简体中文
Nvidia launched a new desktop GPU today that’s designed to bring massive amounts of power to people who are working on machine learning applications. The new Titan V card will provide customers with a Nvidia Volta chip that they can plug into a desktop computer.
According to a press release, the Titan V promises increased performance over its predecessor, the Pascal-based Titan X, while maintaining the same power requirements. The Titan V sports 110 teraflops of raw computing capability, which is nine times that of its predecessor.
It’s a chip that’s meant for machine learning researchers, developers, and data scientists who want to be able to build and test machine learning systems on desktop computers. GPUs are good at executing the types of calculations needed to train a machine learning system, and Nvidia’s Volta architecture is designed to provide further performance boosts at those tasks.
While it’s possible to pay for time on Nvidia Volta-backed systems running in the cloud through Amazon Web Services (and it seems all but certain that other providers will follow in the coming months), researchers who want the convenience of running those workloads on a local machine could use the Titan V.
Customers will also be able to use Nvidia GPU Cloud software to get further performance improvements on their workloads. NGC provides customers with software containers that include optimized versions of popular machine learning development frameworks that are supposed to provide an even more optimized environment for developing AI.
Those containers were only just made available for Titan series GPUs earlier this week, and that support will extend to the Titan V.
All of that power comes at a price, though. Customers will have to pay $2,999 for a single Titan V card, and they’ll only be sold online through the Nvidia Store to users in particular countries. The Titan V is available starting today, so developers can start building machine learning systems with them right away.
(VB)
