AMD today uploaded its Accelerated Data Center Premiere Keynote with Dr. Lisa Su and other company executives, who had a handful of exciting announcements to make regarding the company’s latest products and upcoming technologies. They include the AMD Instinct MI250 and MI250x series accelerators, the latter of which is claimed to be the world’s fastest data center accelerator for high-performance computing and artificial intelligence. The MI250x boasts up to 383 teraflops and tera operations of performance.
From AMD:
Built on AMD CDNA 2 architecture, AMD Instinct MI200 series accelerators deliver leading application performance for a broad set of HPC workloads. The AMD Instinct MI250X accelerator provides up to 4.9X better performance than competitive accelerators for double precision (FP64) HPC applications and surpasses 380 teraflops of peak theoretical half-precision (FP16) for AI workloads to enable disruptive approaches in further accelerating data-driven research.
MI200 Series Specifications
| Models | Compute Units | Stream Processors | FP64 | FP32 Vector (Peak) | FP64 | FP32 Matrix (Peak) | FP16 | bf16(Peak) | INT4 | INT8(Peak) | HBM2e ECC Memory | 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 |
“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,” said Forrest Norrod, senior vice president and general manager, Data Center and Embedded Solutions Business Group, AMD. “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.”
AMD also teased its upcoming “Zen 4” EPYC processors during its Accelerated Data Center event. “Genoa” will feature up to 96 cores and support DDR5 and PCIe 5.0, while “Bergamo” will offer up to 128 cores for cloud-native computing.
From AMD:
- “Genoa” is expected to be the world’s highest performance processor for general purpose computing. It will have up to 96 high-performance “Zen 4” cores produced on optimized 5nm technology, and will support the next generation of memory and I/O technologies with DDR5 and PCIe 5. “Genoa” will also include support for CXL, enabling significant memory expansion capabilities for data center applications. “Genoa” is on track for production and launch in 2022.
- “Bergamo” is a high-core count CPU, tailor made for cloud native applications, featuring 128 high performance “Zen 4c” cores. AMD optimized the new “Zen 4c” core for cloud-native computing, tuning the core design for density and increased power efficiency to enable higher core count processors with breakthrough performance per-socket. “Bergamo” comes with all the same software and security features and is socket compatible with “Genoa.” “Bergamo” is on track to ship in the first half of 2023.
Source: AMD