u003e: obsługa PCIe Gen5 gotowa na przyszłe karty GPU\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e\n\u003cstrong\u003ePlatforma klasy korporacyjnej\u003c\/strong\u003e: komponenty serwerowe do pracy 24\/7\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003e\n\u003cstrong\u003eElastyczna obsługa GPU\u003c\/strong\u003e: gotowość na karty konsumenckie i serwerowe\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003e\n\u003cstrong\u003eZaawansowane zarządzanie zdalne\u003c\/strong\u003e: pełna obsługa IPMI do wdrożeń w centrum danych\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eSpecyfikacja techniczna\u003c\/h3\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003eProcesor i pamięć\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003cstrong\u003eCPU\u003c\/strong\u003e: AMD EPYC™ 9554P\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e64 rdzenie \/ 128 wątków\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eTaktowanie bazowe: 3.1 GHz\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eTDP: 360W\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003eZaawansowana technologia 3D V-Cache™\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e\n\u003cstrong\u003ePamięć\u003c\/strong\u003e: 512GB DDR5-4800 ECC REG\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eKonfiguracja 8x 64GB DIMM\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eObsługa korekcji błędów\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eModuły Registered ECC\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003ePłyta główna\u003c\/strong\u003e: ASRock GENOAD8X-2T\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eFormat EEB (12.63\" x 13\")\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003ePojedyncze gniazdo SP5 (LGA 6096)\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003e4 sloty PCIe 5.0 \/ CXL2.0 x16\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003e3 sloty PCIe 5.0 x16\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"4\"\u003e1 slot PCIe 5.0 x8\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"5\"\u003eObsługa do 16 portów SATA 6Gb\/s\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"6\"\u003eDwa sloty M.2 (PCIe 5.0 x4)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003ePamięć masowa i sieć\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003cstrong\u003ePamięć masowa\u003c\/strong\u003e:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e2x dysk SSD 2TB Fanxiang NVMe M.2 PCIe 5.0\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eOdczyt sekwencyjny: do 12000 MB\/s\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eZapis sekwencyjny: do 11000 MB\/s\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e\n\u003cstrong\u003eSieć\u003c\/strong\u003e:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003ePodwójne 10GbE RJ45 (Broadcom BCM57416)\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003ePełna obsługa odciążania TCP\/IP\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003eZasilanie i chłodzenie\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003cstrong\u003eZasilacz\u003c\/strong\u003e: LX2000W Platinum\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eMoc 2000W\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eSprawność 80+ Platinum\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e\n\u003cstrong\u003eChłodzenie\u003c\/strong\u003e:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e3x przemysłowy wentylator 2.7A bez hot-swap\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eRozwiązanie chłodzenia klasy korporacyjnej\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003eObsługa GPU\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eKarty konsumenckie:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eDo 3x RTX 4090\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eGotowość na RTX 5090\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eKarty do centrów danych:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eDo 5x NVIDIA L40\/L4\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eZgodność z A100\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003e*Wymaga dodatkowego modułu przepływu powietrza\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003eParametry fizyczne\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eFormat: rack 4U\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eWymiary: 178mm (wys.) x 437mm (szer.) x 648mm (gł.)\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eWaga: 45kg (konfiguracja bazowa)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eDostępne dodatki\u003c\/h3\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eModuł przepływu powietrza dla kart GPU do centrów danych\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eZestaw szyn montażowych (przesuwne\/stałe)\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eWsporniki podtrzymujące GPU\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003eOpcje przedłużonej gwarancji\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eGwarancja\u003c\/h3\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e2-letnia gwarancja ograniczona\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eOpcja przedłużonej gwarancji \u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eWsparcie Advanced RMA\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eWsparcie techniczne\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003eKonsultacje przy instalacji\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cem\u003eUwaga: instalacja kart GPU do centrów danych wymaga dodatkowego modułu przepływu powietrza (sprzedawany osobno). Specyfikacja może ulec zmianie bez powiadomienia.\u003c\/em\u003e\u003c\/p\u003e\n\u003ch2 class=\"font-600 text-xl font-bold\" level=\"2\"\u003eZalecane zastosowania\u003c\/h2\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eRozwój modeli AI\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eDuże modele językowe\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eTrening uczenia maszynowego\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003eObliczenia wysokiej wydajności\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"4\"\u003eAnalityka danych\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"5\"\u003eObliczenia naukowe\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003e\u003cem\u003eUwaga dotycząca wysyłki: ten produkt wymaga specjalnego pakowania i uzgodnienia warunków transportu. Prosimy o kontakt w sprawie wyceny wysyłki i terminów dostawy.\u003c\/em\u003e\u003c\/p\u003e","brand":"Pcpraha","offers":[{"title":"Default Title","offer_id":49322616619336,"sku":null,"price":37013.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/untitled.png?v=1737994278"},{"product_id":"barebone-server-bone-64-g5-4kw","title":"Serwer barebone Bone64c - G5 - 4kW","description":"\u003ch2 class=\"font-600 text-xl font-bold\" level=\"2\"\u003eOpis produktu\u003c\/h2\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003eThe Bone - 64 - G5 to kolejna generacja infrastruktury obliczeniowej AI, oparta na najnowszym procesorze AMD EPYC™ 9554P i technologii PCIe Gen5. Ten serwer barebone klasy korporacyjnej zapewnia wyjątkową wydajność w rozwoju AI, uczeniu maszynowym i zastosowaniach obliczeń wysokiej wydajności.