HC32 Archive

Zipfle of full conference proceedings! (~190MB)

Tutorials: Sunday, August 16, 2020

Time (PDT) Title Presenters
8:30AM-11:00AM Tutorial 1A: Machine Learning Scale Out
Chair: Paulius Micikevicius
 
  Fundamentals of Scaling Out DL Training Paulius Micikevicius, NVIDIA
  Scale Out Systems – DGX A100 SuperPOD Michael Houston, NVIDIA
  Scale Out Systems – Google TPU Pod Sameer Kumar and Dehao Chen, Google
  Scale Out Systems – Cerebras System Natalia Vassilieva, Cerebras
11:00AM-11:30AM BREAK  
11:30AM-1:00PM Tutorial 1B: Machine Learning Scale Out
Chair: Paulius Micikevicius
 
  Scale Out Training Experiences – Megatron Language Model Mohammad Shoeybi, NVIDIA
  Scale Out Training Experiences – Distributed Parameter Server for Massive Recommender System Weijie Zhao, Baidu
  Scale Out Training Experiences – GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding Zhifeng Chen and Noam Shazeer, Google
1:00PM-2:00PM BREAK  
2:00PM-3:15PM Tutorial 2A: Quantum Computing
Chair: Misha Smelyanskiy
 
  Introduction Misha Smelyanskiy, Facebook
  Quantum Supremacy Using a Programmable Superconducting Processor John Martinis, UCSB
  Applications and Challenges with Near-term Quantum Hardware Jarrod McClean, Google
3:15PM-3:45PM BREAK  
3:45PM-5:15PM Tutorial 2B: Quantum Computing
Chair: Misha Smelyanskiy
 
  Underneath the Hood of a Superconducting Qubit Quantum Computer Matthias Steffen and Oliver Dial, IBM
  Towards a Large-scale Quantum Computer Using Silicon Spin Qubits James S. Clarke, Intel
  If Only We Could Control Them: Challenges and Solutions in Scaling the Control Interface of a Quantum Computer David Reilly, Microsoft

Conference Day 1: Monday, August 17, 2020

Time (PDT) Title Presenters
9:15AM-9:30AM Opening Remarks  
  Opening Remarks Alisa Scherer, General Chair, Hot Chips 32
9:30AM-11:30AM Server Processors
Chair: Pradeep Dubey
 
  Next Generation Intel Xeon(R) Scalable Server Processor: Icelake-SP Irma Esmer Papazian, Intel
  IBM’s POWER10 Processor William Starke and Brian W Thompto, IBM
  Marvell ThunderX3™ Next Generation Arm-Based Server Processor Rabin Sugumar, Marvell
  The 5.2GHz IBM z15 Processor Anthony Saporito
11:30AM-12:00PM BREAK  
12:00PM-1:00PM Mobile Processors
Chair: Fred Weber
 
  AMD Next Generation 7nm RyzenTM 4000 APU Sonu Arora, AMD
  Inside Tiger Lake: Intel’s Next Generation Mobile Client CPU Xavier Vera, Intel
1:00 PM-2:00 PM BREAK  
2:00PM-3:00PM Keynote
Chair: Priyanka Raina
 
  Intel Keynote Raja M. Koduri, Senior Vice President, Chief Architect, and General Manager of Architecture, Graphics, and Software, Intel
3:00PM-5:00PM Edge Computing and Sensing  
  Xuantie-910: Innovating Cloud and Edge Computing by RISC-V Yu Pu, Alibaba
  A technical overview of the Arm Cortex-M55 and Ethos-U55: ARM’s most capable processors for endpoint AI Allan Skillman and Tomas Edso, ARM
  PGMA: A Scalable Bayesian Inference Accelerator for Unsupervised Learning Glenn G. Ko, Harvard University
4:30PM-5:00PM BREAK  
5:00PM-6:30PM GPUs and Gaming Architectures
Chair: John Sell
 
  NVIDIA’s A100 GPU: Performance and Innovation for GPU Computing Jack Choquette and Wishwesh Gandhi, NVIDIA
  The Xe GPU Architecture David Blythe, Intel
  Xbox Series X System Architecture Jeff Andrews and Mark Grossman, Microsoft

Conference Day 2: Tuesday, August 18, 2020

Time (PDT) Title Presenters
8:30AM-10:00AM FPGAs and Reconfigurable Architectures
Chair: Ralph Wittig
 
  Agilex Generation of Intel FPGAs Ilya Ganusov and Mahesh Iyer, Intel
  Xilinx Versal Premium Series Martin Voogel, Yohan Frans and Matt Ouellette, Xilinx
  Compute substrate for software 2.0 Ljubisa Bajic and Jasmina Vasilijevic, Tenstorrent
10:00AM-10:30AM BREAK  
10:30AM-12:30PM Networking and Distributed Systems
Chair: Yuan Xie
 
  Tofino2 – A 12.9Tbps Programmable Ethernet Switch Anurag Agrawal and Changhoon Kim, Intel/Barefoot
  Pensando Distributed Services Architecture Francis Matus, Pensando
  The DPU: A New Category of Microprocessor Pradeep Sindhu, Fungible
  High-density Multi-tenant Bare-metal Cloud with Memory Expansion SoC and Power Management Justin Song and Xiantao Zhang, Alibaba
12:30PM-1:30PM BREAK  
1:30PM-2:30PM Keynote
Chair: Cliff Young
 
