The industry needs a leap in computing power—power that is scalable and open—to facilitate innovation across the entire developer community. Last week at the Intel Developer Forum (IDF), Intel previewed its AI platform for the industry:
• A commitment to open source frameworks (Caffe, Theano) and libraries (Intel® Math Kernel Library – Deep Learning Neural Network, Intel Deep Learning SDK) optimized for machine learning.
• The unveiling of the next generation of Intel® Xeon™ Phi processors, codenamed Knights Mill, with improved variable precision and high-capacity flexible memory.
• The announcement of plans to acquire Nervana Systems, uniting the Intel engineers who created the Intel® Xeon® and Xeon Phi processors with Nervana's machine learning experts to propel the AI industry to a level that would not have been possible otherwise.
AI is only just beginning, but the value and opportunity it will offer the world make it a key element for the data centers of the future. Intel will play a key role as a leader and catalyst for innovation to expand the reach of AI. While there has been much talk about the value of GPUs for machine learning, the reality is that less than 3% of all servers used for this purpose last year employed a GPU.
The scope, scale, and speed of the industry underscore the importance of broad and open access to AI innovations. The entire industry agrees on this point. Consider the following testimonials: Jing Wang of Baidu: “The increased memory offered by Intel Xeon Phi facilitates the efficient development of our models.” Professor Pedro Domingos of the University of Washington: “Intel is uniquely positioned to offer the hardware and architectures that will drive the open community essential for progress.”.
By Jason Waxman, corporate vice president of the Data Center Group and general manager of the Data Center Solutions Group at Intel.