\u003c\/p\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eNajważniejsze cechy\u003c\/h3\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003cstrong\u003eArchitektura nowej generacji\u003c\/strong\u003e: obsługa PCIe Gen5 zapewniająca wydajność GPU na lata\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e\n\u003cstrong\u003ePlatforma klasy korporacyjnej\u003c\/strong\u003e: podzespoły serwerowe do pracy ciągłej 24\/7\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003e\n\u003cstrong\u003eElastyczna obsługa GPU\u003c\/strong\u003e: gotowość na karty konsumenckie i do centrów danych\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003e\n\u003cstrong\u003eZaawansowane zarządzanie zdalne\u003c\/strong\u003e: pełna obsługa IPMI dla wdrożeń w centrum danych\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eSpecyfikacja techniczna\u003c\/h3\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003eProcesor i pamięć\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003cstrong\u003eCPU\u003c\/strong\u003e: AMD EPYC™ 9554P\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e64 rdzenie \/ 128 wątków\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eZegar bazowy: 3.1 GHz\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eTDP: 360W\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003eZaawansowana technologia 3D V-Cache™\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e\n\u003cstrong\u003ePamięć\u003c\/strong\u003e: 512GB DDR5-4800 ECC REG\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eKonfiguracja 8x 64GB DIMM\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eObsługa korekcji błędów\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eModuły Registered ECC\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003ePłyta główna\u003c\/strong\u003e: ASRock GENOAD8X-2T\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eFormat EEB (12.63\" x 13\")\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eJedno gniazdo SP5 (LGA 6096)\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003e4 sloty PCIe 5.0 \/ CXL2.0 x16\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003e3 sloty PCIe 5.0 x16\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"4\"\u003e1 slot PCIe 5.0 x8\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"5\"\u003eObsługa do 16 portów SATA 6Gb\/s\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"6\"\u003eDwa sloty M.2 (PCIe 5.0 x4)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003eMagazyn danych i sieć\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003cstrong\u003eMagazyn danych\u003c\/strong\u003e:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e2x 2TB Fanxiang NVMe M.2 PCIe 5.0 SSD\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eOdczyt sekwencyjny: do 12000 MB\/s\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eZapis sekwencyjny: do 11000 MB\/s\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e\n\u003cstrong\u003eSieć\u003c\/strong\u003e:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eDwa porty 10GbE RJ45 (Broadcom BCM57416)\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003ePełna obsługa odciążania TCP\/IP\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003eZasilanie i chłodzenie\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e\n\u003cstrong\u003eZasilacz\u003c\/strong\u003e: 2 x LX2000W Platinum\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eMoc wyjściowa 2000W\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eSprawność 80+ Platinum\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003e\n\u003cstrong\u003eChłodzenie\u003c\/strong\u003e:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e3x przemysłowe wentylatory 2.7A bez hotswap\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eRozwiązanie chłodzenia klasy korporacyjnej\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003eObsługa GPU\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eKarty konsumenckie:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eDo 4x RTX 4090\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eGotowość na RTX 5090\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eKarty do centrów danych:\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"1\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eDo 6x NVIDIA L40\/L4\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eZgodność do 6x A100 \/ H100\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003e*Wymaga dodatkowego modułu przepływu powietrza\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cstrong\u003eParametry fizyczne\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eFormat: 4U do montażu w racku\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eWymiary: 178mm (wys.) x 437mm (szer.) x 648mm (gł.)\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eWaga: 45kg (konfiguracja bazowa)\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eDostępne dodatki\u003c\/h3\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eModuł przepływu powietrza dla GPU do centrów danych\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eZestaw szyn (przesuwne\/stałe)\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eWsporniki podtrzymujące GPU\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003eOpcje przedłużonej gwarancji\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3 class=\"font-600 text-lg font-bold\" level=\"3\"\u003eGwarancja\u003c\/h3\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003e2-letnia gwarancja ograniczona\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eOpcja przedłużonej gwarancji \u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eWsparcie Advanced RMA\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eWsparcie techniczne\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003eKonsultacja instalacyjna\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp class=\"whitespace-pre-wrap break-words\"\u003e\u003cem\u003eUwaga: instalacja kart GPU do centrów danych wymaga dodatkowego modułu przepływu powietrza (sprzedawanego osobno). Specyfikacja może ulec zmianie bez uprzedzenia.\u003c\/em\u003e\u003c\/p\u003e\n\u003ch2 class=\"font-600 text-xl font-bold\" level=\"2\"\u003eZalecane zastosowania\u003c\/h2\u003e\n\u003cul class=\"-mt-1 [li\u0026gt;\u0026amp;]:mt-2 list-disc space-y-2 pl-8\" depth=\"0\"\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"0\"\u003eRozwój modeli AI\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"1\"\u003eDuże modele językowe\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"2\"\u003eTrenowanie modeli uczenia maszynowego\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"3\"\u003eObliczenia wysokiej wydajności\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"4\"\u003eAnalityka danych\u003c\/li\u003e\n\u003cli class=\"whitespace-normal break-words\" index=\"5\"\u003eObliczenia naukowe\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003e\u003cem\u003eUwaga dotycząca wysyłki: ten produkt wymaga specjalnego pakowania i indywidualnych ustaleń transportowych. Prosimy o kontakt w sprawie wyceny wysyłki i terminów dostawy.\u003c\/em\u003e\u003c\/p\u003e","brand":"Pcpraha","offers":[{"title":"Default Title","offer_id":49322783867208,"sku":null,"price":37643.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/barebone-server-bone-64-g5-4kw-ai-kentino.jpg?v=1786104577"},{"product_id":"enterprise-accelerated-edge-computing-server-al-v-1","title":"Serwer brzegowy z akceleracją dla firm - AL V.1","description":"\u003cp\u003e\u003cstrong\u003eWydajny, uniwersalny i niezawodny serwer do każdego zadania o wysokich wymaganiach\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eSerwer obliczeniowy AL V.1 zaprojektowano dla tych, którzy szukają uniwersalnego, wysoce wydajnego rozwiązania do szerokiego zakresu zadań wykraczających daleko poza AI i uczenie maszynowe. To niezastąpione narzędzie dla firm wymagających wyjątkowej wydajności, niezawodności i elastyczności w wielu zastosowaniach.