Jiwei Network News, on the evening of March 22, Beijing time, NVIDIA GTC 2022 was held live on the official website of NVIDIA, and CEO Huang Jenxun explored and discussed cutting-edge technologies such as AI and Omniverse in a KEYNOTE SPEECH at GTC.
Topics at the conference include dozens of topics across industries, including accelerated computing and development tools, autonomous vehicles, computer vision/video analytics, cybersecurity, data science, game development, high-performance computing, inference, graphic design collaboration and digital twins, recommendation systems, robotics, and voice AI/NLP.
It is worth mentioning that in the GTC keynote speech, Mr. Jen-Hsun Huang, founder and CEO of NVIDIA, discussed cutting-edge technologies such as AI and Omniverse, as well as how NVIDIA's new computing platforms, innovative robots and virtual collaboration technology breakthroughs can empower all walks of life.
Specifically, first of all, of course, professional users pay more attention to the amber GPU, after a long period of rumors, NVIDIA this time finally officially announced this flagship GPU, the name is the same as before, for the Hopper architecture, the core code name is H100.
The first is the manufacturing process, which is different from the previous rumors, the H100 uses TSMC's 4nm process, but Nvidia calls it a 4N, which should be a 4nm custom version, and based on the CoWoS 2.5D wafer level package, with more than 80 billion transistors, Nvidia said that this is the most powerful GPU at present. In addition, NVIDIA also announced the specific parameters of the H100.
You know, the previous generation of A100 is still only 7nm architecture, before this conference came out, many voices in the outside world speculated that Lao Huang would use a 5nm process, and the result was released to give everyone a big surprise.
The most terrifying thing is that the CUDA core directly soared to 16896, directly reaching nearly 2.5 times that of the A100. (You should know that when you go from V100 to A100, the core is only a little bit more)
Floating-point operations and INT8/FP16/TF32/FP64 tensor operations, the performance is basically all improved by more than 3 times, compared to the previous two generations of architecture upgrades are also a little troublesome.
It is reported that this H100 is also the first GPU to support PCle 5.0 and HBM3, and the data processing speed has further soared - the memory bandwidth has reached 3TB/s.
In addition, the new GPUs of the Hopper architecture were grouped together with the name grace of the NVIDIA CPU, which became the name of the famous female computer scientist Grace Hopper, which was also used by Nvidia to name their superchip.
Grace Hopper invented the world's first compiler and COBOL language, known as the "First Lady of Computer Software Engineering".
In addition to GPUs, cpus, another pillar of NVIDIA's "three-core" strategy for data centers, have also made new progress.
At last year's GTC conference, Nvidia revealed the first dedicated data center CPU Grace, and this year, this year, this super CPU for AI infrastructure and high-performance computing was officially unveiled.
There are two versions of Grace this time, the first version is called Grace-Hopper, which is a "CPU + GPU combination, connected using NVLink technology, with a bandwidth of up to 900GB/s."
The version, called the Grace CPU Super Chip, consists of two CPU chip packages connected to each other via NVLink-C2C. The entire super chip has a total of 144 CPU cores (based on the ARMv9 instruction set), a cache capacity of 396MB, support for LPDDR5X ECC memory, and a bandwidth of up to 1TB/s.
This is called "the ideal CPU for AI factories" by Huang Jenxun.
According to reports, the Grace Hopper super chip module can directly connect between the chip between the CPU and the GPU, and its key driver technology is the NVLink interconnection between memory-consistent chips, and the speed of each link reaches 900GB/s.
The Grace CPU super chip can also be composed of two CPU chips. They are connected to each other via the high-speed, low-latency chip-to-chip interconnect technology NVLink-C2C.
It is difficult not to think of Apple's M1 Ultra, which seems to have progressed in inter-chip interconnect technology and made "assembly" a major trend in the chip industry.
The Grace Super Chip achieved 740 points in SPECrate2017_int_base benchmarks, 1.5 times (460 points) of the CPU currently mounted on the DGX A100.
Grace Super Chips can run on all NVIDIA computing platforms, either as stand-alone pure CPU systems or as GPU-accelerated servers, using NVLink-C2C technology to carry one to eight GPUs based on Hopper architecture.
The Grace CPU Super Chip can run all of NVIDIA's computing software stacks, combined with NVIDIA ConnectX-7 network cards, can be flexibly configured into the server, or as a standalone pure CPU system, or as a GPU acceleration server, can be paired with 1, 2, 4 or 8 Hopper-based GPUs.
In other words, users can maintain only one software stack to optimize performance for their specific workloads.
Huang Said grace super chips are expected to start supplying next year.
In terms of AI, NVIDIA has been able to provide full-stack AI, in addition to AI computing hardware, its AI software has also made a lot of progress.
Huang said that AI has fundamentally changed the ability of software and the way software is developed, and in the past decade, NVIDIA Accelerated Computing has achieved a million-fold acceleration in the field of AI.
At this gtccity conference, Nvidia released more than 60 updates to a series of libraries, tools and technologies for CUDA-X to accelerate research progress in quantum computing and 6G research, cybersecurity, genomics, drug discovery and other fields.
Nvidia will use its first AI digital twin, Earth-2, to address climate change challenges and create a Physics-ML model to simulate dynamic changes in global weather patterns.
NVIDIA believes that the industry can also increase the amount of data for AI training by simulating in a virtual environment, in other words, "doing big training in the metaverse" in other words.
For example, let AI intelligent driving "practice" in the metaverse, use the simulated data to create a semi-real environment, and add some environmental simulations that may suddenly fail:
In addition to digital twins, the production of digital assets is also a major consideration in the early construction phase of the metaverse.
In this regard, Nvidia has launched Omniverse Cloud, which can collaborate anytime, anywhere in the cloud.
Huang Believes that the first wave of AI learning is perception and reasoning, and the development direction of the next wave of AI is robotics, that is, the use of AI to plan actions. NVIDIA's Omniverse platform is also becoming an indispensable tool when manufacturing robotic software.
As a simulation engine for virtual worlds, the Omniverse platform follows the laws of physics and builds a realistic digital world that can be applied to remote collaboration between designers using different tools, as well as industrial digital twins.
Huang Believes that industrial digital twins need a purpose-built new type of computer, so NVIDIA has built OVX servers and OVX SuperPOD super clusters for industrial digital twins.
Venturebeat commented that "these cases give real meaning to the metaverse."
Finally, Huang Jenxun once again sent his virtual digital avatar "Doll Old Huang" Toy Jensen, and had a smooth real-time Q&A dialogue with the vividly animated doll.
(Proofreading/Arden)