NetApp just came back from NVIDIA’s GPU Technology Conference (GTC), where we had had a strong presence on the show floor. Several product announcements led up to the conference, and NetApp hosted a fun VIP customer event, a session presentation, an industry panel discussion, and numerous customer meetings.
A popular sight was NetApp demonstrating a three-DGX-2 NetApp® ONTAP® AI system in a modular, Dynamic Density Control (DDC) liquid-air-cooled cabinet from ScaleMatrix. ScaleMatrix is part of NVIDIA’s DGX-Ready Data Center program, which gives customers access to data center services through a network of colocation partners. The key storage metrics that NetApp tested and published with three DGX-2 servers are:
- 20GBps (buffering data) peak throughput requested
- 600µs average latency achieved
- 40% AFF A800 CPU utilization achieved
- >90% DGX-2 GPU utilization achieved
Also featured at the NetApp booth, in partnership with Cisco, was the FlexPod® AI solution. It includes Cisco UCS blade and rack servers, Cisco Nexus 9000 Series switches, Cisco UCS 6000 Series Fabric Interconnects, and NetApp® AFF A800 flash storage arrays.
Partner H20.ai was in the NetApp booth and highlighted machine learning in the cloud. H20.ai demonstrated the capabilities of its H2O Driverless AI platform in Amazon Web Services (AWS) with NetApp Cloud Volumes Service, NetApp Kubernetes Service, and NVIDIA GPU Cloud (NGC).
Other partners were represented in the NetApp booth, which featured Allegro.ai‘s demonstration of its computer vision platform on an NVIDIA DGX Station, Parabricks’ solutions for genomics, and NetApp HCI with Tesla M10 GPUs used in a healthcare visualization demonstration.
NetApp’s Santosh Rao and Sundar Ranganathan presented to a packed audience on “Architectural Considerations for Federating ML and DL Data Pipelines Across Edge, Core, and Cloud.” NetApp was joined by partners ePlus and Allegro.ai. Rao also participated in an industry ecosystem panel discussion with NVIDIA.
Only NetApp offers smart, powerful, and trusted solutions to help data-driven organizations unlock the potential of data science, from machine learning to deep learning. By combining cloud-connected flash with the best supercomputing solutions, NetApp accelerates AI outcomes at any scale. To learn more, visit NetApp.com/ai.