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eDlaczego AL V.1 to więcej niż serwer AI\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eSerwer AL V.1 powstał nie tylko z myślą o obciążeniach AI i uczenia maszynowego, lecz także o szerokiej gamie zadań obliczeniowych o wysokiej wydajności, takich jak:\u003c\/p\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eAnaliza danych:\u003c\/strong\u003e dzięki zdolności do przetwarzania ogromnych wolumenów danych serwer AL V.1 świetnie sprawdza się w analityce, modelowaniu i prognozowaniu.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePrzetwarzanie Big Data:\u003c\/strong\u003e wydajne rozwiązanie do obsługi Big Data i złożonych baz danych, zapewniające szybkie i efektywne przetwarzanie oraz analizę.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eModelowanie finansowe:\u003c\/strong\u003e idealny dla instytucji finansowych i firm potrzebujących dużej szybkości obliczeń przy złożonych kalkulacjach i budowie modeli.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eRendering i przetwarzanie grafiki:\u003c\/strong\u003e dzięki elastycznym opcjom GPU AL V.1 jest zoptymalizowany pod rendering 3D, postprodukcję wideo oraz inne zadania z zakresu efektów wizualnych i multimediów.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eBadania naukowe:\u003c\/strong\u003e to mocne wsparcie dla firm biotechnologicznych, uczelni i laboratoriów badawczych, umożliwiające złożone obliczenia w genomice, symulacjach procesów chemicznych i innych przedsięwzięciach naukowych.\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003e\u003cstrong\u003eZalety i możliwości:\u003c\/strong\u003e\u003c\/p\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eWysoka niezawodność i gotowość 24\/7:\u003c\/strong\u003e AL V.1 zbudowano do pracy ciągłej, co czyni go idealnym do krytycznych środowisk biznesowych. Komponenty klasy przemysłowej zapewniają stabilność i długą żywotność — warunek konieczny dla każdej firmy polegającej na nieprzerwanej pracy.\u003c\/p\u003e\n\u003c\/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eKonfigurowalny i skalowalny:\u003c\/strong\u003e AL V.1 można dopasować do konkretnych potrzeb każdej firmy. Obsługuje zarówno GPU konsumenckie, jak i klasy centrum danych, a konfigurowalna konstrukcja pozwala na optymalne dostosowanie do praktycznie każdego zadania.\u003c\/p\u003e\n\u003c\/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eŁatwe zarządzanie:\u003c\/strong\u003e dzięki funkcjom zdalnego zarządzania AL V.1 łatwo zintegrować z centrum danych i obsługiwać z dowolnego miejsca na świecie. To szczególnie korzystne dla firm z rozproszonymi zespołami lub wieloma biurami.\u003c\/p\u003e\n\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003e\u003cstrong\u003eAL V.1 — kompletne rozwiązanie dla biznesu\u003c\/strong\u003e\u003c\/p\u003e\n\u003cp\u003eSerwer obliczeniowy AL V.1 oferuje solidne możliwości firmom, które chcą zwiększyć moc obliczeniową i zbudować platformę pod innowacyjny rozwój. Czy to w laboratorium badawczym, firmie finansowej, czy centrum danych, AL V.1 zapewnia idealne połączenie wydajności i elastyczności, pozwalające sprostać każdemu złożonemu wyzwaniu obliczeniowemu.\u003c\/p\u003e","brand":"Pcpraha","offers":[{"title":"Default Title","offer_id":49341096984904,"sku":null,"price":41930.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/bluerack.png?v=1737994278"},{"product_id":"hpe-proliant-dl320-gen11-xeon-4510-12-core-2x-32-gb-ddr5-8x-sff-2x-1000-w-redundant-1u-rack","title":"HPE ProLiant DL320 Gen11 - Xeon 4510 12-core, 2x 32 GB DDR5, 8x SFF, 2x 1000 W Redundant, 1U Rack","description":"\u003cp\u003eHPE ProLiant DL320 Gen11 - Xeon 4510 12-core, 2x 32 GB DDR5, 8x SFF, 2x 1000 W Redundant, 1U Rack - an enterprise rack server supplied by Kentino from EU distribution stock, ready for virtualisation, storage or as a GPU host.\u003c\/p\u003e\n\u003ch3\u003eSpecification\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eProcessor family:\u003c\/strong\u003e Intel Xeon Silver\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eProcessor model:\u003c\/strong\u003e 4510\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eProcessor frequency:\u003c\/strong\u003e 2.4 GHz\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eInternal memory:\u003c\/strong\u003e 64 GB\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eInternal memory type:\u003c\/strong\u003e DDR5-SDRAM\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePower supply:\u003c\/strong\u003e 1000 W\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eChassis type:\u003c\/strong\u003e Rack (1U)\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMemory layout (slots x size):\u003c\/strong\u003e 2 x 32 GB\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3\u003ePart number\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eManufacturer part number:\u003c\/strong\u003e P85575-425\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eUnit weight:\u003c\/strong\u003e 25.8 kg\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3\u003eBuying from Kentino\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eIn EU distribution stock\u003c\/strong\u003e - no import wait.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePrices are ex-VAT.\u003c\/strong\u003e VAT is a separate line: 0 % reverse charge for intra-EU B2B with a valid VAT ID, 21 % for domestic CZ and B2C.\u003c\/li\u003e\n\u003cli\u003e\u003cstrong\u003eQuantity pricing on request.\u003c\/strong\u003e\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eConfigured to order.\u003c\/strong\u003e We can populate CPU, memory, storage and GPUs before dispatch, with burn-in and a commissioning report.\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003eQuestions or a quote with your VAT ID applied: \u003ca href=\"mailto:info@kentino.com\"\u003einfo@kentino.com\u003c\/a\u003e\u003c\/p\u003e","brand":"HPE","offers":[{"title":"Default Title","offer_id":54504293597512,"sku":"P85575-425","price":8890.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/kentino-p85575-425.jpg?v=1793816162"},{"product_id":"hpe-proliant-dl20-gen11-xeon-6333p-6-core-32-gb-ddr5-4x-sff-800-w-1u-rack","title":"HPE ProLiant DL20 Gen11 - Xeon 6333P 6-core, 32 GB DDR5, 4x SFF, 800 W, 1U Rack","description":"\u003cp\u003eHPE ProLiant DL20 Gen11 - Xeon 6333P 6-core, 32 GB DDR5, 4x SFF, 800 W, 1U Rack - an enterprise rack server supplied by Kentino from EU distribution stock, ready for virtualisation, storage or as a GPU host.\u003c\/p\u003e\n\u003ch3\u003eSpecification\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eProcessor family:\u003c\/strong\u003e Intel® Xeon®\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eProcessor model:\u003c\/strong\u003e 6333P\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eProcessor frequency:\u003c\/strong\u003e 3.1 GHz\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eInternal memory:\u003c\/strong\u003e 32 GB\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eInternal memory type:\u003c\/strong\u003e DDR5-SDRAM\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePower supply:\u003c\/strong\u003e 800 W\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eChassis type:\u003c\/strong\u003e Rack (1U)\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eMemory layout (slots x size):\u003c\/strong\u003e 1 x 32 GB\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3\u003ePart number\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eManufacturer part number:\u003c\/strong\u003e P85574-425\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eUnit weight:\u003c\/strong\u003e 10.9 kg\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3\u003eBuying from Kentino\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eIn EU distribution stock\u003c\/strong\u003e - no import wait.