Monday, January 09, 2017

Cloud, Mobile: CPU vs SoC, FPGA, TPU, GPU...

Cloud and mobile are introducing significant changes in types of processors used.
Engineering optimization is not limited to ever higher number of transistors in chips resulting in Moore's law of doubling every 18-24 months. Alternative architectures are also used now.
This is a significant opportunity for businesses to leverage.
It is also a classic "numbers game", where huge sales are driving specialized optimization.

The moonshot that succeeded: How Bing and Azure are using an AI supercomputer in the cloud - Next at Microsoft
"FPGA can be reprogrammed at a moment’s notice to respond to new advances in artificial intelligence or meet another type of unexpected need in a datacenter."

Programmable chips turning Azure into a supercomputing powerhouse | Ars Technica

Microsoft's Internet Business Gets a New Kind of Processor | WIRED

Microsoft's Love For FPGA Accelerators May Be Contagious
"Intel’s $16.7B acquisition of FPGA leader Altera now appears to have been prescient; at the time of the acquisition they predicted that 30% of servers would require FPGA acceleration by 2020."
Developer Preview – EC2 Instances (F1) with Programmable Hardware | AWS Blog
"Field Programmable Gate Array, or FPGA. In contrast to a purpose-built chip which is designed with a single function in mind and then hard-wired to implement it, an FPGA is more flexible. It can be programmed in the field, after it has been plugged in to a socket on a PC board."

"Tensor Processing Unit (TPU), a custom ASIC we built specifically for machine learning — and tailored for TensorFlow. We’ve been running TPUs inside our data centers for more than a year, and have found them to deliver an order of magnitude better-optimized performance per watt for machine learning. This is roughly equivalent to fast-forwarding technology about seven years into the future (three generations of Moore’s Law)."

Applications that scale using GPU Compute (presentation from AzureCon)


How the SoC is Displacing the CPU – Medium
"This shift from mostly performance-centric chips to mostly power-constrained chips and the focus on lowering cost and increasing system-level integration has disrupted the traditional semiconductor landscape within just a decade and continues to reshape the industry today."


"Mobile SoCs are now very competitive to standalone CPUs. They offer matched/better compute performance, matched/better graphics within a lower power envelope and a 10X lower price-point. In addition, a single chip SoC solution is far better suited to a mobile formfactor than a multi-chip CPU solution (Source: Intel/Chipworks)"
Lancelot, an FPGA-based bitcoin mining board.

No comments: