Enquire about this solution

AMD Instinct GPU Accelerators


The new AMD Instinct MI200 series accelerators are the newest data centre GPU from AMD, and are powering the first exascale high performance computer. They are designed to accelerate HPC, scientific compute, research and AI workloads. The MI series delivers dramatically faster Floating Point calculations as well as much faster Tensor matrix calculations.

These AMD GPUs are ideally paired with AMD EPYC CPUs where they can take advantage of AMD’s Infinity Fabric technology which adds direct GPU to CPU connectivity and enables cache coherency.

model no.
description
key features
model no. AMD Instinct™ MI250X Accelerator «NEW»
description

The AMD Instinct™ MI250X Accelerators deliver a quantum leap in HPC and AI performance over competitive data center GPUs today.

key features
  • Delivers industry-leading double precision performance for HPC workloads with up to 47.9TFLOPS peak FP64 performance
  • Provide up to 128GB High-bandwidth HBM2e memory with ECC support at a clock rate of 1.6 GHz.
  • Deliver an ultra-high 3.2 TB/s of memory bandwidth
  • Deliver up to 800 GB/s I/O bandwidth performance

Learn more

model no. AMD Instinct™ MI250 Accelerator «NEW»
description

The AMD Instinct MI250 accelerator handles large data efficiently for training and is the first data center GPU to deliver 383 teraflops of peak FP16 performance for deep learning training.

key features
  • Delivers industry-leading double precision performance for HPC workloads with up to 47.9TFLOPS peak FP64 performance
  • Provide up to 128GB High-bandwidth HBM2e memory with ECC support at a clock rate of 1.6 GHz.
  • Deliver an ultra-high 3.2 TB/s of memory bandwidth
  • Deliver up to 800 GB/s I/O bandwidth performance

Learn more

model no. AMD Instinct™ MI210 Accelerator «NEW»
description

The AMD Instinct™ 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.

key features
  • up to 22.6 TFLOPS peak FP64|FP32 performance
  • delivers an outstanding 181 teraflops peak theoretical FP16 and BF16 performance
  • delivers 64 GB/s CPU to GPU bandwidth without the need for PCIe® switches
  • up to 300 GB/s of Peer-to-Peer (P2P) bandwidth performance through three Infinity Fabric links

Learn more