\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePrices are ex-VAT.\u003c\/strong\u003e VAT is a separate line: 0 % reverse charge for intra-EU B2B with a valid VAT ID, 21 % for domestic CZ and B2C.\u003c\/li\u003e\n\u003cli\u003e\u003cstrong\u003eQuantity pricing on request.\u003c\/strong\u003e\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eConfigured to order.\u003c\/strong\u003e We can populate CPU, memory, storage and GPUs before dispatch, with burn-in and a commissioning report.\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003cp\u003eQuestions or a quote with your VAT ID applied: \u003ca href=\"mailto:info@kentino.com\"\u003einfo@kentino.com\u003c\/a\u003e\u003c\/p\u003e","brand":"HPE","offers":[{"title":"Default Title","offer_id":54504293630280,"sku":"P85574-425","price":6090.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/kentino-p85574-425.jpg?v=1793816162"},{"product_id":"asrock-rack-2u4g-genoa-m3-2u-4-gpu-epyc-barebone","title":"ASRock Rack 2U4G-GENOA\/M3 — 2U 4-GPU EPYC Barebone (PCIe 5.0, 1+1 2700 W)","description":"\u003cdiv style=\"background:linear-gradient(135deg,#0d0d0d 0%,#1a1a2e 100%);color:#fff;padding:32px;border-radius:12px;margin-bottom:32px\"\u003e\n\u003cp style=\"font-size:18px;margin:0 0 20px 0;color:#ccc\"\u003eASRock Rack 2U4G-GENOA\/M3\u003c\/p\u003e\n\u003cp style=\"font-size:26px;font-weight:700;margin:0 0 16px 0;line-height:1.3\"\u003eFour full-size GPU in 2U, on one EPYC socket\u003c\/p\u003e\n\u003cdiv style=\"display:flex;gap:16px;flex-wrap:wrap;margin-top:24px\"\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e4\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003etriple-slot PCIe 5.0 x16\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e2U\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003erack height\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e2700 W\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003e1+1 redundant\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003eGen5\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003eend to end\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cp style=\"margin-top:20px;font-size:15px;color:#aaa\"\u003eThe chassis we reach for when a four-card machine has to fit a rack that is already full.\u003c\/p\u003e\n\u003c\/div\u003e\u003cp style=\"font-size:17px;color:#333;margin-bottom:24px\"\u003eA 2U rackmount built around a single SP5 socket that takes four full-length, triple-slot GPU on native PCIe 5.0 x16 — no switch, no lane sharing. Four hot-swap Gen5 NVMe bays, an OCP 3.0 slot for networking and a genuinely redundant 1+1 2700 W Platinum supply. It does in 2U what most of the market needs 4U to do.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eSpecification\u003c\/h2\u003e\u003ctable style=\"width:100%;border-collapse:collapse;margin-bottom:24px;font-size:15px\"\u003e\n\u003cthead\u003e\u003ctr style=\"background:#0d0d0d;color:#fff\"\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eComponent\u003c\/th\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eDetail\u003c\/th\u003e\n\u003c\/tr\u003e\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eForm factor\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e2U rackmount, 850 × 438 × 87.5 mm\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eProcessor\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003eSingle Socket SP5 (LGA 6096). AMD EPYC 9005 (BIOS update required), 9004 incl. 3D V-Cache, and 97x4. Up to 400 W TDP\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eMemory\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e8 DIMM slots, 1DPC. DDR5 RDIMM \/ RDIMM-3DS to 4800 MHz. 128 GB per RDIMM, 256 GB per 3DS module\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eGPU slots\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003eRear: 1 × FHFL triple-slot PCIe 5.0 x16 + 1 × FHHL PCIe 5.0 x16. Internal: 3 × FHFL triple-slot PCIe 5.0 x16. Plus 1 × FHFL PCIe 5.0 x8\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eGPU clearance\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e312.4 × 137 × 62 mm rear, 312.4 × 130 × 62 mm internal\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eStorage\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e4 × hot-swap 2.5\" NVMe PCIe 5.0 x4 + 2 × hot-swap 2.5\" SATA, 6-port backplane. 2 × M.2 PCIe 5.0 x4 internal\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eNetworking\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1 × OCP NIC 3.0 (PCIe 5.0 x16), dedicated IPMI GLAN\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003ePower\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1+1 CRPS, 2700 W at 200–240 V (1200 W at 110 V), 80 PLUS Platinum\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eCooling\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e3 × 80 × 38 mm + 1 × 80 × 80 mm PWM, front-to-back\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eManagement\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003eASPEED AST2600 BMC, IPMI, mini DisplayPort, 2 × USB 3.2 Gen1\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eRead the power budget before you choose cards\u003c\/h2\u003e\u003cp\u003eThe 1+1 supply means one module has to carry the whole machine, so the honest budget is \u003cstrong\u003e2700 W, not 5400 W\u003c\/strong\u003e. That single number decides which GPU this chassis is for.\u003c\/p\u003e\u003ctable style=\"width:100%;border-collapse:collapse;margin-bottom:24px;font-size:15px\"\u003e\n\u003cthead\u003e\u003ctr style=\"background:#0d0d0d;color:#fff\"\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eComponent\u003c\/th\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eDetail\u003c\/th\u003e\n\u003c\/tr\u003e\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003e4 × 300 W class (e.g. RTX PRO 6000 Max-Q)\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e~1200 W of GPU, ~1800 W total. Comfortable, and the configuration this chassis was designed around\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003e2 × 600 W class (Server \/ Workstation Edition)\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e~1200 W of GPU, ~1800 W total. Also comfortable\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003e4 × 600 W class\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e~2400 W of GPU, ~3000 W total. \u003cstrong\u003eOver budget once a supply fails.\u003c\/strong\u003e Do not specify it\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\u003cp\u003eNote also that the three internal x16 positions have no external cutout, so an add-in card fitted there has no reachable I\/O ports. For compute cards that is irrelevant. For anything you need to plug into, use the rear slots.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhere the 8 DIMM slots bite\u003c\/h2\u003e\u003cp\u003eSP5 is a twelve-channel platform and this board brings out eight slots at one DIMM per channel. You therefore run \u003cstrong\u003eeight of twelve channels\u003c\/strong\u003e, whatever you populate, and memory bandwidth lands around two thirds of what the CPU could do. For GPU work, where the cards hold the working set, that trade buys you the 2U height and is usually the right one. For a memory-bound CPU workload it is the wrong chassis and we will tell you so.