Neural processing unit: Difference between revisions

Revision as of 04:05, 17 June 2016

Revision as of 04:07, 17 June 2016

Line 22:

<ref>{{cite web|title=nvidia introduces supercomputer for self driving cars|url=http://gas2.org/2016/01/06/nvidia-introduces-supercomputer-for-self-driving-cars/}}</ref>

=== Motivation ===

However at least a factor of 10 in efficiency<ref>{{cite web|title=google boosts machine learning with TPU|url=http://techreport.com/news/30155/google-boosts-machine-learning-with-its-tensor-processing-unit}}mentions 10x efficiency</ref>

HoweverWhilst atGPUs leastperform far better than CPUs for these tasks, a factor of 10 in efficiency<ref>{{cite web|title=google boosts machine learning with TPU|url=http://techreport.com/news/30155/google-boosts-machine-learning-with-its-tensor-processing-unit}}mentions 10x efficiency</ref>

can still be gained with a more specific design. The [[memory access pattern]] of AI calculations differs from graphics, with more a more predictable but deeper [[dataflow]] (benefiting from the ability to keep more temporary variables on-chip); GPUs by contrast devote silicon to efficiently dealing with non-linear [[memory access pattern#GATHER_SCATTER|gather-scatter]] addressing (between textures and frame buffers), and [[texture filtering]], as needed for 3D rendering.