During the day-two of the keynote at IDF 2016, Intel briefly announced the next-gen Xeon Phi processor family. The new member, code-named as Knights Mill, will focus on artificial intelligence. Intel is working on this powerful Xeon Phi processor for servers and workstations.
The new processor is being aimed at deep machine learning applications and RAM will be stacked into its die. Featuring many cores, it will focus on high-performance calculations.
The Knights Mill will be available in 2017 as announced during the keynote. According to Intel, Knights Mill “is optimized for scale-out analytics implementations, and will include key enhancements for deep learning training. For today’s machine learning applications, the large memory size of the Intel Xeon Phi processor family helps customers like Baidu make it easier to train their models efficiently.”
Slide shown by Diane M. Bryant – VP & GM Data Center Group for Intel Corporation, during IDF 2016 keynote in San Francisco.
While the upcoming Mill is supposed to run in servers and workstations, the 10nm sister Knights Hill was announced in 2014 to be used in super computers for example US Department of Energy’s Aurora supercomputer.
