AMD Instinct MI100 - GPU computing processor 32GB - 100-506116

AMD Instinct MI100. Graphics processor family: AMD, Graphics processor: Radeon Instinct MI100. Discrete graphics adapter memory: 32 GB, Graphics adapter memory type: High Bandwidth Memory 2 (HBM2), Memory bus: 4096 bit. Interface type: PCI Express x16 4.0. Cooling type: Passive

AMD Instinct MI100 accelerator is the world's fastest HPC GPU, engineered from the ground up for the new era of computing. Powered by the AMD CDNA architecture, the MI100 accelerators deliver a giant leap in compute and interconnect performance, offering a nearly 3.5x the boost for HPC (FP32 matrix) and a nearly 7x boost for AI (FP16) throughput compared to prior generation AMD accelerators.
MI100 accelerators supported by AMD ROCm, the industry's first open software platform, offer customers an open platform that does not lock customers into AMD solutions, enabling developers to enhance existing GPU codes to run everywhere. Combined with the award-winning AMD EPYC processors and AMD Infinity Fabric technology, MI100-powered systems provide scientists and researchers platforms that propel discoveries today and prepare them for exascale tomorrow.

Accelerate Your Discoveries
AMD Instinct™ MI100 accelerator is the world’s fastest HPC GPU, engineered from the ground up for the new era of computing.1 Powered by the AMD CDNA architecture, the MI100 accelerators deliver a giant leap in compute and interconnect performance, offering a nearly 3.5x the boost for HPC (FP32 matrix) and a nearly 7x boost for AI (FP16) throughput compared to prior generation AMD accelerators.2

Read how ORNL is Using MI100 GPUs for Plasma Physics
MI100 accelerators supported by AMD ROCm™, the industry’s first open software platform, offer customers an open platform that does not lock customers into AMD solutions, enabling developers to enhance existing GPU codes to run everywhere. Combined with the award winning AMD EPYC™ processors and AMD Infinity Fabric™ technology, MI100-powered systems provide scientists and researchers platforms that propel discoveries today and prepare them for exascale tomorrow.

World’s Fastest HPC GPU1
Delivering up to 11.5 TFLOPs of double precision (FP64) theoretical peak performance, the AMD Instinct™ MI100 accelerator delivers leadership performance for HPC applications and a substantial up-lift in performance over previous gen AMD accelerators. The MI100 delivers up to a 74% generational double precision performance boost for HPC applications.13

Unleash Intelligence Everywhere
Powered by the all-new Matrix Kärnor technology, the AMD Instinct™ MI100 accelerator delivers nearly a 7x up-lift in FP16 performance compared to prior generation AMD accelerators for AI applications.2 MI100 also greatly expands mixed precision capabilities and P2P GPU connectivity for AI and machine learning workloads.

AMD CDNA™ Architecture
Delivers ground-breaking technologies to fuel the convergence of HPC and AI in the era of Exascale.

AMD Infinity Architecture
With architecture, performance, and security leadership, our approach to processor design accelerates the pace of innovation so that you can break through years of data center stagnation.

AMD ROCm™ - Open, Flexible and Portable
The AMD ROCm™ open software platform brings a rich foundation to advanced computing by seamlessly integrating the CPU and GPU with the goal of solving real-world problems.

• Designed on AMD CDNA Architecture with 120 Compute Units (7,680 Kärnor)
• Up to 11.5 TFLOPs Peak FP64 Performance for HPC
• Up to 46.1 TFLOPs FP32 Matrix Peak Performance with All-New Matrix Kärnor for HPC & AI Workloads
• Up to 184.6 TFLOPs FP16 & 92.3 TFLOPs bFloat16 Peak for Ultra-Fast AI Training
• 32 GB Ultra-fast HBM2 ECC Memory with up to 1.2 TB/s Memory Bandwidth
• Open & portable AMD ROCm Ecosystem
• 2nd Gen Infinity Architecture with up to 340 GB/s of aggregate P2P GPU I/O bandwidth
• PCIe Gen 4 x16 Ready GPU