\u003c\/p\u003e\u003cp\u003eIt also sets the ceiling: 1 TB with 128 GB RDIMMs, 2 TB with 256 GB 3DS. If you are sizing system memory at twice total VRAM — which is what we recommend — four 96 GB cards want 768 GB, and that needs 96 GB or larger modules.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eRack and cooling\u003c\/h2\u003e\u003cp\u003eFront-to-back airflow, rated for continuous 24\/7 operation in a datacentre rack. Plan a dedicated feed sized for the real load: around 2 kW for a four-card Max-Q build. Cold aisle below 24 °C and real airflow through the rack, not just perforated doors — 2U at this power density is unforgiving of a cabinet that merely looks ventilated.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat is included\u003c\/h2\u003e\u003cp\u003e\u003cstrong\u003eWhat is in the box:\u003c\/strong\u003e chassis, backplane, motherboard, the full fan wall and the power supplies listed above, pre-fitted and cabled. It is a complete, working platform less the parts you choose.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eWhat is not:\u003c\/strong\u003e CPU, memory, drives, GPU and any add-in NIC. Those are specified with you and priced separately — that is the point of buying a barebone rather than a finished server.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat Kentino adds\u003c\/h2\u003e\u003cp\u003eEvery barebone we ship is unpacked, assembled, powered up and tested before it leaves Prague: fan wall and PSU redundancy checked, BMC firmware brought up to the current release, IPMI configured and the network details handed over in writing. You get a machine that has already been switched on once by somebody who knows what it should sound like.\u003c\/p\u003e\u003cp\u003eTell us the workload instead of the part numbers and we will specify the CPU, memory and storage for you, build it, burn it in and ship it as a finished server. The same chassis, the same price for the chassis — we simply add the parts and the work.\u003c\/p\u003e\u003cp style=\"font-size:14px;color:#666\"\u003e\u003cem\u003ePrice shown is EUR 5,526 excluding VAT for the barebone as described, \u003cstrong\u003eEXW Prague\u003c\/strong\u003e (Incoterms 2020). Freight, insurance and any destination duties are for the buyer’s account. Typical lead time 10–28 days. Image: ASRock Rack.\u003c\/em\u003e\u003c\/p\u003e","brand":"Kentino \/ ASRock Rack","offers":[{"title":"Default Title","offer_id":54509188579656,"sku":"2U4G-GENOA\/M3","price":5593.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/01-2u4g-genoa-m3-1.jpg?v=1793969136"},{"product_id":"asrock-rack-1u12e-genoa-2l2t-1u-epyc-nvme-barebone","title":"ASRock Rack 1U12E-GENOA\/2L2T — 1U EPYC Barebone, 12× Gen5 NVMe, 24 DIMM","description":"\u003cdiv style=\"background:linear-gradient(135deg,#0d0d0d 0%,#1a1a2e 100%);color:#fff;padding:32px;border-radius:12px;margin-bottom:32px\"\u003e\n\u003cp style=\"font-size:18px;margin:0 0 20px 0;color:#ccc\"\u003eASRock Rack 1U12E-GENOA\/2L2T\u003c\/p\u003e\n\u003cp style=\"font-size:26px;font-weight:700;margin:0 0 16px 0;line-height:1.3\"\u003eTwelve Gen5 NVMe bays and every memory channel, in 1U\u003c\/p\u003e\n\u003cdiv style=\"display:flex;gap:16px;flex-wrap:wrap;margin-top:24px\"\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e12\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003ehot-swap Gen5 NVMe\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e24\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003eDIMM slots\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e1U\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003erack height\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e10GbE\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003eon board\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cp style=\"margin-top:20px;font-size:15px;color:#aaa\"\u003eThe dense storage and compute node — the one ASRock Rack system here that populates all twelve channels.\u003c\/p\u003e\n\u003c\/div\u003e\u003cp style=\"font-size:17px;color:#333;margin-bottom:24px\"\u003eA single-socket SP5 node in 1U with twelve hot-swap PCIe 5.0 NVMe bays and \u003cstrong\u003etwenty-four DIMM slots\u003c\/strong\u003e — the full twelve-channel memory width that most compact EPYC systems give up. Dual 10GbE and dual 1GbE on board plus an OCP 3.0 slot, so a standard build needs no add-in networking at all.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eSpecification\u003c\/h2\u003e\u003ctable style=\"width:100%;border-collapse:collapse;margin-bottom:24px;font-size:15px\"\u003e\n\u003cthead\u003e\u003ctr style=\"background:#0d0d0d;color:#fff\"\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eComponent\u003c\/th\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eDetail\u003c\/th\u003e\n\u003c\/tr\u003e\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eForm factor\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1U rackmount\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eProcessor\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003eSingle Socket SP5 (LGA 6096). AMD EPYC 9005 (BIOS update required), 9004 incl. 3D V-Cache, and 97x4. Up to 300 W TDP\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eMemory\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e24 DIMM slots, 2DPC. DDR5 RDIMM \/ RDIMM-3DS, 4800 MHz at 1DPC. 128 GB per RDIMM\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eStorage\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e12 × hot-swap 2.5\" NVMe PCIe 5.0 x4 \/ SATA bays. 2 × M.2 PCIe 5.0 x4 internal\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eExpansion\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e2 × FHHL PCIe 5.0 x16 (half length), 1 × OCP NIC 3.0 (PCIe 5.0 x16)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eNetworking\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e2 × 10GbE Broadcom BCM57416, 2 × 1GbE Intel i210, dedicated IPMI LAN\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003ePower\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1+1 redundant 1000 W slim, 80 PLUS Platinum\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eManagement\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003eASPEED AST2600 BMC, IPMI\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat it is for, and what it is not\u003c\/h2\u003e\u003cp\u003eTwelve Gen5 NVMe bays behind a 300 W-class EPYC with full memory width is a very good shape for a storage node, a database node, a Ceph or vSAN member, a virtualisation host, or the CPU tier underneath a GPU cluster.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eIt is not a GPU server, and we would rather say so here than after you order.\u003c\/strong\u003e The two x16 slots are \u003cstrong\u003ehalf length\u003c\/strong\u003e. A full-length accelerator — RTX PRO 6000, L40S — does not fit, and no riser changes that. A low-profile card such as the NVIDIA L4 fits comfortably, and for light inference alongside storage that is a sensible pairing. If you want four full-size GPU, the 2U4G-GENOA\/M3 above is the chassis you actually want.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eRack and cooling\u003c\/h2\u003e\u003cp\u003eFront-to-back airflow, 24\/7 rated. 1U fan walls are loud by physics, not by fault — this belongs in a datacentre rack, not in an office. A fully populated node with a 300 W CPU and twelve NVMe drives sits comfortably inside the 1000 W redundant budget.