{"product_id":"hp-p54871-001-nvidia-a40-pcie-48gb-gpu-for-hpe","title":"HP P54871-001 NVIDIA A40 PCIe 48GB GPU *New Sealed* P54743-001 R7E31C R9S37C R7E31A","description":"\u003cdiv class=\"c4-rich\"\u003e\n\u003cp\u003eThe HP P54871-001 is an NVIDIA A40 data center GPU with 48GB of memory on a PCIe card, supplied for HPE servers. Built on the NVIDIA Ampere architecture, the A40 handles virtual workstations, rendering, and visualization, and it also runs AI training, inference, and compute work in the data center. It is also listed under HPE numbers P54743-001, R7E31C, R9S37C, and R7E31A.\u003c\/p\u003e\n\u003ch3 style=\"margin:1.4em 0 0.35em\"\u003eAt a glance\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003eNVIDIA A40 with 48GB GDDR6 ECC memory\u003c\/li\u003e\n\u003cli\u003ePCIe Gen 4 x16, dual-slot, passively cooled, 300W\u003c\/li\u003e\n\u003cli\u003eHPE version, sold by HPE as R9S37C\u003c\/li\u003e\n\u003cli\u003eNew, factory sealed\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3 style=\"margin:1.4em 0 0.35em\"\u003eOutfitting several servers with A40s?\u003c\/h3\u003e\n\u003cp\u003eNeed more cards than we have listed? Core 4 can source additional quantities, along with NVLink bridges and the servers to run them. Tell a sales engineer about your project. \u003ca href=\"\/pages\/contact?part_number=P54871-001\u0026amp;additional_requirements=Product%3A%20HP%20P54871-001%20NVIDIA%20A40%20PCIe%2048GB%20GPU%20*New%20Sealed*%20P54743-001%2C%20R7E31C%20%28https%3A%2F%2Fwww.core4solutions.com%2Fproducts%2Fhp-p54871-001-nvidia-a40-pcie-48gb-gpu-for-hpe%29\"\u003e\u003cstrong\u003eTalk to a sales engineer\u003c\/strong\u003e\u003c\/a\u003e or call 1-855-55-CORE4.\u003c\/p\u003e\n\u003ch3 style=\"margin:1.4em 0 0.35em\"\u003eTechnical specifications\u003c\/h3\u003e\n\u003ctable style=\"border-collapse:collapse;width:100%;margin:0 0 1em\"\u003e\u003ctbody\u003e\n\u003ctr\u003e\n\u003cth scope=\"row\" style=\"text-align:left;vertical-align:top;font-weight:600;padding:6px 16px 6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eGPU\u003c\/th\u003e\n\u003ctd style=\"padding:6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eNVIDIA A40\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth scope=\"row\" style=\"text-align:left;vertical-align:top;font-weight:600;padding:6px 16px 6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eArchitecture\u003c\/th\u003e\n\u003ctd style=\"padding:6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eNVIDIA Ampere\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth scope=\"row\" style=\"text-align:left;vertical-align:top;font-weight:600;padding:6px 16px 6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eGPU memory\u003c\/th\u003e\n\u003ctd style=\"padding:6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003e48GB GDDR6 with ECC\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth scope=\"row\" style=\"text-align:left;vertical-align:top;font-weight:600;padding:6px 16px 6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eMemory bandwidth\u003c\/th\u003e\n\u003ctd style=\"padding:6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003e696 GB\/s\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth scope=\"row\" style=\"text-align:left;vertical-align:top;font-weight:600;padding:6px 16px 6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eCUDA cores\u003c\/th\u003e\n\u003ctd style=\"padding:6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003e10,752\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth scope=\"row\" style=\"text-align:left;vertical-align:top;font-weight:600;padding:6px 16px 6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eRT cores\u003c\/th\u003e\n\u003ctd style=\"padding:6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003e84, second generation\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth scope=\"row\" style=\"text-align:left;vertical-align:top;font-weight:600;padding:6px 16px 6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eTensor cores\u003c\/th\u003e\n\u003ctd style=\"padding:6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003e336, third generation\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth scope=\"row\" style=\"text-align:left;vertical-align:top;font-weight:600;padding:6px 16px 6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eInterface\u003c\/th\u003e\n\u003ctd style=\"padding:6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003ePCIe Gen 4 x16\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth scope=\"row\" style=\"text-align:left;vertical-align:top;font-weight:600;padding:6px 