At Intel “Architecture Day,” top executives, architects and fellows revealed next-generation technologies and discussed progress on a strategy to power an expanding universe of data-intensive workloads for PCs and other smart consumer devices, high-speed networks, ubiquitous artificial intelligence (AI), specialized cloud data centers and autonomous vehicles.
Intel demonstrated a range of 10nm-based systems in development for PCs, data centres and networking, and previewed other technologies targeted at an expanded range of workloads.
Highlights:
- Industry-First 3D Stacking of Logic Chips
Intel demonstrated a new 3D packaging technology, “Foveros,” which brings the benefits of 3D stacking to enable logic-on-logic integration.
- New Sunny Cove CPU Architecture:
Intel introduced Sunny Cove, Intel’s next-generation CPU microarchitecture designed to increase performance per clock and power efficiency for general purpose computing tasks, includes new features to accelerate special purpose computing tasks like AI and cryptography.
Sunny Cove features:
-
- Enhanced micro architecture to execute more operations in parallel.
- New algorithms to reduce latency.
- Increased size of key buffers and caches to optimize data-centric workloads.
- Architectural extensions for specific use cases and algorithms.
- Next-Generation Graphics:
Intel unveiled new Gen11 integrated graphics with 64 enhanced execution units, more than double previous Intel Gen9 graphics (24 EUs), designed to break the 1 TFLOPS barrier. The new integrated graphics will be delivered in 10nm-based processors beginning in 2019.
- “One API” Software:
Intel announced the “One API” project to simplify the programming of diverse computing engines across CPU, GPU, FPGA, AI and other accelerators.
- Memory and Storage:
Intel Optane DC persistent memory is a new product that converges memory-like performance with the data persistence and large capacity of storage. The revolutionary technology brings more data closer to the CPU for faster processing of bigger data sets like those used in AI and large databases.
Deep Learning Reference Stack:
Intel is releasing the Deep Learning Reference Stack; an integrated, highly-performant open source stack optimized for Intel Xeon Scalable platforms.
- Operating System: Clear Linux OS is customizable to individual development needs, tuned for Intel platforms and specific use cases like deep learning;
- Orchestration: Kubernetes manages and orchestrates containerized applications for multi-node clusters with Intel platform awareness;Containers: Docker* containers and Kata* containers utilize Intel Virtualization Technology to help secure container;
- Libraries: Intel Math Kernel Library for Deep Neural Networks is Intel’s highly optimized math library for mathematical function performance;
- Runtimes: Python providing application and service execution runtime support is highly tuned and optimized for Intel architecture;
- Frameworks: TensorFlow is a leading deep learning and machine learning framework;
Deployment: KubeFlow is an open-source industry-driven deployment tool that provides a fast experience on Intel architecture, ease of installation and simple use.
