Intel has revealed new details of upcoming high-performance artificial intelligence (AI) accelerators: Intel Nervana neural network processors, with the NNP-T for training and the NNP-I for inference.

Intel engineers also presented technical details on hybrid chip packaging technology, Intel Optane DC persistent memory and chiplet technology for optical I/O at yesterday’s Hot Chips 2019 event.

“To get to a future state of ‘AI everywhere,’ we’ll need to address the crush of data being generated and ensure enterprises are empowered to make efficient use of their data, processing it where it’s collected when it makes sense and making smarter use of their upstream resources,” says Naveen Rao, Intel vice-president and GM: artificial intelligence products group. “Data centres and the cloud need to have access to performant and scalable general purpose computing and specialized acceleration for complex AI applications.

“In this future vision of AI everywhere, a holistic approach is needed – from hardware to software to applications.”

Turning data into information and then into knowledge requires hardware architectures and complementary packaging, memory, storage and interconnect technologies that can evolve and support emerging and increasingly complex use cases and AI techniques.

Dedicated accelerators like the Intel Nervana NNPs are built from the ground up, with a focus on AI to provide customers the right intelligence at the right time.

Intel has presented the following:

* Intel Nervana NNP-T: Built from the ground up to train deep learning models at scale: Intel Nervana NNP-T (Neural Network Processor) pushes the boundaries of deep learning training. It is built to prioritise two key real-world considerations: training a network as fast as possible and doing it within a given power budget. This deep learning training processor is built with flexibility in mind, striking a balance among computing, communication and memory. While Intel Xeon Scalable processors bring AI-specific instructions and provide a great foundation for AI, the NNP-T is architected from scratch, building in features and requirements needed to solve for large models, without the overhead needed to support legacy technology. To account for future deep learning needs, the Intel Nervana NNP-T is built with flexibility and programmability so it can be tailored to accelerate a wide variety of workloads – both existing ones today and new ones that will emerge.

* Intel Nervana NNP-I: High-performing deep learning inference for major data centre workloads: Intel Nervana NNP-I is purpose-built specifically for inference and is designed to accelerate deep learning deployment at scale, introducing specialised leading-edge deep learning acceleration while leveraging Intel’s 10nm process technology with Ice Lake cores to offer industry-leading performance per watt across all major datacenter workloads. Additionally, the Intel Nervana NNP-I offers a high degree of programmability without compromising performance or power efficiency. As AI becomes pervasive across every workload, having a dedicated inference accelerator that is easy to program, has short latencies, has fast code porting and includes support for all major deep learning frameworks allows companies to harness the full potential of their data as actionable insights.

* Lakefield: Hybrid cores in a three-dimensional package: Lakefield introduces the industry’s first product with 3D stacking and IA hybrid computing architecture for a new class of mobile devices. Leveraging Intel’s latest 10nm process and Foveros advanced packaging technology, Lakefield achieves a dramatic reduction in standby power, core area and package height over previous generations of technology. With best-in-class computing performance and ultra-low thermal design power, new thin form-factor devices, 2 in 1s, and dual-display devices can operate always-on and always-connected at very low standby power.

* TeraPHY: An in-package optical I/O chiplet for high-bandwidth, low-power communication: Intel and Ayar Labs demonstrated the industry’s first integration of monolithic in-package optics (MIPO) with a high-performance system-on-chip (SOC). The Ayar Labs TeraPHY optical I/O chiplet is co-packaged with the Intel Stratix 10 FPGA using Intel Embedded Multi-die Interconnect Bridge (EMIB) technology, offering high-bandwidth, low-power data communication from the chip package with determinant latency for distances up to 2 km. This collaboration will enable new approaches to architecting computing systems for the next phase of Moore’s Law by removing the traditional performance, power and cost bottlenecks in moving data.

* Intel Optane DC persistent memory: Architecture and performance: Intel Optane DC persistent memory, now shipping in volume, is the first product in the memory/storage hierarchy’s entirely new tier called persistent memory. Based on Intel 3D XPoint technology and in a memory module form factor, it can deliver large capacity at near-memory speeds, latency in nanoseconds, while also natively delivering the persistence of storage. Details of the two operational modes (memory mode and app direct mode) as well performance examples show how this new tier can support a complete re-architecting of the data supply subsystem to enable faster and new workloads.