Earlier today, AMD announced the new AMD Instinct MI200 series accelerators, the first exascale-class GPU accelerators. The new Instinct MI200 series accelerators include the world’s fastest high-performance computing(HPC) and artificial intelligence(AI) accelerator, the AMD Instinct MI250X.
AMD Instinct MI200 Series Specifications
|Models||Compute Units||Stream Processors||FP64 | FP32 Vector (Peak)||FP64 | FP32 Matrix (Peak)||FP16 | bf16
|INT4 | INT8
|Memory Bandwidth||Form Factor|
|AMD Instinct MI250x||220||14,080||Up to 47.9 TF||Up to 95.7 TF||Up to 383.0 TF||Up to 383.0 TOPS||128GB||3.2 TB/sec||OCP Accelerator Module|
|AMD Instinct MI250||208||13,312||Up to 45.3 TF||Up to 90.5 TF||Up to 362.1 TF||Up to 362.1 TOPS||128GB||3.2 TB/sec||OCP Accelerator Module|
The new Instinct MI200 series accelerators are built on the AMD CDNA 2 architecture delivering leading application performance for a wide set of HPC workloads. The MI250X provides up to 4.9X better performance for FP64 double-precision FP64 HPC applications and surpasses 380 teraflops of peak theoretical FP16 half-precision for AI workloads.
AMD Instinct MI200 accelerators deliver leadership HPC and AI performance, helping scientists make generational leaps in research that can dramatically shorten the time between initial hypothesis and discovery.
With key innovations in architecture, packaging and system design, the AMD Instinct MI200 series accelerators are the most advanced data center GPUs ever, providing exceptional performance for supercomputers and data centers to solve the world’s most complex problems.
-Forrest Norrod, AMD Data Center and Embedded Solution Business Group Senior Vice President and General Manager
AMD, together with the US Department of Energy, Oak Ridge National Laboratory, and HPE designed the Frontier a supercomputer expected to deliver more than 1.5 exaflops of peak computing power powered by AMD 3rd generation EPYC CPUs and AMD Instinct MI250X accelerators. Frontier aims to push the boundaries of scientific discovery by dramatically enhancing the performance of AI, analytics, and simulation at scale. Frontier will help scientists to pack in more calculations, identify new patterns in data and develop innovative data analysis methods to accelerate the pace of scientific discovery.
The Frontier supercomputer is the culmination of a strong collaboration between AMD, HPE, and the U.S. Department of Energy, to provide an exascale-capable system that pushes the boundaries of scientific discovery by dramatically enhancing the performance of artificial intelligence, analytics, and simulation at scale
-Thomas Zacharia, Oak Ridge National Laboratory Director]
AMD Instinct MI200 Key Capabilities and Features
- AMD CDNA 2 architecture – 2nd Gen Matrix Cores accelerating FP64 and FP32 matrix operations, delivering up to 4X the peak theoretical FP64 performance vs. AMD previous-gen GPUs.
- Leadership Packaging Technology – Industry-first multi-die GPU design with 2.5D Elevated Fanout Bridge (EFB) technology delivers 1.8X more cores and 2.7X higher memory bandwidth vs. AMD previous-gen GPUs, offering the industry’s best aggregate peak theoretical memory bandwidth at 3.2 terabytes per second.
- 3rd Gen AMD Infinity Fabric technology – Up to 8 Infinity Fabric links connect the AMD Instinct MI200 with 3rd Gen EPYC CPUs and other GPUs in the node to enable unified CPU/GPU memory coherency and maximize system throughput, allowing for an easier on-ramp for CPU codes to tap the power of accelerators.
AMD ROCm is an open software platform allowing researchers to tap the AMD Instinct Accelerators for scientific discoveries. ROCm 5.0 extends the platform’s open portability powering top HPC and AI applications effectively its increasing accessibility to developers.
AMD Infinity Hub allows researchers, data scientists, and end-users to easily find, download and install containerized HPC apps and ML frameworks that are optimized and supported on AMD Instinct Accelerators and ROCm. Infinity Hub currently offers support for Radeon Instinct MI50, AMD Instinct MI100, or AMD Instinct MI200 accelerators including several applications like Chroma, CP2k, LAMMPS, NAMD, OpenMM, and more, along with popular ML frameworks TensorFlow and PyTorch. New containers are continually being added to the hub.
Available Server Solutions
The newly released AMD Instinct MI250X and Instinct M250 are available in the open-hardware accelerator module or OCP accelerator module(OAM) form factor. The AMD Instinct MI210 will be available for OEM servers in the PCIe card form factor. The AMD MI250X accelerator is currently available from HPE in the HPE Cray EX Supercomputer. Additional MI 200 series accelerators are expected in systems from major OEM and ODM enterprise partners in Q1 2020 including ASUS, ATOS, Dell Technologies, Gigabyte, Hewlett Packard Enterprise (HPE), Lenovo, Penguin Computing, and Supermicro.