16px 6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eMaximum power\u003c\/th\u003e\n\u003ctd style=\"padding:6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003e300W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003cth scope=\"row\" style=\"text-align:left;vertical-align:top;font-weight:600;padding:6px 16px 6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eForm factor\u003c\/th\u003e\n\u003ctd style=\"padding:6px 0;border-bottom:1px solid rgba(128,128,128,.2)\"\u003eDual-slot, full height, full length, passively cooled\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\n\u003ch3 style=\"margin:1.4em 0 0.35em\"\u003eCompatibility\u003c\/h3\u003e\n\u003cp\u003eThis is the HPE-supplied version of the NVIDIA A40, sold by HPE as the NVIDIA A40 48GB PCIe Non-CEC Accelerator for HPE (R9S37C) and listed among HPE's accelerators for ProLiant servers.\u003c\/p\u003e\n\u003cp\u003eThe card needs a PCIe x16 slot with room for a dual-slot, full-length card, a 300W auxiliary power feed, and server chassis airflow, since the A40 has no fan of its own. A Core 4 sales engineer can confirm fit for a specific server model and riser configuration.\u003c\/p\u003e\n\u003ch3 style=\"margin:1.4em 0 0.35em\"\u003eCondition\u003c\/h3\u003e\n\u003cp\u003eThis A40 is new and factory sealed: it comes in the original manufacturer packaging with the factory seal intact, ready to go straight into your build.\u003c\/p\u003e\n\u003cp\u003eIt is covered by Core 4's 1-year advanced replacement warranty. If the card fails within a year, we ship a replacement right away, before the failed unit comes back to us.\u003c\/p\u003e\n\u003ch3 style=\"margin:1.4em 0 0.35em\"\u003eLifecycle\u003c\/h3\u003e\n\u003cp\u003eThe A40 is part of NVIDIA's Ampere architecture generation of data center GPUs.\u003c\/p\u003e\n\u003ch3 style=\"margin:1.4em 0 0.35em\"\u003eDatasheets and documentation\u003c\/h3\u003e\n\u003cul\u003e\n\u003cli\u003e\u003ca href=\"https:\/\/buy.hpe.com\/sg\/en\/options\/server-accelerators\/computational-graphics-accelerators-for-servers\/computational-graphics-accelerators-for-proliant-servers\/nvidia-a40-48gb-pcie-non-cec-accelerator-for-hpe\/p\/r9s37c\" target=\"_blank\" rel=\"noopener\"\u003eNVIDIA A40 48GB PCIe Non-CEC Accelerator for HPE (HPE store page)\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"https:\/\/www.nvidia.com\/content\/dam\/en-zz\/Solutions\/Data-Center\/a40\/proviz-print-nvidia-a40-datasheet-us-nvidia-1469711-r8-web.pdf\" target=\"_blank\" rel=\"noopener\"\u003eNVIDIA A40 datasheet\u003c\/a\u003e\u003c\/li\u003e\n\u003cli\u003e\u003ca href=\"https:\/\/www.nvidia.com\/en-us\/data-center\/a40\" target=\"_blank\" rel=\"noopener\"\u003eNVIDIA A40 Data Center GPU for Visual Computing\u003c\/a\u003e\u003c\/li\u003e\n\u003c\/ul\u003e\n\u003ch3 style=\"margin:1.4em 0 0.35em\"\u003eCommon questions\u003c\/h3\u003e\n\u003cp\u003e\u003cstrong\u003eCan the A40 drive a monitor directly?\u003c\/strong\u003e The A40 has three DisplayPort 1.4 outputs, but it ships configured for virtualization with the display connectors turned off. NVIDIA's management software can switch them on when you need physical display output.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eDoes the A40 support virtual GPU workloads?\u003c\/strong\u003e Yes. The A40 is designed for NVIDIA virtual GPU software, which lets one card be shared across multiple virtual workstations or virtual machines, each with its own slice of the 48GB frame buffer.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eCan two A40 cards be linked together?\u003c\/strong\u003e Two A40 cards can be joined with an NVIDIA NVLink bridge, giving 112.5 GB\/s of bidirectional bandwidth between them and a combined 96GB of GPU memory for applications that support it. The bridge is a separate part.\u003c\/p\u003e\n\u003cp\u003e\u003cstrong\u003eWhy does the A40 need server airflow?\u003c\/strong\u003e The A40 is passively cooled, with a heatsink and no onboard fan. It relies on the server's fans pushing air across the card, so it belongs in a chassis and slot designed for a 300W passive GPU.\u003c\/p\u003e\n\u003c\/div\u003e","brand":"HP","offers":[{"title":"Default Title","offer_id":68095676088555,"sku":"hp p54871-001 new","price":4495.0,"currency_code":"USD","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0843\/8769\/2779\/files\/1_f0868ad2-9f7b-4534-9c3c-f075408a1a20.jpg?v=1793191269","url":"https:\/\/www.core4solutions.com\/products\/hp-p54871-001-nvidia-a40-pcie-48gb-gpu-for-hpe","provider":"Core 4 Solutions","version":"1.0","type":"link"}