Nvidia reveals first AI platform for edge devices

by

in

In a bid to bring artificial intelligence to edge devices, Nvidia has declared its new Nvidia EGX platform which has the capacity to understand and act on information in real time without needing to send it into the cloud or information center first.

Up until today, AI calculations were so complicated that they required machines to run them in cloud information centers. This required a excellent deal of bandwidth also resulted in higher latency which is why Nvidia along with other hardware producers are attempting to bring AI into the edge of the system instead.

Ideally AI ought to be conducted where information is made whether that’s in stores, production centers or warehouses.

Nvidia closes in on Mellanox bargain in data center pushMichael Dell: Why edge computing could be the Upcoming big thingCompanies That Aren’t buying AI might risk falling behind

Nvidia’s entrance to the AI edge market makes a great deal of sense because of its graphics processing units are famous for how well they could manage AI together using the Tesla V100 used for deep learning and the Quadro GV100 producing realistic images in real time with ray tracing.

Nvidia EGX

One of the best aspects of Nvidia’s new EGX system is the simple fact it is scalable and can be utilised in a number of configurations.  

For instance, EGX begins with the tiny Nvidia Jetson Nano which can provide half hour trillion operations per minute (TOPS) using only a couple of watts for tasks like image recognition and crosses all the way into a stand of Nvidia T4 servers to deliver within 10,000 TOPS for real-time speech recognition and other real-time AI tasks.

Vice president and general manager of Enterprise and Edge Computing at Nvidia, Bob Pette clarified how the EGX system was designed to meet the needs of enterprise customers within an press release, saying:

“Enterprises need powerful computing at the edge to process their own oceans of raw information — flowing in from innumerable interactions with customers and centers — to create rapid, AI-enhanced decisions which may drive their enterprise. A scalable platform like NVIDIA EGX enables them to deploy systems to fulfill their requirements at the cloud or even both, on premises. ” 

The system also incorporates security, storage and networking technologies from Mellanox that Nvidia intends to get from the end of the year.

Nvidia is working using 13 server manufacturers to market its EGX platform including Cisco, Dell-EMC, HPE, Lenovo and many others. EGX can be compatible with AI software working on AWS and Microsoft Azure along with the stage can connect to IoT providers from both suppliers.

This is what you Want to know about AI

Via Silicon Angle

Buy Tickets for every event – Sports, Concerts, Festivals and more buy tickets

Discover more from Teslas Only

Subscribe now to keep reading and get access to the full archive.

Continue reading