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat is included\u003c\/h2\u003e\u003cp\u003e\u003cstrong\u003eWhat is in the box:\u003c\/strong\u003e chassis, backplane, motherboard, the full fan wall and the power supplies listed above, pre-fitted and cabled. It is a complete, working platform less the parts you choose.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eWhat is not:\u003c\/strong\u003e CPU, memory, drives, GPU and any add-in NIC. Those are specified with you and priced separately — that is the point of buying a barebone rather than a finished server.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat Kentino adds\u003c\/h2\u003e\u003cp\u003eEvery barebone we ship is unpacked, assembled, powered up and tested before it leaves Prague: fan wall and PSU redundancy checked, BMC firmware brought up to the current release, IPMI configured and the network details handed over in writing. You get a machine that has already been switched on once by somebody who knows what it should sound like.\u003c\/p\u003e\u003cp\u003eTell us the workload instead of the part numbers and we will specify the CPU, memory and storage for you, build it, burn it in and ship it as a finished server. The same chassis, the same price for the chassis — we simply add the parts and the work.\u003c\/p\u003e\u003cp style=\"font-size:14px;color:#666\"\u003e\u003cem\u003ePrice shown is EUR 4,070 excluding VAT for the barebone as described, \u003cstrong\u003eEXW Prague\u003c\/strong\u003e (Incoterms 2020). Freight, insurance and any destination duties are for the buyer’s account. Typical lead time 10–28 days. Image: ASRock Rack.\u003c\/em\u003e\u003c\/p\u003e","brand":"Kentino \/ ASRock Rack","offers":[{"title":"Default Title","offer_id":54509189431624,"sku":"1U12E-GENOA\/2L2T","price":4119.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/01-1u12e-genoa-2l2t-1.jpg?v=1793969132"},{"product_id":"asrock-rack-2u12l2s-siena-2u-storage-barebone","title":"ASRock Rack 2U12L2S-SIENA — 2U Storage Barebone, 12× 3.5\" (EPYC 8004\/8005, Titanium)","description":"\u003cdiv style=\"background:linear-gradient(135deg,#0d0d0d 0%,#1a1a2e 100%);color:#fff;padding:32px;border-radius:12px;margin-bottom:32px\"\u003e\n\u003cp style=\"font-size:18px;margin:0 0 20px 0;color:#ccc\"\u003eASRock Rack 2U12L2S-SIENA\u003c\/p\u003e\n\u003cp style=\"font-size:26px;font-weight:700;margin:0 0 16px 0;line-height:1.3\"\u003eTwelve 3.5-inch bays on a low-power EPYC, with a Titanium supply\u003c\/p\u003e\n\u003cdiv style=\"display:flex;gap:16px;flex-wrap:wrap;margin-top:24px\"\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e12\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003ehot-swap 3.5\" bays\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003eTitanium\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003e1+1 800 W\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003eSP6\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003eEPYC Siena\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e2U\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003erack height\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cp style=\"margin-top:20px;font-size:15px;color:#aaa\"\u003eBulk capacity that is cheap to run — the sensible bottom tier under a GPU cluster.\u003c\/p\u003e\n\u003c\/div\u003e\u003cp style=\"font-size:17px;color:#333;margin-bottom:24px\"\u003eA 2U storage platform on AMD’s SP6 socket — the low-power, lower-core-count EPYC built for exactly this job. Twelve hot-swap 3.5-inch bays plus two 2.5-inch, and a 1+1 \u003cstrong\u003e80 PLUS Titanium\u003c\/strong\u003e supply, which is a grade above almost everything else in this class and shows up directly on the power bill of a box that runs permanently.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eSpecification\u003c\/h2\u003e\u003ctable style=\"width:100%;border-collapse:collapse;margin-bottom:24px;font-size:15px\"\u003e\n\u003cthead\u003e\u003ctr style=\"background:#0d0d0d;color:#fff\"\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eComponent\u003c\/th\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eDetail\u003c\/th\u003e\n\u003c\/tr\u003e\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eForm factor\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e2U rackmount\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eProcessor\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003eSingle Socket SP6 (LGA 4844). AMD EPYC 8005 \/ 8004 series (Siena)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eMemory\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e8 DIMM slots. DDR5 RDIMM, 4800 MHz, 96 GB per module — 768 GB maximum\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eStorage\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e12 × hot-swap 3.5\" SATA + 2 × hot-swap 2.5\" SATA (7 mm). 2 × M.2 PCIe 5.0 x4\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eExpansion\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e2 × HHHL PCIe 5.0 x16 (half height, half length)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eNetworking\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e2 × 1GbE Intel i210, dedicated IPMI LAN\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003ePower\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1+1 redundant 800 W CRPS, 80 PLUS \u003cstrong\u003eTitanium\u003c\/strong\u003e\n\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eManagement\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003eASPEED AST2600 BMC, IPMI\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eHonest limits\u003c\/h2\u003e\u003cp\u003eThe drive bays are \u003cstrong\u003eSATA\u003c\/strong\u003e, so this is a capacity tier, not a performance tier — put the fast working set on NVMe elsewhere and use this for what spinning disks and bulk SSDs are good at. The two expansion slots are half height and half length, which is enough for an HBA or a 25\/100GbE NIC and not enough for any GPU worth fitting. Siena tops out well below a 9005-series part on cores and clock; that is the trade that buys the low idle draw.\u003c\/p\u003e\u003cp\u003eUsed for what it is — backup target, media or dataset store, Ceph OSD node, the archive tier under a training cluster — it is one of the cheapest ways to rack real capacity with redundant Titanium power.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eRack and cooling\u003c\/h2\u003e\u003cp\u003eFront-to-back airflow, 24\/7 rated. Twelve 3.5-inch drives are heavy: use proper rails and load the rack from the bottom. Budget around 400–500 W in service for a full set of drives, well inside the 800 W redundant envelope.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat is included\u003c\/h2\u003e\u003cp\u003e\u003cstrong\u003eWhat is in the box:\u003c\/strong\u003e chassis, backplane, motherboard, the full fan wall and the power supplies listed above, pre-fitted and cabled. It is a complete, working platform less the parts you choose.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eWhat is not:\u003c\/strong\u003e CPU, memory, drives, GPU and any add-in NIC. Those are specified with you and priced separately — that is the point of buying a barebone rather than a finished server.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat Kentino adds\u003c\/h2\u003e\u003cp\u003eEvery barebone we ship is unpacked, assembled, powered up and tested before it leaves Prague: fan wall and PSU redundancy checked, BMC firmware brought up to the current release, IPMI configured and the network details handed over in writing. You get a machine that has already been switched on once by somebody who knows what it should sound like.