Nvidia unveiled a new processor, and it's the first of its kind with incredible new technologies. The Grace processor features artificial intelligence and high-performance computing work. This is Nvidia's first datacenter CPU with the purpose of running applications operating on a giant scale.
The announcement first came from a tweet by Nvidia introducing the Grace CPU with its ARM data center architecture designed to accelerate the most extensive High-Performance Computing (HPC) and Artificial Intelligence (AI) workloads. The Verge reported that the company has spent $40 billion on ARM technology to "create the world's premier computing company for the age of AI," and this new processor might be the first of their projects.
Meet the NVIDIA Grace #CPU, leveraging the flexibility of @Arm’s data center architecture and designed from the ground up to accelerate the largest HPC and AI workloads. #GTC21 https://t.co/PHDaxrfzQv pic.twitter.com/uck0akde3a— NVIDIA GTC (@NVIDIAGTC) April 12, 2021
Nvidia Grace CPU
The Grace processor is reported by Venturebeat to have combined energy-efficient ARM CPU cores, which is used for creating innovative-low power memory subsystems and deliver high performance. The chip is programmed to address the computing requirements for some of the most advanced applications available in the current technological age, including the recommender systems, processing, and AI supercomputing that analyze enormous data sets. These data sets also require a massive memory and ultra-fast computing performance.
The processor also uses the LPDDR5x memory subsystem, which is 10 times better in energy efficiency than the current DDR4 memory. The processor has a new architecture with unified cache coherence and single memory address space, combining system and HBM GPU memory.
Grace is also created to process workloads with over 1 trillion parameters like training the next-generation Neuro-Linguistic Programming (NLP) models. When running with Nvidia's GPUs, the Grace processor delivers up to 10 times faster performance than Nvidia's DGX-based systems and x86 CPUs.
Nvidia's ARM CPU Technology
Jensen Huang, CEO of Nvidia, proudly announced their technology in the recent GTC 2021 AI conference for innovators event. "Leading-edge AI and data science are pushing today's computer architecture beyond its limits - processing unthinkable amounts of data... using licensed ARM IP, Nvidia has designed Grace as a CPU specifically for giant-scale AI and HPC. Coupled with the GPU and DPU, Grace gives us the third foundational technology for computing and the ability to re-architect the datacenter to advance AI. Nvidia is now a three-chip company."
Grace processor is designed for many great things; however, Nvidia has yet to disclose the official release date of the processor. They have also withheld some of the unit's details, such as its number of transistors and estimated price range. All these rumors and leaked performance are subject to possible improvements until Nvidia makes their official release of the Grace Processor in the market.
It stands for the fact that the new Grace Processor is one of the most advanced technologies available today. It could also revolutionize new computing capabilities and push forward the advancement of other technological industries. Keep an eye out for more news relating to this fantastic new processor by bookmarking Nvidia's webpage and Twitter account.
Related Article: Nvidia GeForce GTX 1650 Restock: Latest Stock News, Price and Where to Buy