Nvidia edited by Graphics Technology Conference(GTC), the company’s new GPU architecture and CPU were announced. Among Nvidia CEO Jensen Huang’s statements, First CPU named Nvidia Gracewhich will go on sale in 2022. new generation Hopper GPU architecture included.
The development that excites those waiting for the RTX 40 series from NVIDIA: 4nm chips are on the way!
NVIDIA shared new information about the Hopper architecture to be used in the RTX 40 graphics cards, which are highly anticipated by the players.
What does Nvidia offer with Grace CPU and Hopper architecture?
Grace CPU SuperchipIt will be Nvidia’s first standalone chip ever, but up front, the game won’t grace your PC. Nvidia originally introduced the Grace CPU in 2021, but the term Superchip was first used at the GTC conference. This is similar to the Apple M1 Ultra Nvidia’s NVLink technologyIt means that it combines two Grace CPUs.
But unlike the M1 Ultra, the Grace Superchip was not designed for general use. 144 core CPU; Developed for artificial intelligence, data science and applications with high memory requirements. Nvidia’s first chipset ARM kernels uses.
In addition to Grace Superchip, Nvidia next generation Hopper architecture exhibited. Again, let’s say up front, this will not be the architecture that powers the RTX 4080. Hopper technology is built for data center tools. The company uses this architecture H100 GPU will be released to the market.
Built using TSMC’s N4 process, the H100 is “according to Nvidia”world’s most advanced chip“. The GPU, which includes 80 billion transistors, is also the first hardware to support PCIe 5.0 and HBM3 memory.
Nvidia also made a crazy claim to show the power of PCIe 5.0 and HBM3. Nvidia CEO stated that he can control the equivalent of the whole world’s internet traffic with only 20 H100 GPUs.
With the new DGX servers featuring eight H100 GPUs and 640GB of HBM3 memory next generation Hopper architecturee will be accessible. According to Nvidia, these machines are capable of six times higher performance than the previous generation A100. 32 petaFLOPs of artificial intelligence support offers.
While all these developments are exciting for the world of artificial intelligence and high-performance computing, Nvidia’s announcements about the next-generation RTX 4080 graphics card are still eagerly awaited.
What about the Nvidia Grace CPU Superchip and the next generation Hopper architecture? Don’t forget to share your feedback with us on the SDN Forum or in the comments!