Servers & Systems: The Right Compute

Accelerate your AI with HPE ProLiant, Red Hat OCP and NVIDIA GPU collaboration

Artificial Intelligence is now more widely adopted than ever before. HPE, Red Hat, and NVIDIA have collaborated to produce a solution that drastically accelerates and simplifies the deployment and implementation of AI and machine learning solutions.

In this age of digital transformation, artificial intelligence (AI) and machine learning (ML) have the potential to transform every business in the same way as (and possibly more HPE-RedHat-NVIDIA-containers-blog.jpgthan) the internet has. From smarter products and services, to better business decisions and optimized business processes, to freeing up employees to be more creative and produce higher-value work—AI has the power to change almost everything. If your business doesn’t capitalize on this digital transformation, you risk being left behind.

Deploying AI solutions can be a complicated and difficult journey. That’s why HPE, Red Hat, and NVIDIA have removed the guess work of AI implementation by developing a scalable integrated solution. This solution makes it simple to acquire a complex solution (hardware configuration), removing the requirement of your business to engineer the hardware platform.

In addition, the solution leverages the industry-leading repository of NVIDIA NGC containers fine-tuned for specific AI use cases. As your business needs grow, worker nodes (HPE ProLiant DL380 with GPU) can be added to scale. The HPE ProLiant DL380 Server supports a dense GPU configuration of up to two double-wide or four single-wide GPUs for AI workload acceleration

Red Hat OpenShift increases AI adoption and removes the knowledge roadblocks to speed delivery of accelerated AI. It includes an enterprise-grade Linux® operating system, container runtime, networking, monitoring, container registry, authentication, and authorization solutions. These components are tested together for unified operations on a complete Kubernetes platform spanning every cloud.

Concerned about getting it right? HPE Pointnext Services can guide you through the stages of AI implementation by working with you to define a strategy that works with your needs. In addition, it’s about successful outcomes and Pointnext can help you define objectives, identify KPIs, and track ROI so you can get the most out of your AI initiatives.

Leverage 2019’s bestselling server: HPE ProLiant

The HPE ProLiant DL380 Gen10 Server is the industry’s most trusted compute platform and delivers the latest in workload-optimized performance, security, and management automation, backed by a comprehensive warranty. The HPE ProLiant DL380 Gen10 server is securely designed to reduce costs and complexity, featuring the First and Second Generation Intel® Xeon® Scalable Processor Family with up to a 60% performance gain (1) and 27% increase in cores (2), plus up to 3.0 TB of HPE 2933 MT/s DDR4 SmartMemory. It supports 12 Gb/s SAS, and up to 20 NVMe drives plus a broad range of compute options.

The HPE ProLiant DL360 Gen10 Server delivers security, agility, and flexibility without compromise. It also supports the Intel® Xeon® Scalable Processor Family with up to a 60% performance gain  and 27% increase in cores, along with 2933 MT/s HPE DDR4 SmartMemory supporting up to 3.0 TB with an increase in performance of up to 82% (3). 

How HPE ProLiant provides the intelligent foundation for hybrid cloud

HPE ProLiant offers a fresh, flexible, software-defined approach, delivering unmatched workload optimization, security, and automation—all available as a service.

  • Workload optimizedThe foundational intelligence of HPE ProLiant transforms IT with insights that help optimize workload performance, placement, and efficiency to deliver better outcomes faster. Key attributes are:
  • 360-degree security—Already the world’s most secure industry-standard server, HPE ProLiant provides an enhanced holistic, 360-degree approach to security that begins in the manufacturing supply chain and concludes with a safeguarded, end-of-life decommissioning.
  • Intelligent automation—The intelligence built into HPE ProLiant simplifies and automates management tasks, establishing a solid foundation for an open, hybrid cloud platform enabled by composability.
  • Delivered as a serviceHPE provides our customers choice in how you acquire and consume IT. Beyond traditional financing and leasing, HPE offers options that free trapped capital, accelerate infrastructure updates, and provide on-premises pay-per-use consumption.

Red Hat OpenShift Container Platform

OpenShift Container Platform is Red Hat’s enterprise-grade Kubernetes distribution that provides enterprises the ability to build, deploy, and manage container-based applications.  Red Hat OpenShift Container Platform provides enterprises with a full-featured Kubernetes environment that includes automated operations, cluster services, developer services, and application services to build platform as a service (PaaS) and containers as a service (CaaS) on-premises hybrid cloud solution. Red Hat OpenShift Container Platform provides integrated logging and metrics, authentication and scheduling, high availability, automated over the air updates, and an integrated application container registry.   

Accelerating AI initiatives with NVIDIA GPUs and NGC Containers

The NVIDIA T4 GPU accelerates diverse cloud workloads, including high-performance computing, deep learning training and inference, machine learning, data analytics, and graphics. Based on the new NVIDIA Turing™ architecture and packaged in an energy-efficient 70W, small PCIe form factor, T4 is optimized for mainstream computing environments and features multi-precision Turing Tensor Cores and new RT Cores. Combined with accelerated containerized software stacks from NGC, T4 delivers revolutionary performance at scale.

NGC containers deliver powerful and easy-to-deploy software proven to deliver the fastest results. By taking care of the plumbing, NGC enables users to focus on building lean models, producing optimal solutions and gathering faster insights.

Learn more about HPE ProLiant DL servers.

Download the HPE Reference Architecture for accelerated AI and machine learning on HPE ProLiant DL380 Gen10 and HPE ProLiant DL360 Gen10 servers with rapid deployment using Red Hat OpenShift Container Platform and NVIDIA T4 GPUs. 

(1) HPE measurements: Up to 60% performance increase of Intel Xeon Platinum vs. previous generation E5-2600 v4 average gains of STREAM, LINPACK, SPEC CPU 2006 & SPEC CPU2017 metrics on HPE servers comparing 2-socket Intel Xeon Platinum 8280 to E5-2699 v4 family processors. Any difference in system hardware or software design or configuration may affect actual performance. April 2019.

(2) Up to 27% cores increase of Intel Xeon Platinum vs. previous generation comparing 2-socket Intel Xeon Platinum 8280 (28 cores) to E5-2699 v4 (22 cores). Calculation 28 cores/22 cores= 1.27 = 27%. April 2019.

(3) HPE measurements conducted in April 2019. Percentage compares Gen10 vs Gen9; Gen10 = 12 channels x 2933 data rate x 8 bytes = 281 GB/sec; Gen9 = 8 channels x 2400 x 8 bytes = 154 GB/Sec; 281/154 = 1.82 or Gen10 is 82% greater bandwidth. Any difference in system hardware or software design or configuration may affect actual performance.

Tom Lattin
Hewlett Packard Enterprise

0 Kudos
About the Author


Meet HPE Blogger Tom Lattin. Tom is the Vice President and General Manager of the Hewlett Packard Enterprise Workload Optimized Solutions and Mass Market organization responsible for driving the P&L and market share position for HPE compute solutions and products in the mass market.