Servers: The Right Compute
Showing results for 
Search instead for 
Did you mean: 

Pioneering a new age of intelligence: highlights from GTC 2018



GTC 2018 empowered organizations to harness AI and deep learning, leveraging breakthrough solutions like the HPE Apollo 6500 Gen10 and NVIDIA GPUs to accelerate intelligence and expand the scope of human knowledge. 

3.jpgThe NVIDIA GPU Technology Conference (GTC) returned last month to Silicon Valley, CA. Every year, GTC invites thought leaders and industry experts to collaborate with top developers like NVIDIA and Hewlett Packard Enterprise (HPE). This flagship artificial intelligence (AI) and deep learning event is an opportunity to gain valuable insight and participate in hands-on training through hundreds of labs, presentations, and demos, as well as to discover how NVIDIA GPUs and revolutionary deep learning solutions are accelerating digital transformation and expanding the scope of human knowledge.

From March 26th–29th, industry leaders from around the globe convened to innovate, connect, and discover a new era of AI and deep learning capabilities. This year’s featured industries included financial services, government, healthcare, automotive and transportation, manufacturing, media and entertainment, and telecommunications. Organizations from each of these sectors—and others—gathered at GTC 2018 for a grand total of 7,100 attendees from 64 different countries. The conference also highlighted 571 sessions (including deep neural networks, voice recognition, autonomous vehicles, seismic processing, remote visualization, and much more), 150 exhibitors (showcasing in-memory compute, supercomputing, alternative energy sources, cluster management, computer-aided engineering, and other topics), and 138 various research posters.

Insights from GTC 2018 promise to transform the way organizations perform, operate, and compete, and HPE and NVIDIA are leading the charge.

Pioneering innovation at GTC

The focus of GTC is to aid AI and deep learning adoption and to inspire business innovation. Cognitive computing is the foundation for a new generation of intelligence, and equipping organizations with the actionable insight to predict and even prevent disease, build smart cities, leverage AI-based analytics, and extend deep learning applications. To tap into these capabilities, organizations are looking to the massively parallel processing power of high performance computing (HPC) systems from HPE accelerated by NVIDIA GPUs. Together, HPC and GPU-accelerated computing are empowering AI/deep learning tools and applications for a variety of game-changing uses.

Some remarkable innovations include accelerated data analytics, which enables enhanced data processing and interactive visualization for deeper, more accurate insights. GPU virtualization powered by NVIDIA Virtual GPU technology is helping organizations deploy a virtualized environment, increasing data security, and dramatically reducing time to insight, while reducing data center footprint and cost. These advancements not only support sophisticated forms of analysis—such as intelligent video analytics and predictive analytics—but they are also preparing us for the future. By combining the robust HPC performance of HPE solutions with the unmatched processing power of NVIDIA GPUs, these leaders are helping organizations create a smarter, safer world—at GTC and beyond. 

Among the informative presentations at the GTC conference, HPE’s Greg Schmidt discussed how to “Unlock AI Business Value with the HPE Apollo 6500 Gen10,” empowering attendees to expand their AI utilization in order to rapidly derive and employ intelligent insight. Leveraging enterprise platforms for accelerated computing with NVIDIA® NVLink™, organizations can train deep learning models in minute or hours—not days or weeks—which is a key operational advantage.

Natalia Vassilieva and Sergey Serebryakov of Hewlett Packard Labs presented the components of the HPE Deep Learning Cookbook, a revolutionary tool designed to guide organizations in selected the right hardware and software to execute their most demanding deep learning workloads. Complete with research benchmarks, open source solutions, and expert guidance, the HPE Deep Learning Cookbook is a vital first step on the journey to deep insight.

HPE’s Bill Dunmire wrapped up these session with a presentation on in-memory HPC. During this talk, attendees explored the HPE Superdome Flex, which is purpose-built platform designed for in-memory computing at unparalleled scale. Integrated with NVIDIA GPU technology, this breakthrough solution equips organizations to solve complex, data-heavy problems, supercharge data analytics, and seamlessly visualize data for faster discovery.

Fueling the next generation of intelligence

On March 21st, days before the GTC kick-off, HPE announced the new HPE Apollo 6500 Gen10 server. This groundbreaking platform is the next step in AI innovation, designed to perform lightning-quick analytics on escalating datasets in order to tackle the most compute-intensive HPC and deep learning workloads.

The HPE Apollo 6500 Gen10 solves highly complex problems at breakneck speeds by autonomously analyzing and learning from endless streams of data. The platform is expertly engineered for ease of use with reliability, availability, and performance in mind, to help organizations innovation quickly and continuously. Furthermore, the addition of eight NVIDIA® Tesla® V100 32GB GPUs—the world’s most advanced data center GPU—work to dramatically increase application performance, providing 2X the memory capacity to expedite HPC and deep learning workloads. These new powerhouse GPUs improve AI accuracy by minimizing the training time required by AI and deep learning models. By harnessing the robust HPC performance, massively parallel processing power, and ultra-scalability of the HPE Apollo 6500 Gen10, organizations can achieve superior levels of speed and productivity. 

GTC recap 2.jpgHPE’s and NVIDIA’s common goal is to help organizations realize unmatched data-driven intelligence. And with leading-edge solutions and services for HPC and AI/deep learning, we are committed to industrializing deep learning for faster intelligence. 

For organizations that require further assistance, HPE Pointnext services for AI are available to help you get started on the path to AI/deep learning. To learn more about industry-leading advancements that are accelerate intelligence, I invite you to follow me on Twitter at @pango. And be sure to visit @HPE_HPC and @NVIDIADC to explore the latest news and updates in HPC and GPU innovation.

0 Kudos
About the Author


Pankaj is building HPE’s Artificial Intelligence business. He is excited by the potential of AI to improve our lives, and believes HPE has a huge role to play. In his past life, he has been a computer science engineer, an entrepreneur, and a strategy consultant. Reach out to him to discuss everything AI @HPE.