HOT CHIPS 2021 Preview

As usual in August, it is HOT CHIPS. I always find this one of the most interesting conferences of the year. It gives a lot of insight into the specific products being presented, but also a feel for what technologies are being used for the most advanced designs. For example, the move towards chiplets and advanced packaging as the way of the future was one of the things that was hard to miss at HOT CHIPS a couple of years ago. The conference will be completely virtual again this year, as it was last year. As usual, the first day is a tutorial day, and then two days of presentations about the chips. This year, there is no separate fee for tutorials or for single days, there is simply one fee to attend and then you can attend anything. It is all available for replay afterward, too. I will attend and report on some of what gets presented, as I have done for several years.

All times are Pacific Daylight Time.

Tutorials, Sunday, August 22

If you had to name two big trends in semiconductor systems, then you might well pick machine learning and advanced packaging. Those are the subjects of the two tutorials on August 22 (yes, it's a Sunday, you have to suffer for knowledge).

• Tutorial 1: ML Performance and Real World Applications: Machine learning is a rich, varied, and rapidly evolving field. This tutorial will explore the applications, performance characteristics, and key challenges of many different unique workloads across training and inference. In particular, we will focus on hardware/software co-optimization for the industry-standard MLPerf benchmarks and selected applications and considerations at prominent cloud players.
• Tutorial 2: Advanced Packaging: This tutorial will discuss advanced packaging technologies that enable performance and density improvements. Descriptions of the technologies and how they are used in cutting-edge applications will be made by industry leaders in packaging and chip design.

Monday, August 23

9:00am-11:00am CPUs

• Intel Alder Lake CPU Architectures. Efraim Rotem, Intel
• AMD Next Generation “Zen 3” Core. Mark Evers, AMD
• The >5GHz next generation IBM Z processor chip. Christian Jacobi, IBM
• Next-Gen Intel Xeon CPU — Sapphire Rapids. Arijit Biswas and Sailesh Kottapalli, Intel

11:30am-12:30pm Academic Spinout Chips

• Mozart: Designing for Software Maturity and the Next Paradigm for Chip Architectures. Karu Sankaralingam, University of Wisconsin- Madison
• Morpheus II: A RISC-V Security Extension for Protecting Vulnerable Software and Hardware. Todd Austin, University of Michigan

12:30pm-1:30pm Keynote

• Does Artificial Intelligence Require Artificial Architects? Aart de Geus, co-CEO of Synopsys

2:30pm-4:00pm Infrastructure and Data Processors

• Arm Neoverse N2: Arm’s second-generation high-performance infrastructure CPUs and system products. Andrea Pellegrini, Arm
• NVIDIA DATA Center Processing Unit (DPU) Architecture. Idan Burstein, NVIDIA
• Intel’s Hyperscale-Ready SmartNIC for Infrastructure Processing. Bradley Burres, Intel

4:00pm-5:00pm Keynote

• Skydio Autonomy Engine: Enabling the Next Generation of Autonomous Flight. Abraham Bachrach, Skydio CTO

5:30pm-7:00pm Enabling Technologies

• Heterogeneous computing to enable the highest level of safety in automotive systems. Ramanujan Venkatadri, Infineon
• Architecting an Open RISC-V 5G and AI SoC for Next Generation 5G Open Radio Access Network. Sriram Rajagopal, EdgeQ
• Aquabolt-XL: Samsung HBM2-PIM with in-memory processing for machine learning accelerators. Jin Hyun Kim, Samsung Electronics

Tuesday, August 24

8:30am-10:00am ML Inference for the Cloud

• Accelerating ML Recommendation with over a Thousand RISC-V/Tensor Processors on Esperanto’s ET-SoC-1 Chip. David Ditzel, Esperanto Technologies
• AI Compute Chip from Enflame. Ryan Liu and Chuang Feng, Enflame Technology
• Qualcomm Cloud AI 100: 12 TOPs/W Scalable, High Performance and Low Latency Deep Learning Inference Accelerator. Karam Chatha, Qualcomm Inc

10:00am-11:00am Keynote

• Architectural Challenges: AI Chips, Decision Support, and High-Performance Computing. Dimitri Kusnezov, Deputy Under Secretary for AI and Technology, Department of Energy

11:30am-1:30pm ML and Computation Platforms

• Graphcore Colossus Mk2 IPU. Simon Knowles, Graphcore
• The Multi-Million Core, Multi-Wafer AI Cluster. Sean Lie, Cerebras Systems
• SambaNova SN10 RDU: Accelerating Software 2.0 with Dataflow. Raghu Prabhakar and Sumti Jairath, SambaNova Systems, Inc
• The Anton 3 ASIC: a Fire-Breathing Monster for Molecular Dynamics Simulations. J. Adam Butts and David E. Shaw, D.E. Shaw Research

2:30pm-4:30pm Graphics and Video

• Intel’s Ponte Vecchio GPU Architecture. David Blythe, Intel
• AMD RDNA 2 Graphics Architecture. Andrew Pomianowski, AMD
• Google’s Video Coding Unit (VCU) Accelerator. Aki Kuusela and Clint Smullen, Google
• Xilinx 7nm Edge Processors. Juanjo Noguera, Xilinx

5:00pm-7:00pm New Technologies

• Mojo Lens—AR Contact Lenses for Real People. Michael Wiemer and Renaldi Winoto, Mojo Vision
• World Largest Mobile Image Sensor with All Directional Phase Detection Auto Focus Function. Sukki Yoon, Samsung Electronics
• New Value Creation by Nano-Tactile Sensor Chip Exceeding our Fingertip Discrimination Ability. Hidekuni Takao, Kagawa University
• The IonQ Trapped Ion Quantum Computer Architecture. Christopher Monroe, IonQ, Inc  

Posters

There are also a couple of dozen "posters". I'm not quite sure how that works in a virtual-only conference. There are several posters on photonics and several on AI.

Learn More

All the details are on the HOT CHIPS website, including a link for registration. Early (cheaper) registration ends on August 6.

Sign up for Sunday Brunch, the weekly Breakfast Bytes email.