AMD’s upcoming Instinct MI300 APUs with next-generation CDNA 3 GPUs and Zen 4 CPU cores will power the El Capitan supercomputer.
El Capitan Supercomputer with AMD’s Instinct MI300 APUs (Zen 4 CPU and CDNA 3 GPU) to deliver up to 2 Exaflops of double-precision computing power
According to HPC Wire, during the 79th HPC User Forum at Oak Ridge National Laboratory (ORNL), Associate Director for High Performance Computing at Lawrence Livermore National Laboratory (LLNL), Terri Quinn, revealed that the El Capitan supercomputer that will be released in late 2023, will use AMD’s next-generation Instinct MI300 APUs.
The El Capitan supercomputer will use several nodes, each equipped with different Instinct MI300 APU accelerators that feature a socketed design and can be housed by AMD’s SP5 (LGA 6096) socket. The system is said to offer 10x better performance than the Sierra, which uses IBM’s Power9 CPUs and NVIDIA’s Volta GPUs and has been operational since 2018. In the FP64 (double precision) theoretical calculation, the system should deliver an astounding value of two Exaflops of power and will do so under 40 MW.
“This is the first time we have publicly stated this,” said Quinn, associate director for HPC at LLNL. “I cut these words out [AMD’s] investor document and that’s what it says: it’s a 3D chiplet design with AMD CDNA3 GPU, Zen 4 CPU, cache memory and HBM chiplet.
“I can’t give you all the specifics, but [El Capitan] is at least 10 times the performance of Sierra in terms of average merit, “Quinn said.” The theoretical peak is two double-precision exaflops, [and we’ll] keep it under 40 megawatts, the same reason as Oak Ridge, the operating cost.
via HPC cable
El Capitan will be designed by HPE and will be equipped with Slingshot-11 interconnects to connect each HPE Cray XE rack together. The AMD Instinct MI300 will pave the way for future Exascale-class APUs. During AMD Financial Day 2022, the company confirmed that the MI300 will be a multi-chip, multi-IP Instinct accelerator that not only features the next-generation CDNA 3 GPU cores, but is also equipped with the Zen 4 CPU. next generation core.
To enable more than 2 exaflops of double-precision processing power, the US Department of Energy, Lawrence Livermore National Laboratory and HPE have teamed up with AMD to design El Capitan, which is expected to be the world’s fastest supercomputer with delivery scheduled for early 2023. Capitan will leverage next generation products that incorporate custom processor design enhancements in Frontier.
- The next generation AMD EPYC processors, codenamed “Genoa”, will feature the “Zen 4” processor core to support next generation memory and I / O subsystems for AI and HPC workloads
- Next-generation AMD Instinct GPUs based on a new compute-optimized architecture for HPC and AI workloads will utilize next-generation high-bandwidth memory for optimal deep learning performance
This project will excel in artificial intelligence and machine learning data analytics to create models that are faster, more accurate and able to quantify the uncertainty of their predictions.
AMD will use the 5nm process node for its Instinct MI300 “CDNA 3” GPUs. The chip will be equipped with the next generation of Infinity Cache and will feature the fourth generation Infinity architecture that enables support for the CXL 3.0 ecosystem. The Instinct MI300 accelerator will rock a unified memory APU architecture and new math formats, allowing for a 5x performance increase per watt over CDNA 2, which is huge. AMD also predicts more than 8x the performance of AI compared to CDNA 2-based Instinct MI250X accelerators. The CDNA 3 GPU UMAA will connect the CPU and GPU to a unified HBM memory package, eliminating redundant copies of memory and offering at the same time a low TCO.
AMD’s Instinct MI300 APU accelerators are expected to be available by the end of 2023, which coincides with the implementation of the El Capitan supercomputer mentioned above.
Thanks toFor the tip!