• All-new Matrix kärnor technology for machine learning
  The AMD Instinct MI100 GPU brings customers all-new Matrix kärnor Technology with superior performance for a full range of mixed precision operations bringing you the ability to work with large models and enhance memory-bound operation performance for whatever combination of machine learning workloads you need to deploy. The MI100 offers optimized BF16, INT4, INT8, FP16, FP32 and FP32 Matrix capabilities bringing you supercharged compute performance to meet all your AI system requirements. The AMD Instinct MI100 handles large data efficiently for training complex neural networks used in deep learning and delivers a nearly 7x boost for AI (FP16) performance compared to AMD's prior generation accelerators.
• AMD Infinity Fabric Link technology
  AMD Instinct MI100 GPUs provide advanced I/O capabilities in standard off-the-shelf servers with Infinity Fabric technologies and PCIe Gen4 support. The MI100 GPU delivers 64GB/s CPU to GPU bandwidth without the need for PCIe switches, and up to 276 GB/s of peer-to-peer (P2P) bandwidth performance through three Infinity Fabric Links designed with AMD's 2nd Gen Infinity architecture.
• AMD's Infinity technologies allow
  for platform designs with dual direct-connect quad GPU hives enabling superior P2P connectivity and delivering up to 1.1 TB/s of total theoretical GPU bandwidth within a server design.
• Ultra-fast HBM2 memory
  The AMD Instinct MI100 GPU provides 32GB High-bandwidth HBM2 memory at a clock rate of 1.2 GHz and delivers an ultra-high ~1.2 TB/s of memory bandwidth to support your largest data sets and help eliminate bottlenecks in moving data in and out of memory. Combine this performance with the MI100's advanced I/O capabilities and you can push workloads closer to their full potential.
• Industry's latest PCIe Gen 4.0
  The AMD Instinct MI100 GPU is designed to support the latest PCIe Gen 4.0 technology which provides up to 64GB/s peak theoretical transport data bandwidth from CPU to GPU per card.
• Leading FP64 performance for HPC workloads
  The AMD Instinct MI100 GPU delivers industry-leading double precision performance with up to 11.5 TFLOPS peak FP64 performance, enabling scientists and researchers across the globe to more efficiently process HPC parallel codes across several industries including life sciences, energy, finance, academics, government, defense and more.

https://www.amd.com/en/products/server-accelerators/instinct-mi100

Accelerate Your Discoveries
AMD InstinctΓäó MI100 accelerator is the worldΓÇÖs fastest HPC GPU, engineered from the ground up for the new era of computing.1 Powered by the AMD CDNA architecture, the MI100 accelerators deliver a giant leap in compute and interconnect performance, offering a nearly 3.5x the boost for HPC (FP32 matrix) and a nearly 7x boost for AI (FP16) throughput compared to prior generation AMD accelerators.2

Read how ORNL is Using MI100 GPUs for Plasma Physics
MI100 accelerators supported by AMD ROCmΓäó, the industryΓÇÖs first open software platform, offer customers an open platform that does not lock customers into AMD solutions, enabling developers to enhance existing GPU codes to run everywhere. Combined with the award winning AMD EPYCΓäó processors and AMD Infinity FabricΓäó technology, MI100-powered systems provide scientists and researchers platforms that propel discoveries today and prepare them for exascale tomorrow.

WorldΓÇÖs Fastest HPC GPU1
Delivering up to 11.5 TFLOPs of double precision (FP64) theoretical peak performance, the AMD InstinctΓäó MI100 accelerator delivers leadership performance for HPC applications and a substantial up-lift in performance over previous gen AMD accelerators. The MI100 delivers up to a 74% generational double precision performance boost for HPC applications.13

Unleash Intelligence Everywhere
Powered by the all-new Matrix Cores technology, the AMD InstinctΓäó MI100 accelerator delivers nearly a 7x up-lift in FP16 performance compared to prior generation AMD accelerators for AI applications.2 MI100 also greatly expands mixed precision capabilities and P2P GPU connectivity for AI and machine learning workloads.

AMD CDNAΓäó Architecture
Delivers ground-breaking technologies to fuel the convergence of HPC and AI in the era of Exascale.

AMD Infinity Architecture
With architecture, performance, and security leadership, our approach to processor design accelerates the pace of innovation so that you can break through years of data center stagnation.

AMD ROCmΓäó - Open, Flexible and Portable
The AMD ROCmΓäó open software platform brings a rich foundation to advanced computing by seamlessly integrating the CPU and GPU with the goal of solving real-world problems.