NVIDIA GTC Highlights Memory And Storage Importance For The Future

Enterprise Tech NVIDIA GTC Highlights Memory And Storage Importance For The Future Tom Coughlin Contributor Opinions expressed by Forbes Contributors are their own. Following New! Follow this author to stay notified about their latest stories. Got it! Sep 26, 2022, 11:02pm EDT | New! Click on the conversation bubble to join the conversation Got it! Share to Facebook Share to Twitter Share to Linkedin Exterior view of Nvidia’s headquarters in Silicon Valley.

getty At the NVIDIA Global Technology Conference (GTC) this September, Jensen Huang focused on three major corporate initiatives, NVIDIA RTX, NVIDIA AI and NVIDIA’s cloud Omniverse offerings. All of these initiatives drive the use of GPUs for big data analysis and generate a lot of information that needs to be retained in memory and long-term storage devices. NVIDIA announced its Ada Lovelace GPU architecture (below).

Targeted for gamers and content creators, it includes shader execution re-ordering to improve the parallel processing of imaging data in order to create simulated worlds. Ada Lovelace L40 data center GPUS in OVX servers are targeted for 3D video-based internet applications. NVIDIA also announce its LLM NeMo large language model and its automotive focused Thor with 2,000 teraflops of performance.

NVIDIAs Ada Lovelace GPU Image Captured by Tom Coughlin The game focused GeForce RTX 4080 12GB has 7,680 CUDA cores and 12GB of Micron GDDR6X memory, and with DLSS 3 is faster than the RTX 3090 Ti, the previous-generation flagship GPU. Micron’s DRAM received several mentions in NVIDIA announcements. The company’s Grace CPU and Grace Hopper Superchip are geared for high performance enterprise computing applications.

NVIDIA’s Grace CPU uses LPDDR5X memory to provide 50% more bandwidth, while using an eighth of the power per gigabyte of traditional DDR5 memory subsystems. NVIDIA also says that its NVLink can be used to provide pools of GPU memory and NVLink-C2C requires just 1. 3 picojoule per bit transferred, 5X the energy efficiency of PCIe Gen 5.

NVIDIA says that the result is that recommendation engine with 4X more performance and greater efficiency using Grace Hopper than using Hopper with traditional CPUs as shown below. MORE FOR YOU The 5 Biggest Technology Trends In 2022 ‘Enthusiastic Entrepreneurs’: Pre-IPO Statements On Profitability Prove To Be Larger Than Real Life The 7 Biggest Artificial Intelligence (AI) Trends In 2022 Improved Training Performance with Grace Hopper Superchip Image from NVIDIA GTC NVIDIA’s Omniverse Cloud is an infrastructure-as-a-service that connects Omniverse applications running in the cloud, on premises or on a device. Omniverse is NVIDIAs name for virtual worlds.

Omniverse is reported in use with retailers like Lowe’s and for 4G and 5G digital twins for Charter. GM is also creating a digital twin of its Michigan Design Studio in Omniverse, where designers, engineers and marketers can collaborate. Huang also described NVIDIA’s second-generation processor for robotics, Orin.

He announced the Jetson Orin Nano, a tiny robot computer that is 80X faster than the previous Jetson Nano product. At the GTC, DDN announced its next generation of reference architectures for NVIDIA DGX BasePOD at the September NVIDIA GTC. The company says that DDN’s A 3 I AI400X2 all-NVME appliance is supporting thousands of NVIDIA DGX systems around the world for AI and other data analytics.

DNN says it has deployed more than 2. 5 Exabytes of AI storage in 2021. The system can be configured as all flash or hybrid (with HDDs).

According to DNN, “DDN’s A3I powered by NVIDIA DGX BasePOD is an evolution of what was previously known as NVIDIA DGX POD configurations. These new configurations increase flexibility and customer choice, while maintaining the users’ abilities to start deployments at small scales and grow their DGX cluster over time. Additionally, DDN is collaborating with NVIDIA on vertical-specific DGX BasePOD solutions tailored specifically to financial services, healthcare and life sciences, and natural language processing.

Customers using these DGX BasePOD configurations will not only get integrated deployment and management, but also software tools including the NVIDIA AI Enterprise software suite, tuned for their specific applications in order to speed up developer success. ” The NVIDIA GTC brought out new GPU and other hardware that will accelerate data processing in gaming, enterprise computing, cloud computing, automotive and media creation and require large amounts of memory and storage to support the creation, analysis and use of this data. Follow me on Twitter or LinkedIn .

Check out my website . Tom Coughlin Editorial Standards Print Reprints & Permissions.