\u003c\/p\u003e\u003cp\u003eTell us the workload instead of the part numbers and we will specify the CPU, memory and storage for you, build it, burn it in and ship it as a finished server. The same chassis, the same price for the chassis — we simply add the parts and the work.\u003c\/p\u003e\u003cp style=\"font-size:14px;color:#666\"\u003e\u003cem\u003ePrice shown is EUR 3,705 excluding VAT for the barebone as described, \u003cstrong\u003eEXW Prague\u003c\/strong\u003e (Incoterms 2020). Freight, insurance and any destination duties are for the buyer’s account. Typical lead time 10–28 days. Image: ASRock Rack.\u003c\/em\u003e\u003c\/p\u003e","brand":"Kentino \/ ASRock Rack","offers":[{"title":"Default Title","offer_id":54509189464392,"sku":"2U12L2S-SIENA","price":3716.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/01-2u12l2s-siena-1.jpg?v=1793969137"},{"product_id":"asrock-rack-2u1g-b650-evac-2u-single-gpu-barebone","title":"ASRock Rack 2U1G-B650\/EVAC — 2U Single-GPU Barebone, Short Depth, 1+1 1200 W","description":"\u003cdiv style=\"background:linear-gradient(135deg,#0d0d0d 0%,#1a1a2e 100%);color:#fff;padding:32px;border-radius:12px;margin-bottom:32px\"\u003e\n\u003cp style=\"font-size:18px;margin:0 0 20px 0;color:#ccc\"\u003eASRock Rack 2U1G-B650\/EVAC\u003c\/p\u003e\n\u003cp style=\"font-size:26px;font-weight:700;margin:0 0 16px 0;line-height:1.3\"\u003eOne full-size GPU, redundant power, 478 mm deep\u003c\/p\u003e\n\u003cdiv style=\"display:flex;gap:16px;flex-wrap:wrap;margin-top:24px\"\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e1\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003eFHFL dual-slot GPU\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e1200 W\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003e1+1 redundant\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e478 mm\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003echassis depth\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003eAM5\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003eEPYC 4005 \/ Ryzen\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cp style=\"margin-top:20px;font-size:15px;color:#aaa\"\u003eThe entry rack GPU node — and the one that fits a shallow cabinet.\u003c\/p\u003e\n\u003c\/div\u003e\u003cp style=\"font-size:17px;color:#333;margin-bottom:24px\"\u003eA 2U node on AMD’s AM5 socket that takes one full-length, dual-slot GPU and backs it with a genuinely redundant 1+1 1200 W Platinum supply. At \u003cstrong\u003e478 mm deep\u003c\/strong\u003e it goes into shallow and wall-mount cabinets that will not accept a normal 850 mm server — which is often the actual constraint in a comms room or at an edge site.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eSpecification\u003c\/h2\u003e\u003ctable style=\"width:100%;border-collapse:collapse;margin-bottom:24px;font-size:15px\"\u003e\n\u003cthead\u003e\u003ctr style=\"background:#0d0d0d;color:#fff\"\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eComponent\u003c\/th\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eDetail\u003c\/th\u003e\n\u003c\/tr\u003e\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eForm factor\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e2U rackmount, 477.7 × 430.4 × 88 mm\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eProcessor\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003eSingle Socket AM5 (LGA 1718). AMD EPYC 4005 \/ 4004, or Ryzen 9000 \/ 8000 \/ 7000. Up to 170 W TDP. Chipset B650E\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eMemory\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e4 DIMM slots, 2DPC. DDR5 ECC or non-ECC UDIMM, 48 GB per module — 192 GB maximum. 5600 MHz at 1DPC, 3600 MHz at 2DPC\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eGPU slot\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1 × FHFL dual-slot PCIe 5.0 x16 on riser, plus 1 × FHHL PCIe 4.0 x4\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eStorage\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e2 × M.2, 1 × optional fixed 2.5\" bay\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eNetworking\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e2 × 1GbE Intel i210, dedicated IPMI LAN\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003ePower\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1+1 hot-swap CRPS, 1200 W at 200–240 V (1000 W at 110 V), 80 PLUS Platinum\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eManagement\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003eASPEED BMC, IPMI\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat it will and will not carry\u003c\/h2\u003e\u003cp\u003e1200 W redundant against a 170 W CPU leaves real room for the card: a 600 W class GPU fits the budget with margin, and a 300 W Max-Q card leaves the machine barely working. One card is the limit — there is one x16 slot and no second position.\u003c\/p\u003e\u003cp\u003eThe constraint to watch is \u003cstrong\u003ememory\u003c\/strong\u003e. AM5 takes UDIMM, capped at 48 GB per module and 192 GB in the chassis. If you size system memory at twice total VRAM, this comfortably carries one 96 GB card and nothing larger. Storage is thin too — two M.2 and one optional 2.5-inch bay — so plan on network storage or keep the dataset on the M.2 pair.\u003c\/p\u003e\u003cp\u003eWithin those limits it is the cheapest way to put a serious GPU in a rack with redundant power: a single-model inference endpoint, a render or transcode node, a CI runner with a real card in it, or a branch box that has to live in a shallow cabinet.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eRack and cooling\u003c\/h2\u003e\u003cp\u003eFront-to-back airflow, 24\/7 rated. Check the cabinet depth before anything else — the short chassis is the reason to choose this machine, and it is also why it will look lost in a standard 1000 mm rack.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat is included\u003c\/h2\u003e\u003cp\u003e\u003cstrong\u003eWhat is in the box:\u003c\/strong\u003e chassis, backplane, motherboard, the full fan wall and the power supplies listed above, pre-fitted and cabled. It is a complete, working platform less the parts you choose.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eWhat is not:\u003c\/strong\u003e CPU, memory, drives, GPU and any add-in NIC. Those are specified with you and priced separately — that is the point of buying a barebone rather than a finished server.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat Kentino adds\u003c\/h2\u003e\u003cp\u003eEvery barebone we ship is unpacked, assembled, powered up and tested before it leaves Prague: fan wall and PSU redundancy checked, BMC firmware brought up to the current release, IPMI configured and the network details handed over in writing. You get a machine that has already been switched on once by somebody who knows what it should sound like.\u003c\/p\u003e\u003cp\u003eTell us the workload instead of the part numbers and we will specify the CPU, memory and storage for you, build it, burn it in and ship it as a finished server. The same chassis, the same price for the chassis — we simply add the parts and the work.\u003c\/p\u003e\u003cp style=\"font-size:14px;color:#666\"\u003e\u003cem\u003ePrice shown is EUR 2,967 excluding VAT for the barebone as described, \u003cstrong\u003eEXW Prague\u003c\/strong\u003e (Incoterms 2020). Freight, insurance and any destination duties are for the buyer’s account. Typical lead time 10–28 days. Image: ASRock Rack.