Deploy AI Training Faster: Putting Inferencing to Work

New NVIDIA A10 and A30 GPUs available in next-generation Dell EMC PowerEdge servers.

Enterprise artificial intelligence (AI) has exploded in recent years, with use cases ranging from customer engagement to forecasting to product development and more. With the introduction of these increasingly complex, demanding and diverse applications, enterprises have a real need for simple, powerful solutions that can provide the high-performance computing (HPC) power that will drive the future of the enterprise.

Dell Technologies has worked with NVIDIA® for years, investing in solutions that incorporate leading-edge GPUs and software from NVIDIA. That legacy continues, with the all-new NVIDIA A30 and A10 GPUs. These enterprise-class GPUs combine high performance and low power consumption, making them ideal for a broad range of AI, graphics-intensive and traditional enterprise workloads.

  • The NVIDIA A30 Tensor Core GPU delivers versatile performance for industry-standard Dell EMC PowerEdge servers. According to NVIDIA, it supports a broad range of AI inference workloads as well as mainstream enterprise applications. These might include recommender systems, conversational AI, computer vision and HPC workloads. It combines fast memory bandwidth and low, 165W power consumption in a PCIe form factor.
  • The NVIDIA A10 Tensor Core GPU is a single-slot 150W GPU that supports mixed AI and graphics workloads on a common infrastructure. It can accelerate deep learning inference, interactive rendering, computer-aided design, virtual desktop infrastructure (VDI), cloud gaming and more.

Launching summer 2021, these new GPUs are optimal to deliver business outcomes with the latest accelerated servers in the Dell EMC PowerEdge portfolio:

  • The PowerEdge R750xa, R750 and R7525 enterprise servers boost workloads to new performance heights with GPU and accelerator support for demanding workloads including enterprise AI. With its enhanced, air-cooled design and supporting up to 4 NVIDIA double-width GPUs, the PowerEdge R750xa server is purpose-built for optimal performance for the entire spectrum of HPC, AI-ML/DL training and inferencing workloads.
  • The new GPUs are also a great addition to our Dell EMC XE2420 and XR12 edge servers, as requirements for accelerated edge application and data analytics continue to evolve in 5G and enterprise core.

The latest generation of PowerEdge servers feature PCIe Gen 4.0, which doubles throughput performance over the previous generation. With PCIe Gen 4.0, servers can host more NVIDIA GPUs and high-speed NVIDIA networking to allow for greater compute and network acceleration in the same form factor. Combined with the new PowerEdge lineup, these breakthrough NVIDIA accelerators enable companies of all sizes to maximize AI training models for several applications and deliver higher inferencing throughput.

“NVIDIA and our partners are focusing not only on delivering world-leading performance for AI, but on democratizing AI with a coming wave of enterprise servers powered by our new A30 and A10 GPUs,” says Ian Buck, general manager and vice president of Accelerated Computing at NVIDIA.

With this announcement, NVIDIA and Dell Technologies are furthering their partnership, delivering next-generation hardware and software to accelerate AI, HPC and other enterprise workloads. We’re working closely to help our customers capitalize on the promise of enterprise AI and HPC.

To learn more, come visit us at ISC 2021.

Robert Hartman

About the Author: Robert Hartman

Robert Hartman is a Senior Product Marketing Manager for PowerEdge FX servers and other products in the Dell EMC PowerEdge family. He has always had a fascination with technology and how it can add value to people, businesses and ongoing trends to improve productivity. His perspective is to communicate how customers and business can leverage and maximize their investments and increase their results. Robert recently joined the Dell EMC team a second time to drive server marketing, following joining in 2010 in the Networking business unit. He has over 20 years of marketing experience in semiconductors, networking and servers, with many more years in technology.