  DeepMind Keynote Dan Belov, Distinguished Engineer, DeepMind
2:30PM-4:00PM ML Training
Chair: Ron Diamant
 
  Google’s Training Chips Revealed: TPUv2 and TPUv3 Thomas Norrie and Nishant Patil, Google
  Software Co-design for the First Wafer-Scale Processor (and Beyond) Sean Lie, Cerebras
  Manticore: A 4096-core RISC-V Chiplet Architecture for Ultra-efficient Floating-point Computing Florian Zaruba, ETH Zurich
4:00PM-4:30PM BREAK  
4:30PM-6:30PM ML Inference
Chair: Forest Baskett
 
  Baidu Kunlun – An AI Processor for Diversified Workloads Jian Ouyang, Baidu
  Hanguang 800 NPU – The Ultimate AI Inference Solution for Data Centers Yang Jiao, Alibaba
  Silicon Photonics for Artificial Intelligence Acceleration Carl Ramey, Lightmatter
6:30PM-6:40PM Closing Remarks  
  Closing Remarks Alisa Scherer, General Chair, Hot Chips 32

Posters

Title Authors & Affiliation
A Fully Integrated Genetic Variant Discovery SoC for Next-Generation Sequencing Yi-Chung Wu, Yen-Lung Chen, Chung-Hsuan Yang, Chao-Hsi Lee, Chao-Yang Yu, Nian-Shyang Chang, Ling-Chien Chen, Jia-Rong Chang, Chun-Pin Lin, Hung-Lieh Chen, Chi-Shi Chen, Jui-Hung Hung and Chia-Hsiang Yang; National Taiwan University
Chipyard – Integrated Design, Simulation, and Implementation of Custom RISC-V SoCs Alon Amid, Abraham Gonzalez, David Biancolin, Daniel Grubb, Sagar Karandikar, Harrison Liew, Albert Magyar, Howard Mao, Albert Ou, Nathan Pemberton, Paul Rigge, Colin Schmidt, John Wright, Jerry Zhao, Yakun Sophia Shao, Krste Asanovic and Borivoje Nikolic; UC Berkeley
A Mutual Information Accelerator for Autonomous Robot Exploration Peter Li, Sertac Karaman and Vivienne Sze; MIT
GANPU: A Versatile Many-Core Processor for Training GAN on Mobile Devices with Speculative Dual-Sparsity Exploitation Sanghoon Kang, Donghyeon Han, Juhyoung Lee, Dongseok Im, Sangyeob Kim, Soyeon Kim, Junha Ryu and Hoi-Jun Yoo; KAIST
Vortex: An Open Source Reconfigurable RISC-V GPGPU Accelerator for Architecture Research Fares Elsabbagh, Blaise Tine, Apurve Chawda, Will Gulian, Yaotian Feng, Da Eun Shim, Priyadarshini Roshan, Ethan Lyons, Lingjun Zhu, Sung Kyu Lim and Hyesoon Kim; Georgia Tech
Light-in-the-loop: Using a Photonics Co-processor for Scalable Training of Neural Networks Julien Launay, Iacopo Poli, Kilian Müller, Igor Carron, Laurent Daudet, Florent Krzakala and Sylvain Gigan; LightOn
ELearn: Edge Learning Processor with Bidirectional Speculation and Sparsity & Mixed-Precision Aware Dataflow Parallelism Reconfiguration Fengbin Tu, Weiwei Wu, Yang Wang, Hongjiang Chen, Feng Xiong, Man Shi, Ning Li, Jinyi Deng, Tianbao Chen, Leibo Liu, Shaojun Wei and Shouyi Yin; Tsinghua University
Fully Autonomous Mixed Signal SoC Design & Layout Generation Platform Tutu Ajayi, Yaswanth K Cherivirala, Kyumin Kwon, Sumanth Kamineni, Mehdi Saligane, Morteza Fayazi, Shourya Gupta, Chien-Hen Chen, Dennis Sylvester, David Blaauw, Ronald Dreslinski Jr, Benton Calhoun and David Wentzloff; University of Michigan
OmniCIM: A Sparsity-Aware Computing-in-Memory based Processor for Accelerating Arbitrary Quantized Neural Networks Jianxun Yang, Yuyao Kong, Yiqi Wang, Zhao Zhang, Jing Zhou, Zhuangzhi Liu, Yonggang Liu, Chenfu Guo, Te Hu, Congcong Li, Leibo Liu, Jun Yang, Shaojun Wei and Shouyi Yin; Tsinghua University
SAINT-S – 3D SRAM Stacking Solution Based on TSV Technology Kyoungsun Cho, Jinhong Park, Billy Koo, Sunkyoung Seo, Yoonjae Hwang, Sungcheol Park and Mijung Noh; Samsung
A 3.2Gbps/pin HBM2E PHY with Low Power I/O and Enhanced Training Scheme for 2.5D System-in-Package Solutions Sangyun Hwang, Kwanyeob Chae, Taekyung Yeo, Sangsoo Park, Won Lee, Shinyoung Lee, Soo-Min Lee, Kihwan Seong, Eunkyoung Ha, Eunsu Kim, Jihun Oh, Kyoung-Hoi Koo, Sanghune Park, Jongshin Shin; Samsung