{"product_id":"hpe-nvidia-l4-24gb-gddr6-pcie-accelerator-graphics-card-s0k89c-401233","title":"HPE NVIDIA L4 24GB GDDR6 PCIe Accelerator Graphics Card S0K89C","description":"\u003ch2\u003eHPE NVIDIA L4 24GB GDDR6 PCIe Accelerator Graphics Card S0K89C\u003c\/h2\u003e\n\n\u003cp\u003eThe HPE NVIDIA L4 24GB PCIe Accelerator is designed to bring efficient GPU acceleration to supported HPE compute platforms. Powered by the NVIDIA Ada Lovelace architecture, it supports enterprise workloads including artificial intelligence inference, video processing, virtual workstations, graphics, rendering and data analytics.\u003c\/p\u003e\n\n\u003cp\u003e\u003cimg alt=\"NVIDIA L4 24GB PCIe accelerator reference image\" src=\"https:\/\/www.nvidia.com\/content\/nvidiaGDC\/au\/en_AU\/data-center\/l4\/_jcr_content\/root\/responsivegrid\/nv_container_902983955\/nv_image.coreimg.jpeg\/1694423302287\/nvidia-l4-qtr-left.jpeg\"\u003e\u003c\/p\u003e\n\n\u003ch3\u003eVersatile Enterprise Acceleration\u003c\/h3\u003e\n\n\u003cp\u003eThe NVIDIA L4 combines fourth-generation Tensor Cores with third-generation RT Cores to accelerate a broad selection of AI, visual computing and graphics applications. Its 24GB of GDDR6 memory and 300 GB\/s memory bandwidth provide capacity and throughput for data-intensive enterprise workloads.\u003c\/p\u003e\n\n\u003ch3\u003eDesigned for AI and Video Processing\u003c\/h3\u003e\n\n\u003cp\u003eThe accelerator supports mixed-precision processing for deep learning and machine learning applications. Two NVENC encoders, four NVDEC decoders and four JPEG decoders also provide dedicated hardware resources for video transcoding, streaming, visual effects and image-processing workflows.\u003c\/p\u003e\n\n\u003ch3\u003eCompact, Energy-Conscious Design\u003c\/h3\u003e\n\n\u003cp\u003eA half-height, half-length, single-slot form factor allows the L4 to fit into a range of supported HPE server configurations. The card operates within a maximum 72W power envelope and uses passive cooling, making correctly configured server airflow essential.\u003c\/p\u003e\n\n\u003ch3\u003eKey Features:\u003c\/h3\u003e\n\n\u003cul\u003e\n\t\u003cli\u003eNVIDIA Ada Lovelace architecture supports AI, graphics, video and virtualisation workloads.\u003c\/li\u003e\n\t\u003cli\u003e24GB of GDDR6 memory provides capacity for demanding datasets and visual applications.\u003c\/li\u003e\n\t\u003cli\u003e300 GB\/s memory bandwidth supports efficient movement of data within the accelerator.\u003c\/li\u003e\n\t\u003cli\u003ePCIe Gen4 x16 connectivity provides a high-bandwidth connection to supported HPE servers.\u003c\/li\u003e\n\t\u003cli\u003eLow-profile, half-length, single-slot construction helps conserve valuable server space.\u003c\/li\u003e\n\t\u003cli\u003e72W maximum power envelope supports efficient deployment in compatible data-centre and edge systems.\u003c\/li\u003e\n\t\u003cli\u003ePassive thermal design works with the airflow supplied by an approved server configuration.\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3\u003ePerformance for Mixed Workloads:\u003c\/h3\u003e\n\n\u003ctable\u003e\n\t\u003cthead\u003e\n\t\t\u003ctr bgcolor=\"#444444\"\u003e\n\t\t\t\u003cth scope=\"col\" style=\"color: #ffffff;\"\u003eProcessing Format\u003c\/th\u003e\n\t\t\t\u003cth scope=\"col\" style=\"color: #ffffff;\"\u003ePerformance\u003c\/th\u003e\n\t\t\u003c\/tr\u003e\n\t\u003c\/thead\u003e\n\t\u003ctbody\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eFP32\u003c\/td\u003e\n\t\t\t\u003ctd\u003e30.3 TFLOPS\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eTF32 Tensor Core\u003c\/td\u003e\n\t\t\t\u003ctd\u003eUp to 120 TFLOPS*\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eFP16 Tensor Core\u003c\/td\u003e\n\t\t\t\u003ctd\u003eUp to 242 TFLOPS*\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eBFLOAT16 Tensor Core\u003c\/td\u003e\n\t\t\t\u003ctd\u003eUp to 242 TFLOPS*\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eFP8 Tensor Core\u003c\/td\u003e\n\t\t\t\u003ctd\u003eUp to 485 TFLOPS*\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eINT8 Tensor Core\u003c\/td\u003e\n\t\t\t\u003ctd\u003eUp to 485 TOPS*\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\u003c\/tbody\u003e\n\u003c\/table\u003e\n\n\u003cp\u003e\u003cem\u003e*Tensor Core performance is shown with sparsity. NVIDIA states that these figures are halved without sparsity.