Wednesday, August 17, 2016

Competition For NVIDIA: The Nervana Systems Chip That Will Let Intel Advance Its Deep Learning (INTC; NVDA)

NVDA has a $2 billion R&D head start but this is worth keeping an eye on.
In addition, with the stock down two days in a row and $1.32 below the most recent all-time high this is probably a good time to point out the drawdowns can be big in this type of investing. That said, NVDA is the class of the field. $61.98 down 62 cents.

From IEEE Spectrum:
Deep-learning artificial intelligence has mostly relied upon the general-purpose GPU hardware used in many other computing tasks. But Intel’s recent acquisition of the startup Nervana Systems will give the tech giant ownership of a specialized chip designed specifically for deep learning AI applications. That could give Intel a huge lead in the race to develop next-generation artificial intelligence capable of swiftly finding patterns in huge data sets and learning through imitation.

Nervana has leaned heavily on GPU hardware to build its own portfolio of deep-learning AI services for both companies and independent developers. But the startup has also been developing its own specialized deep-learning hardware, called Nervana Engine, that includes only the components necessary for running deep-learning algorithms and eliminates the extra components used for general-purpose GPU tasks. Nervana claims that when the Engine chip comes out in 2017,  it will deliver around 10 times as much computing power for deep learning as the best of today’s GPUs.

“Nervana’s AI expertise combined with Intel’s capabilities and huge market reach will allow us to realize our vision and create something truly special,” said Naveen Rao, CEO and cofounder of Nervana, in a blog post.

Software algorithms known as artificial neural networks are the heart of deep-learning AI. Such algorithms learn how to perform certain tasks through imitation and by observing correctly labeled examples as they sift through huge amounts of data. To accommodate deep learning’s voracious appetite for data, Nervana’s Engine hardware design includes High Bandwidth Memory technology that has stacked memory and densely packed data channels to swiftly move around large amounts of data.

The end result: 32 gigabytes of on-chip storage and up to 8 terabits per second of memory access speed. By comparison, the GDDR5 memory technology used in GPUs has memory access speeds of just 224 gigabits per second....MORE