[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-672776-105":59,"doc-detail-672776-en":129},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":122,"head_meta":124,"extra_data":126,"updated_unix":128},105,"en","nvidia-a100-tensor-core-gpu-datasheet-jun20","NVIDIA A100 Tensor Core GPU - Datasheet - Jun20","","NVIDIA A100 Tensor Core GPU datasheet describes unprecedented acceleration for AI training and inference, data analytics, and HPC. It highlights scaling from single workloads to thousands of GPUs, and introduces Multi-Instance GPU (MIG) to partition one A100 into up to seven isolated GPU instances. The document also details third-generation Tensor Cores, HBM2 memory performance, next-generation NVLink/NVSwitch interconnect scaling, and structural sparsity for faster sparse-model execution. System specifications cover peak compute, memory, interfaces, power, and software APIs.",{"@graph":69,"@context":121},[70,84,104],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":30,"@type":76,"position":81},"https://docshare.wps.com/document/technology/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/nvidia-a100-tensor-core-gpu-datasheet-jun20/672776/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":98,"encodingFormat":97,"isAccessibleForFree":99,"interactionStatistic":100},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/nvidia-a100-tensor-core-gpu-datasheet-jun20/672776.png","ImageObject",300,407,{"name":92,"@type":93},"miraaa_2004","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-02",true,{"@type":101,"interactionType":102,"userInteractionCount":14},"InteractionCounter",{"@type":103},"ViewAction",{"@type":105,"mainEntity":106},"FAQPage",[107,113,117],{"name":108,"@type":109,"acceptedAnswer":110},"A100如何支持从单节点到大规模多GPU的扩展？","Question",{"text":111,"@type":112},"A100可通过MIG对单GPU进行硬件级分区，或通过NVLink连接多GPU以加速大规模工作负载，从最小任务到多节点规模均可覆盖。","Answer",{"name":114,"@type":109,"acceptedAnswer":115},"MIG（Multi-Instance GPU）提供了什么能力？",{"text":116,"@type":112},"A100最多可分成七个GPU实例，这些实例在硬件层面完全隔离，并各自拥有高带宽内存、缓存和计算核心。",{"name":118,"@type":109,"acceptedAnswer":119},"A100的主要关键创新包括哪些方面？",{"text":120,"@type":112},"包括第三代Tensor Core、HBM2高带宽内存、下一代NVLink/NVSwitch互联扩展，以及结构稀疏（Structural Sparsity）以提升稀疏模型性能。","https://schema.org",{"og:url":83,"og:type":123,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":125,"canonical":83},"index,follow",{"doc_id":127,"site_id":62},672776,1793607139,{"code":4,"msg":5,"data":130},{"doc_id":127,"user_id":131,"nickname":92,"user_avatar":132,"doc_module":4,"category_id":29,"category_name":30,"doc_title":65,"doc_description":67,"doc_content":133,"file_id":134,"file_url":135,"file_type":136,"file_size":137,"view_count":14,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":14,"language":138,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":139,"faqs":140,"seo_title":141,"seo_description":67,"update_tm":128,"read_time":24},962094249165,"https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0","NVIDIA A100  \nTENSOR CORE GPU  \nUnprecedented Acceleration at Every Scale  \nThe NVIDIA A100 Tensor Core GPU delivers unprecedented acceleration at every scale for AI, data analytics, and HPC to tackle the world’s toughest computing challenges. As the engine of the NVIDIA data center platform, A100 can efficiently scale up to thousands of GPUs or, using new Multi-Instance GPU (MIG) technology, can be partitioned into seven isolated GPU instances to accelerate workloads of all sizes. A100’s thirdgeneration Tensor Core technology now accelerates more levels of precision for diverse workloads, speeding time to insight as well as time to market.  \n\n| Up to 6X Higher Out-of-the-Box\u003Cbr>Performance with TF32 for AI Training1 BERT Large Training |  |  |  |\n| --- | --- | --- | --- |\n| Relative Performance | 7X\u003Cbr>6X\u003Cbr>5X\u003Cbr>4X\u003Cbr>3X\u003Cbr>2X\u003Cbr>1X\u003Cbr>0 | 6X \u003Cbr>1X\u003Cbr>|  |\n|  |  | NVIDIA V100 FP32 | NVIDIA A100 TF32 |\n\n\n| Up to 7X Higher Performance with\u003Cbr>Multi-Instance GPU (MIG) for AI Inference2 BERT Large Inference |  |  |  |  |\n| --- | --- | --- | --- | --- |\n| Sequences/second | 7,000 6,000 5,000 4,000 3,000 2,000\u003Cbr>1,000 0 | 7X 0.6X   1X \u003Cbr>|  |  |\n|  |  | NVIDIA T4 | NVIDIA V100 | NVIDIA A100 |\n\n\n| Up to 2X More HPC performance3\u003Cbr>Molecular Dynamics Physics Geo Science Physics |  |\n| --- | --- |\n| 2.0X\u003Cbr>1.5X\u003Cbr>1.0X\u003Cbr>0.5X\u003Cbr>0.0X | |\n\nSYSTEM SPECIFICATIONS (PEAK PERFORMANCE)  \n\n|  | NVIDIA A100 for\u003Cbr>NVIDIA HGX™ | NVIDIA A100 for\u003Cbr>PCIe |\n| --- | --- | --- |\n| GPU Architecture | NVIDIA Ampere |  |\n| Double-Precision Performance | FP64: 9.7 TFLOPS FP64 Tensor Core: 19.5 TFLOPS |  |\n| Single-Precision Performance | FP32: 19.5 TFLOPS Tensor Float 32 (TF32): 156 TFLOPS | 312 TFLOPS* |  |\n| Half-Precision Performance | 312 TFLOPS | 624 TFLOPS* |  |\n| Bfloat16 | 312 TFLOPS | 624 TFLOPS* |  |\n| Integer Performance | INT8: 624 TOPS | 1,248 TOPS*\u003Cbr>INT4: 1,248 TOPS | 2,496 TOPS* |  |\n| GPU Memory | 40 GB hBM2 |  |\n| Memory Bandwidth | 1.6 TB/sec |  |\n| Error-Correcting Code | Yes |  |\n| Interconnect Interface | PCIe Gen4: 64 GB/\u003Cbr>sec Third generation\u003Cbr>NVIDIA® NVLink® :\u003Cbr>600 GB/sec** | PCIe Gen4: 64 GB/\u003Cbr>sec Third generation\u003Cbr>NVIDIA® NVLink® :\u003Cbr>600 GB/sec** |\n| Form Factor | 4/8 SXM GPUs in\u003Cbr>NVIDIA hGX™ A100 | PCIe |\n| Multi-Instance GPU (MIG) | Up to 7 GPU instances |  |\n| Max Power Consumption | 400 W | 250 W |\n| Delivered Performance for Top Apps | 100% | 90% |\n| Thermal Solution | Passive |  |\n| Compute APIs | CUDA®, DirectCompute,\u003Cbr>OpenCL™, OpenACC® |  |\n\n* Structural sparsity enabled  \n** SXM GPUs via HGX A100 server boards; PCIe GPUs via NVLink Bridge for up to 2 GPUs  \nNVIDIA A100 | DATAShEET | JUN20  \nGROUNDBREAKING INNOVATIONS  \nNVIDIA AMPERE ARCHITECTURE  \nA100 accelerates workloads big and small. Whether using MIG to partition an A100 GPU into smaller instances, or NVLink to connect multiple GPUs to accelerate large-scale workloads, A100 can readily handle differentsized acceleration needs, from the smallest job to the biggest multi-node workload. A100’s versatility means IT managers can maximize the utility of every GPU in their data center around the clock.  \nMULTI-INSTANCE GPU (MIG)  \nAn A100 GPU can be partitioned into as many as seven GPU instances, fully isolated atthe hardware level with their own high-bandwidth memory, cache, and compute cores.  \nMIG gives developers access to breakthrough acceleration for all their applications, and IT administrators can offer rightsized GPU acceleration for every job, optimizing utilization and expanding access to every user and application.  \nTHIRD-GENERATION TENSOR CORES  \nA100 delivers 312 teraFLOPS (TFLOPS) of deep learning performance. That’s 20X Tensor FLOPS for deep learning training and 20X Tensor TOPS for deep learning inference compared to NVIDIA Volta™ GPUs.  \nHBM2  \nWith 40 gigabytes (GB) of highbandwidth memory (HBM2), A100 delivers improved raw bandwidth of 1. 6TB/sec, as well as higher dynamic random-access memory (DRAM) utilization efficiency a","cbCainm7TQhKY5DR","https://ap.wps.com/l/cbCainm7TQhKY5DR","pdf",655380,"English","# System Specifications (Peak Performance)\n## NVIDIA A100 for NVIDIA HGX\n## NVIDIA A100 for PCIe\n# Groundbreaking Innovations\n## NVIDIA Ampere Architecture\n## Multi-Instance GPU (MIG)\n## Third-Generation Tensor Cores\n## HBM2\n## Next-Generation NVLink\n## Structural Sparsity\n# Product Overview","[{\"question\":\"A100如何支持从单节点到大规模多GPU的扩展？\",\"answer\":\"A100可通过MIG对单GPU进行硬件级分区，或通过NVLink连接多GPU以加速大规模工作负载，从最小任务到多节点规模均可覆盖。\"},{\"question\":\"MIG（Multi-Instance GPU）提供了什么能力？\",\"answer\":\"A100最多可分成七个GPU实例，这些实例在硬件层面完全隔离，并各自拥有高带宽内存、缓存和计算核心。\"},{\"question\":\"A100的主要关键创新包括哪些方面？\",\"answer\":\"包括第三代Tensor Core、HBM2高带宽内存、下一代NVLink/NVSwitch互联扩展，以及结构稀疏（Structural Sparsity）以提升稀疏模型性能。\"}]","NVIDIA A100 Tensor Core GPU - Datasheet - Jun20 | PDF"]