AMD Instinct™ MI210 Accelerator
NEWExascale-Class Technologies for the Data Center
The AMD Instinct™ MI210 accelerator extends AMD industry performance leadership in accelerated compute for double precision (FP64) on PCIe® form factors for mainstream HPC and AI workloads in the data center.1,2 Built on AMD Exascale-class technologies with the 2nd Gen AMD CDNA™ architecture, the MI210 enables scientists and researchers to tackle our most pressing challenges from climate change to vaccine research.MI210 accelerators, combined with the AMD ROCm™ 5 software ecosystem, allow innovators to tap the power of HPC and AI data center PCIe® GPUs to accelerate their time to science and discovery.
Get a Quote
GPU Specifications
- GPU Architecture: CDNA2
- Stream Processors: 6656
- Peak Half Precision (FP16) Performance: 181 TFLOPs
- Peak Single Precision Matrix (FP32) Performance: 45.3 TFLOPs
- Peak Single Precision (FP32) Performance: 22.6 TFLOPs
- Peak INT4 Performance: 181 TOPs
- Peak bfloat16: 181 TFLOPs
- Compute Units: 104
- Peak Engine Clock: 1700 MHz
- Peak Double Precision Matrix (FP64) Performance: 45.3 TFLOPs
- Peak Double Precision (FP64) Performance: 22.6 TFLOPs
- Peak INT8 Performance: 181 TOPs
- OS Support Linux: x86_64
Requirements
- External Power Connectors: 1x8 pin 12V EPS
- Total Board Power (TBP): 300W Peak
GPU Memory
- Dedicated Memory Size: 64 GB
- Memory Interface: 4096-bit
- Peak Memory Bandwidth: Up to 1638.4 GB/s
- Dedicated Memory Type: HBM2e
- Memory Clock: 1.6 GHz
- Memory ECC Support: Yes (Full-Chip)
Board Specifications
- Form Factor: PCIe® Add-in Card
- Infinity Fabric™ Links: 3
- Cooling Passive: Passive
- Bus Type PCIe®: 4.0 x16
- Peak Infinity Fabric™ Link Bandwidth: 100 GB/s
Dimensions
- Board Height: Full Height
- Board Width: Double Slot
- Board Length: 10.5" (267mm)
Additional Features
- Supported Technologies: AMD CDNA™ 2 Architecture
- AMD Infinity Architecture
- Page Retirement: Yes
- RAS Support: Yes
Software API Support
- OpenMP®: Yes
- HIP: Yes
- OpenCL™: Yes
Frameworks
- TensorFlow: Yes
- Kokkos: Yes
- PyTorch: Yes
- RAJA: Yes
Product Basics
- Product Family: AMD Instinct™
- Platform: Server
- Product Line: AMD Instinct™ MI Series
- Launch Date: 3/22/22