\u003c\/em\u003e\u003c\/p\u003e\n\n\u003ch3\u003eSupported Applications:\u003c\/h3\u003e\n\n\u003cul\u003e\n\t\u003cli\u003eArtificial intelligence and machine learning inference\u003c\/li\u003e\n\t\u003cli\u003eGenerative AI workloads\u003c\/li\u003e\n\t\u003cli\u003eComputer vision and video analytics\u003c\/li\u003e\n\t\u003cli\u003eVideo transcoding and streaming\u003c\/li\u003e\n\t\u003cli\u003eRendering and visual computing\u003c\/li\u003e\n\t\u003cli\u003eVirtual desktops and virtual workstations\u003c\/li\u003e\n\t\u003cli\u003eData science and data analytics\u003c\/li\u003e\n\u003c\/ul\u003e\n\n\u003ch3\u003eTechnical Specifications:\u003c\/h3\u003e\n\n\u003ctable\u003e\n\t\u003cthead\u003e\n\t\t\u003ctr bgcolor=\"#444444\"\u003e\n\t\t\t\u003cth scope=\"col\" style=\"color: #ffffff;\"\u003eSpecification\u003c\/th\u003e\n\t\t\t\u003cth scope=\"col\" style=\"color: #ffffff;\"\u003eDetails\u003c\/th\u003e\n\t\t\u003c\/tr\u003e\n\t\u003c\/thead\u003e\n\t\u003ctbody\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eBrand\u003c\/td\u003e\n\t\t\t\u003ctd\u003eHPE \/ NVIDIA\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eModel\u003c\/td\u003e\n\t\t\t\u003ctd\u003eNVIDIA L4\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eHPE Product Number\u003c\/td\u003e\n\t\t\t\u003ctd\u003eS0K89C\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eProduct Type\u003c\/td\u003e\n\t\t\t\u003ctd\u003eServer PCIe graphics and compute accelerator\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eGPU Architecture\u003c\/td\u003e\n\t\t\t\u003ctd\u003eNVIDIA Ada Lovelace\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eNumber of Accelerators\u003c\/td\u003e\n\t\t\t\u003ctd\u003eOne per card\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eGPU Memory\u003c\/td\u003e\n\t\t\t\u003ctd\u003e24GB GDDR6\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eMemory Bandwidth\u003c\/td\u003e\n\t\t\t\u003ctd\u003e300 GB\/s\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eMemory Bus Width\u003c\/td\u003e\n\t\t\t\u003ctd\u003e192-bit\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eSystem Interface\u003c\/td\u003e\n\t\t\t\u003ctd\u003ePCIe Gen4 x16\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eInterface Bandwidth\u003c\/td\u003e\n\t\t\t\u003ctd\u003eUp to 64 GB\/s\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eVideo Engines\u003c\/td\u003e\n\t\t\t\u003ctd\u003e2 × NVENC, 4 × NVDEC and 4 × JPEG decoders\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eECC Support\u003c\/td\u003e\n\t\t\t\u003ctd\u003eSupported and enabled by default\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eThermal Design\u003c\/td\u003e\n\t\t\t\u003ctd\u003ePassive cooling; compatible server airflow required\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eMaximum Power\u003c\/td\u003e\n\t\t\t\u003ctd\u003e72W\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eMinimum Configurable Power\u003c\/td\u003e\n\t\t\t\u003ctd\u003e40W\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eForm Factor\u003c\/td\u003e\n\t\t\t\u003ctd\u003eHalf-height, half-length, single-slot, low-profile PCIe card\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eDimensions\u003c\/td\u003e\n\t\t\t\u003ctd\u003e16.85 × 6.89 × 1.87 cm\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eWeight\u003c\/td\u003e\n\t\t\t\u003ctd\u003eApproximately 0.27 kg\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr\u003e\n\t\t\t\u003ctd\u003eSupported Platform\u003c\/td\u003e\n\t\t\t\u003ctd\u003eSelected HPE mainstream compute platforms\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\t\u003ctr bgcolor=\"#f2f2f2\"\u003e\n\t\t\t\u003ctd\u003eWarranty\u003c\/td\u003e\n\t\t\t\u003ctd\u003eHPE Qualified Options limited warranty with three-year parts coverage; terms and regional conditions apply\u003c\/td\u003e\n\t\t\u003c\/tr\u003e\n\t\u003c\/tbody\u003e\n\u003c\/table\u003e\n\n\u003ch3\u003eHPE Server Compatibility\u003c\/h3\u003e\n\n\u003cp\u003eHPE lists the S0K89C for selected configurations across its ProLiant, Alletra, Apollo, edge and scale-up server families. These include supported models from the HPE ProLiant DL and ML Gen10 Plus, Gen11 and Gen12 ranges.\u003c\/p\u003e\n\n\u003cp\u003e\u003cstrong\u003eImportant:\u003c\/strong\u003e GPU support can depend on the server model, processor platform, riser configuration, available PCIe slot, power arrangement, airflow requirements and any required enablement kit. Confirm compatibility in the current HPE QuickSpecs for your exact server configuration before ordering.\u003c\/p\u003e\n\n\u003cp\u003eAdd the HPE NVIDIA L4 accelerator to a compatible server to support AI, video, graphics and virtualisation workloads with a compact 24GB enterprise GPU.\u003c\/p\u003e\n","brand":"FirstShop","offers":[{"title":"Default Title","offer_id":53257261973871,"sku":"S0K89C","price":66649.0,"currency_code":"ZAR","in_stock":false}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0457\/5893\/9300\/files\/s0k89c-graphics-cards-1251672931.jpg?v=1785486997","url":"https:\/\/www.firstshop.co.za\/products\/hpe-nvidia-l4-24gb-gddr6-pcie-accelerator-graphics-card-s0k89c-401233","provider":"FirstShop","version":"1.0","type":"link"}