Pankaj Goyal, VP, AI Business & Data Center Strategy, Hewlett Packard Enterprise
The artificial intelligence (AI) revolution is quickly changing every industry, and modern data centers must be equipped to capitalize on these extraordinary new capabilities. Hewlett Packard Enterprise (HPE) and NVIDIA® are partnering to bring best-of-breed AI solutions to every customer, offering AI-integrated systems, services, and support capabilities to help all organizations seamlessly optimize their AI foundation, deliver differentiated outcomes, and gain competitive advantage.
High performance computing (HPC) has become key to solving many of the world’s grand challenges in the realms of science, industry, and engineering. However, traditional CPUs are increasingly failing to deliver the performance gains they used to, and the path forward for HPC data centers is GPU-accelerated computing. GPUs leverage a massively parallel architecture consisting of many cores that are designed to handle multiple tasks at once, while CPUs have fewer cores that are optimized for sequential serial processing. The ability to break computationally-intensive assignments into pieces and begin solving them simultaneously is the main reason why GPU computing is the ideal fit for AI, deep learning, and other highly analytical workloads.
NVIDIA’s industry-leading, advanced data center GPUs accelerate the most demanding HPC applications and power many of the world’s fastest supercomputers. HPE’s purpose-built deep learning platforms, advanced service capabilities, and solutions expertise help customers easily deploy and maximize the value of their deep learning systems. Together, this collaboration is delivering a portfolio of GPU-optimized solutions that enable data scientists, researchers, and engineers to uncover ground-breaking AI innovation and solve problems that were once thought impossible.
The new driving force behind AI
NVIDIA Volta™ is a brand new GPU architecture that is designed to bring AI to every industry. Featuring over 21 billion transistors, Volta is the most powerful GPU architecture that’s ever been made for computational and data science, delivering the performance of an AI supercomputer in a single GPU. The NVIDIA® Tesla® V100 accelerator incorporates the powerful new Volta GV100 GPU, which offers significant performance and scalability gains over the previous-generation Pascal™ architecture. NVIDIA Tesla V100 GPUs offer performance equivalent to 100 CPUs, delivering more than 120 teraflops per second (TFLOPS) of deep learning performance per GPU.
As the complexity and size of deep neural networks (DNNs) continue to grow, so does the demand for even higher compute performance and faster training times. NVIDIA Volta introduces Tensor Core technology, which provides the exponential leap in computing performance required to train large neural networks. Each Volta GPU is equipped with 640 Tensor Cores that deliver up to 12x higher peak TFLOPS for training and 6x higher peak TFLOPS for inference.
Accelerating real-time insights and intelligence
It was recently estimated that at least 80 percent of the recent advances in AI can be attributed to the availability of more computing power. However, not every server can efficiently handle the compute-intensive nature of AI environments, and traditional IT infrastructure solutions are falling short of delivering the required performance, bandwidth, and speed. To succeed with AI techniques like deep learning, organizations need optimal platforms and solutions expertise.
HPE’s portfolio of deep learning servers are purpose-built for the demands of AI and deep learning workloads and optimized with the latest GPU technologies from NVIDIA. HPE’s deep learning solutions include the most powerful NVIDIA GPUs in a high GPU-to-CPU configuration, and are designed with ease-of-use in mind to drive AI adoption across all organizations, regardless of size or industry.
For organizations lacking the integral requirements to implement deep learning, HPE offers a range of services to help businesses quickly overcome adoption challenges and get started with deep learning. HPE’s Deep Learning Cookbook is a book of “recipes” for deep learning workloads, and a comprehensive set of tools to guide customers through the process of selecting the optimal hardware/software environment for various deep learning tasks. The HPE GreenLake Flex Capacity service allows customers to consume their deep learning infrastructure using a flexible, on-demand consumption model, paying only for what they use in terms of servers, storage, networks, software, and services. And HPE Pointnext Advisory and Operational Services provide AI expertise for planning, deployment, and upgrades of the latest technologies, including NVIDIA Volta, Tensor Core, and Tesla V100 technologies.
Realize faster time-to-value for AI investments
A recent survey found that over three-quarters of today’s enterprises are investing in AI, but 91 percent expect to encounter barriers to AI realization, with lack of IT infrastructure (40 percent) and lack of access to talent (34 percent) among the top challenges. Whether you’re interested in AI but simply don’t have the expertise needed to get started, or you’re already leveraging AI techniques but want additional competitive advantage, HPE and NVIDIA offer AI expertise and services geared to maximizing performance.
Five global Centers of Excellence (CoEs) were recently set-up by HPE and NVIDIA to assist IT departments and data scientists in accelerating their deep learning applications and realizing better ROI from their deep learning deployments. These centers offer select customers access to the latest technology and expertise including the latest NVIDIA GPU technologies running on HPE systems. New HPE-NVIDIA deep learning tutorials and workshops are also available to help organizations train their teams quickly and build new competencies to master the latest AI techniques.
HPE and NVIDIA are helping enterprises demystify deep learning in order to realize incredible performance and innovation. To learn more about how you can build your AI foundation, please follow me on Twitter at @pango. And for the latest news and updates on HPC and GPU computing, please visit @HPE_HPC and @NVIDIADC.