Nvidia unveiled the latest RTX server configuration at GTC 2019, offering a huge leap in cloud rendering density, efficiency, and scalability.
The configuration comprises 1,280 Turing GPUs on 32 RTX blade servers, with each RTX blade server containing 40 GPUs into an 8U space. Multiple users with Nvidia GRID vGaming or container software can share it. Mellanox technology is used as the backbone storage and networking interconnect to deliver the apps and updates instantly to thousands of concurrent users.
The servers can be used in renderfarms and for other high-end computational needs. Moreover, they have been optimized for use by cloud gaming operators for rendering and streaming games at the performance levels of GeForce RTX 2080 GPUs to any client device. Think of it as virtually renting a high-end PC gaming system.
Cloud-rendered AR and VR applications are employing RTX servers using low-latency mobile-edge computing. With RTX servers on 5G Edge networks, users will have access to cloud gaming services like GeForce NOW and AR and VR applications on just about any device. And, Nvidia is working with HTC to bring cloud gaming and VR wirelessly into homes through the HTC 5G Hub, a 5G hotspot.
System makers such as Dell, HPE, Lenovo, ASUS, and Supermico have unveiled RTX servers for on-demand rendering and virtual workstation solutions from the data center.
Both 2U and 4U RTX servers are available from OEM partners today. The new 8U RTX blade servers will be available from Nvidia in Q3.