AMD Instinct MI200 CDNA 2 ‘Aldebaran’ MCM HPC GPU Accelerator Launching Later This Year, Will Compete Against Intel Ponte Vecchio & NVIDIA Ampere

Submit

AMD's CEO has confirmed that the successor to the CDNA architecture-powered Instinct GPU HPC accelerator is on its way for launch later this year.

AMD Instinct MI200 With CDNA 2 MCM GPU Architecture Landing Later This Year, Will Power HPC Workloads

The confirmation came during JPMorgan's 49th Annual Global Technology Communications and Media conference. AMD's CEO, Lisa Su, stated that they will be launching the next generation of CDNA architecture later this year. Following is the transcript from the conference (Source: Seeking Alpha).

Intel CEO on Alder Lake CPUs: AMD Is In The Rearview Mirror in Client & Never Again Will They Be In The Windshield

Last year, we talked about our first-generation CDNA architecture. This year, as I said, we’re putting together our next-generation CDNA architecture. This is actually a key component that enabled us to win the largest supercomputer bids in the US around the Frontier Oak Ridge National Labs installment as well as the Lawrence Livermore National Labs installment with El Capitan and many others.

But it’s a coherent interconnect between CPUs and GPUs that allow us to fully optimize for HPC and for AI and ML applications. And we will be launching the next generation of that architecture, actually, later this year. We’re very excited about it. I think it’s progressed extremely well. It’s the next big step in sort of innovation around the data center architectures.

Dr. Lisa Su (AMD CEO)

Here's Everything We Know About AMD's CDNA 2 Architecture Powered Instinct Accelerators

The AMD CDNA 2 architecture will be powering the next-generation AMD Instinct HPC accelerators. We know that one of those accelerators will be the MI200 which will feature the Aldebaran GPU. It's going to be a very powerful chip and possibly the first GPU to feature an MCM design. The Instinct MI200 is going to compete against Intel's 7nm Ponte Vecchio and NVIDIA's refreshed Ampere parts. Intel and NVIDIA are also following the MCM route on their next-generation HPC accelerators but it looks like Ponte Vecchio is going to be available in 2022 and the same can be said for NVIDIA's next-gen HPC accelerator as their own roadmap confirmed.

In the previous Linux patch, it was revealed that l that the AMD Instinct MI200 'Aldebaran' GPU will feature HBM2E memory support. NVIDIA was the first to hop on board the HBM2E standard and will offer a nice boost over the standard HBM2 configuration used on the Arcturus-based MI100 GPU accelerator. HBM2E allows up to 16 GB memory capacity per stack so we can expect up to 64 GB HBM2E memory at blisteringly fast speeds for Aldebaran.

The latest Linux Kernel Patch revealed that the GPU carries 16 KB of L1 cache per CU which makes up 2 MB of the total L1 cache considering that the GPU will be packing 128 Compute Units. The GPU also carries 8 MB of shared L2 cache but carries 14 CUs per Shader Engine compared to 16 CUs per SE in the previous Instinct lineup. Regardless, it is stated that each CU on Aldebaran GPUs will have a significantly higher computing output.

Other features listed include SDMA (System Direct Memory Access) support which will allow data transfers over PCIe and XGMI/Infinity Cache subsystems. As far as Infinity Cache is concerned, it's looking like that won't be happening on HPC GPUs. Do note that AMD's CDNA 2 GPU will be fabricated on a brand new process node & are confirmed to feature a 3rd Generation AMD Infinity architecture that extends to Exascale by allowing up to 8-Way coherent GPU connectivity.

AMD Radeon Instinct Accelerators 2020

Accelerator NameAMD Instinct MI300AMD Instinct MI250XAMD Instinct MI250AMD Instinct MI210AMD Instinct MI100AMD Radeon Instinct MI60AMD Radeon Instinct MI50AMD Radeon Instinct MI25AMD Radeon Instinct MI8AMD Radeon Instinct MI6
GPU ArchitectureTBA (CDNA 3)Aldebaran (CDNA 2)Aldebaran (CDNA 2)Aldebaran (CDNA 2)Arcturus (CDNA 1)Vega 20Vega 20Vega 10Fiji XTPolaris 10
GPU Process NodeAdvanced Process Node6nm6nm6nm7nm FinFET7nm FinFET7nm FinFET14nm FinFET28nm14nm FinFET
GPU Dies4 (MCM)?2 (MCM)2 (MCM)1 (MCM)1 (Monolithic)1 (Monolithic)1 (Monolithic)1 (Monolithic)1 (Monolithic)1 (Monolithic)
GPU Cores28,160?14,08013,3126656768040963840409640962304
GPU Clock SpeedTBA1700 MHz1700 MHz~1700 MHz?~1500 MHz1800 MHz1725 MHz1500 MHz1000 MHz1237 MHz
FP16 ComputeTBA383 TOPs362 TOPs~176 TOPs185 TFLOPs29.5 TFLOPs26.5 TFLOPs24.6 TFLOPs8.2 TFLOPs5.7 TFLOPs
FP32 ComputeTBA95.7 TFLOPs90.5 TFLOPs~44 TFLOPs23.1 TFLOPs14.7 TFLOPs13.3 TFLOPs12.3 TFLOPs8.2 TFLOPs5.7 TFLOPs
FP64 ComputeTBA47.9 TFLOPs45.3 TFLOPs~22 TFLOPs11.5 TFLOPs7.4 TFLOPs6.6 TFLOPs768 GFLOPs512 GFLOPs384 GFLOPs
VRAMTBA128 GB HBM2e128 GB HBM2e64 GB HBM2e32 GB HBM232 GB HBM216 GB HBM216 GB HBM24 GB HBM116 GB GDDR5
Memory ClockTBA3.2 Gbps3.2 Gbps3.2 Gbps?1200 MHz1000 MHz1000 MHz945 MHz500 MHz1750 MHz
Memory BusTBA8192-bit8192-bit4096-bit4096-bit bus4096-bit bus4096-bit bus2048-bit bus4096-bit bus256-bit bus
Memory BandwidthTBA3.2 TB/s3.2 TB/s1.6 TB/s1.23 TB/s1 TB/s1 TB/s484 GB/s512 GB/s224 GB/s
Form FactorTBAOAMOAMDual Slot CardDual Slot, Full LengthDual Slot, Full LengthDual Slot, Full LengthDual Slot, Full LengthDual Slot, Half LengthSingle Slot, Full Length
CoolingTBAPassive CoolingPassive CoolingPassive CoolingPassive CoolingPassive CoolingPassive CoolingPassive CoolingPassive CoolingPassive Cooling
TDPTBA560W500W?300W?300W300W300W300W175W150W
Submit