\u003c\/em\u003e\u003c\/p\u003e","brand":"Kentino \/ ASRock Rack","offers":[{"title":"Default Title","offer_id":54509189497160,"sku":"2U1G-B650\/EVAC","price":2975.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/01-2u1g-b650-evac-1.jpg?v=1793969132"},{"product_id":"asrock-rack-1u2n2g-am5-2t-1u-twin-node-barebone","title":"ASRock Rack 1U2N2G-AM5\/2T — 1U Twin-Node Barebone, GPU + 10GbE per Node","description":"\u003cdiv style=\"background:linear-gradient(135deg,#0d0d0d 0%,#1a1a2e 100%);color:#fff;padding:32px;border-radius:12px;margin-bottom:32px\"\u003e\n\u003cp style=\"font-size:18px;margin:0 0 20px 0;color:#ccc\"\u003eASRock Rack 1U2N2G-AM5\/2T\u003c\/p\u003e\n\u003cp style=\"font-size:26px;font-weight:700;margin:0 0 16px 0;line-height:1.3\"\u003eTwo independent servers, each with its own GPU, in one 1U\u003c\/p\u003e\n\u003cdiv style=\"display:flex;gap:16px;flex-wrap:wrap;margin-top:24px\"\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e2\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003eindependent nodes\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e1\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003eGPU per node\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e10GbE\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003e×2 per node\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cdiv style=\"background:rgba(51,79,180,0.12);border:1px solid #334fb4;border-radius:8px;padding:14px 16px;text-align:center;flex:1;min-width:110px\"\u003e\n\u003cdiv style=\"font-size:22px;font-weight:800;color:#334fb4\"\u003e1600 W\u003c\/div\u003e\n\u003cdiv style=\"font-size:11px;color:#888;text-transform:uppercase;letter-spacing:1px\"\u003e1+1 shared\u003c\/div\u003e\n\u003c\/div\u003e\n\u003c\/div\u003e\n\u003cp style=\"margin-top:20px;font-size:15px;color:#aaa\"\u003eTwice the node count per rack unit — and two machines that fail separately.\u003c\/p\u003e\n\u003c\/div\u003e\u003cp style=\"font-size:17px;color:#333;margin-bottom:24px\"\u003eOne 1U chassis, two complete and independent servers. Each node has its own AM5 processor, its own memory, its own dual 10GbE and its own full-height dual-slot PCIe 5.0 x16 slot for a GPU. They share only the chassis and a 1+1 1600 W Platinum supply, so a fault on one node does not take the other with it.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eSpecification\u003c\/h2\u003e\u003ctable style=\"width:100%;border-collapse:collapse;margin-bottom:24px;font-size:15px\"\u003e\n\u003cthead\u003e\u003ctr style=\"background:#0d0d0d;color:#fff\"\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eComponent\u003c\/th\u003e\n\u003cth style=\"padding:12px 16px;text-align:left;color:#fff !important\"\u003eDetail\u003c\/th\u003e\n\u003c\/tr\u003e\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eForm factor\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1U rackmount, 2 nodes\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eProcessor (per node)\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003eSingle Socket AM5 (LGA 1718). AMD EPYC 4005 \/ 4004, or Ryzen 9000 \/ 8000 \/ 7000. Up to 120 W TDP\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eMemory (per node)\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e4 DIMM slots, 2DPC. DDR5 ECC or non-ECC UDIMM, 48 GB per module — 192 GB maximum. 5600 MHz at 1DPC, 3600 MHz at 2DPC\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eExpansion (per node)\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1 × FH dual-slot PCIe 5.0 x16, or 2 × single-slot PCIe 5.0 x8\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eStorage (per node)\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1 × M.2 PCIe 4.0 x4\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#fff\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003eNetworking (per node)\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e2 × 10GbE Broadcom BCM57416, IPMI\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr style=\"background:#f8f8f8\"\u003e\n\u003ctd style=\"padding:10px 16px;font-weight:600;width:34%\"\u003ePower (shared)\u003c\/td\u003e\n\u003ctd style=\"padding:10px 16px\"\u003e1+1 CRPS, 1600 W at 200–240 V (1000 W at 110 V), 80 PLUS Platinum\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhere the twin layout pays, and where it does not\u003c\/h2\u003e\u003cp\u003eTwo nodes per rack unit halves the rack space and the power distribution for a given node count, and gives you genuine hardware separation — two tenants, a live and a standby, a pair of CI runners, or two inference endpoints that must not share a fault domain.\u003c\/p\u003e\u003cp\u003eThe limits are the 120 W TDP ceiling per node, 192 GB of UDIMM per node, and a single M.2 for local storage. This is a fleet-density machine, not a machine for one big job: it will not train anything of consequence and it is not where a large model should live. Run two modest models, or two small services that each want their own card.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eRack and cooling\u003c\/h2\u003e\u003cp\u003eFront-to-back airflow, 24\/7 rated, and loud in the way all 1U twin chassis are. Budget power for both nodes plus both cards against the 1600 W redundant figure — and remember the supply is shared, so size it for the pair, not for one.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat is included\u003c\/h2\u003e\u003cp\u003e\u003cstrong\u003eWhat is in the box:\u003c\/strong\u003e chassis, backplane, motherboard, the full fan wall and the power supplies listed above, pre-fitted and cabled. It is a complete, working platform less the parts you choose.\u003c\/p\u003e\u003cp\u003e\u003cstrong\u003eWhat is not:\u003c\/strong\u003e CPU, memory, drives, GPU and any add-in NIC. Those are specified with you and priced separately — that is the point of buying a barebone rather than a finished server.\u003c\/p\u003e\u003ch2 style=\"font-size:22px;font-weight:700;margin:40px 0 16px 0;padding-bottom:8px;border-bottom:3px solid #334fb4\"\u003eWhat Kentino adds\u003c\/h2\u003e\u003cp\u003eEvery barebone we ship is unpacked, assembled, powered up and tested before it leaves Prague: fan wall and PSU redundancy checked, BMC firmware brought up to the current release, IPMI configured and the network details handed over in writing. You get a machine that has already been switched on once by somebody who knows what it should sound like.\u003c\/p\u003e\u003cp\u003eTell us the workload instead of the part numbers and we will specify the CPU, memory and storage for you, build it, burn it in and ship it as a finished server. The same chassis, the same price for the chassis — we simply add the parts and the work.\u003c\/p\u003e\u003cp style=\"font-size:14px;color:#666\"\u003e\u003cem\u003ePrice shown is EUR 4,178 excluding VAT for the barebone as described, \u003cstrong\u003eEXW Prague\u003c\/strong\u003e (Incoterms 2020). Freight, insurance and any destination duties are for the buyer’s account. Typical lead time 10–28 days. Image: ASRock Rack.\u003c\/em\u003e\u003c\/p\u003e","brand":"Kentino \/ ASRock Rack","offers":[{"title":"Default Title","offer_id":54509189529928,"sku":"1U2N2G-AM5\/2T","price":4190.0,"currency_code":"EUR","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/5479\/3800\/files\/01-1u2n2g-am5-2t-1.jpg?v=1793969132"}],"url":"https:\/\/kentino.com\/pl\/collections\/servers.oembed","provider":"Kentino","version":"1.0","type":"link"}