Intel has announced a new portfolio of chips to support the development of artificial intelligence in IT networks and data centres. Building on its acquisition of Nervana Systems earlier this year, the new Intel Nervana portfolio also includes Intel Xeon Phi processors and FPGAs (field-programmable gate arrays).
The next generation of the products is currently in development and Intel said it will test first silicon (code-named Lake Crest) in the first half of 2017. The products should be available to key customers later in the year. In addition, Intel announced a new product (code-named Knights Crest) on the roadmap that integrates best-in-class Intel Xeon processors with the technology from Nervana. Lake Crest is optimized specifically for neural networks to deliver high performance for deep learning and compute density with a high-bandwidth interconnect.
To further AI research and strategy, Intel announced the formation of the Intel Nervana AI board, which will feature industry and academic thought leaders to advise the company. Education provider Coursera will provide a series of AI online courses to the academic community to support further development, and Intel will start a Kaggle Competition in January jointly with Mobile ODT where the academic community can put their AI skills to the test to solve real-world socioeconomic problems, such as early detection for cervical cancer in developing countries through the use of AI for soft tissue imaging.
For developers, Intel started the Nervana AI Academy to provide access to training and tools, and introduced the Intel Nervana Graph Compiler to accelerate deep learning frameworks on Intel silicon. Intel also delivers APIs that extend across Intel’s portfolio of processors from edge to cloud, as well as embedded technologies such as Intel RealSense cameras and Movidius vision processing units.
Intel also announced at the presentation for industry partners that it expects the next generation of Xeon Phi processors (code-named Knights Mill) will deliver up to 4x better performance than the previous generation for deep learning and will be available in 2017.
In addition, Intel announced it is shipping a preliminary version of the next generation of Xeon processors (code-named Skylake) to select cloud service providers. With AVX-512, an integrated acceleration advancement, these Xeon processors will significantly boost the performance of inference for machine learning workloads, according to the company. Additional capabilities and configurations will be available when the platform family launches in mid-2017 to meet the full breadth of customer segments and requirements.
Finally, Intel announced an alliance with Google to drive the integration of new AI capabilities in multi-cloud environments and enterprise workloads. The collaboration includes optimising Intel architecture for the Kubernetes open source container management platform and the open source TensorFlow library for machine learning. In addition, the companies will work together to integrate Google Cloud Platform capabilities to Intel IoT edge devices for enterprise customers.
