Nvidia’s new EGX Platform will bring AI computing to the edge, meaning it will take place at the sensors, before being sent to cloud-based data centers. The platform is intended to handle the immense growth in sensors embedded in Internet of Things devices and wearables, which has resulted in a massive amount of raw data. According to Nvidia senior director of enterprise and edge computing Justin Boitano, “we will soon hit a crossover point where there is more computing power at the edge than in data centers.”
VentureBeat reports that Nvidia EGX, which was shown at Computex in Taiwan, “will offer a new class of servers for instant AI on real-time streaming data in markets such as telecommunications, medicine, manufacturing, retail, and transportation … [with] low-latency transactions, or those with minimal time delays between interactions.”
“AI is one of the most important computing challenges of our time, but CPUs have not been able to keep up,” said Boitano. By 2025, an expected “150 billion machine sensors and IoT devices will stream continuous data that will need to be processed.”
Nvidia described the EGX Platform as a “hyperscale cloud-in-a-box,” which marries AI computing with “Mellanox security, networking, and storage technologies.” The scalable EGX goes from the Jetson Nano, “which in a few watts can provide half a trillion operations per second (TOPS) of processing for tasks such as image recognition,” to the high-end T4 servers that provide “more than 10,000 TOPS for real-time speech recognition and other real-time AI tasks.”
Nvidia EGX is also “optimized for running enterprise-grade Kubernetes container platforms such as Red Hat OpenShift.” The latter has been tested with Nvidia Cloud Stack … to simplify setup, provisioning, and management of GPU-accelerated infrastructure for running Tensor RT, Tensor RT Inference Server, and NGC Nvidia GPU Cloud registry.”
Nvidia EGX, Red Hat OpenShift and Nvidia Edge Stack combine to create “a state-of-the-art edge to cloud infrastructure.” For those enterprises interested in hybrid cloud computing, “Nvidia EGX-powered systems and devices can connect to cloud IoT services … [and] customers can remotely manage their service from AWS IoT Greengrass and Microsoft Azure IoT Edge.” BMW and Foxconn are among 40+ early adopters.