Program
The times in the program below are Pacific Time (PT). The link to the conference’s platform is provided in the invitation email that follows your registration.
Conference Day 1: Wednesday, August 18, 2021
| Time | Title | Presenters |
|---|---|---|
| 08:45 | Introduction and Welcome | General Co-chair: Pavel Shamis |
| 09:00 |
Keynote 1: Compute Express Link (CXL) changing the game for Cloud Computing
[Abstract] Session chair: Dan Pitt (Palo Alto Innovation Advisors) |
Debendra Das Sharma, Intel [Bio]
|
| 10:00 | Diamond Sponsor Talk | Intel |
| 10:20 | Break | |
| 10:35 | Technical Paper Session A: Specialized Network Hardware | Session chair: Scott Levy (Sandia National Laboratories) |
| In-network reductions on a multi-dimensional HyperX | Kartik Lakhotia (USC), Fabrizio Petrini (Intel), Rajgopal Kannan (US Army Research Lab), and Viktor Prasanna (USC) | |
| Hardware SYN Attack Protection For High Performance Load Balancers | Reuven Cohen (Technion), Matty Kadosh (Nvidia), Alan Lo (Nvidia), and Qasem Sayah (Technion) | |
| 11:35 | Lunch | |
| 12:30 | Technical Paper Session B: Deep Learning | Session chair: Vishwanath Venkatesan (Nvidia) |
| Accelerating CPU-based Distributed DNN Training on Modern HPC Clusters using BlueField-2 DPUs | Arpan Jain, Nawras Alnaasan, Aamir Shafi, Hari Subramoni, and Dhabaleswar K Panda (The Ohio State University) | |
| Efficient Multi-Path NVLink/PCIe-Aware UCX based Collective Communication for Deep Learning | Yiltan Hassan Temucin, Amirhossein Sojoodi, Pedram Alizadeh and Ahmad Afsahi (Queen’s University) | |
| 13:30 | Gold Sponsor Talks | Session chair: Oscar Hernandez (Nvidia) |
|
Extending Network Transport Across Multiple Cloud Providers
[Abstract] |
Douglas Gourlay (Vice President/General Manager, Arista)
[Bio] |
|
| IBM | IBM | |
| 13:50 | Break | |
| 14:05 |
Panel – Von Neumann vs. Programmable Logic Smart NICs: Which is the Dead End?
[Abstract] |
Moderator: Scott Schweitzer (Achronix) Team Von Neumann: Mario Baldi (Pensando), Michael Kagan (NVIDIA), Cary Ussery (Marvell) Team FPGA: Jim Dworkin (Intel), Nick Ilyadis (Achronix), Ripduman Sohan (Xilinx) |
| 15:35 | Head Bubba Memorial Cocktail Reception |
Conference Day 2: Thursday, August 19, 2021
| Time | Title | Presenters |
|---|---|---|
| 09:00 |
Keynote 2: What is Next for Optics?
[Abstract] Session chair: Raj Channa (RBCCM) |
Andy Bechtolsheim, Arista [Bio]
|
| 10:00 | Platinum Sponsor Talks | Session chair: Luis E. Peña (Arm Research) |
|
Meet Your Kinetic Data Centre - Composable Disaggregated Infrastructure
[Abstract] |
Summit Puri (LIQID), David Nguyen (Dell)
[Bios] |
|
|
The Path to Resource Pooling from PCIe to CXL, Reducing Network Bottlenecks to Boot
[Abstract] |
Alan Benjamin (President and CEO, GigaIO)
[Bio] |
|
| 10:30 | Break | |
| 10:45 | Technical Paper Session C: Buses and Memory | Session chair: William (Whit) Schonbein (Sandia National Laboratories) |
| A low latency approach to delivering alternate protocols with coherency and memory semantics using PCI Express® 6.0 PHY at 64.0 GT/s | Debendra Das Sharma (Intel) | |
| Programmable FPGA-based Memory Controller | Sasindu Wijeratne, Sanket Pattnaik, Zhiyu Chen, Rajgopal Kannan and Viktor Prasanna (University of Southern California) | |
| 11:45 | Gold Sponsor Talks | Session chair: Raj Channa (RBCCM) |
| Algo-Logic
[Abstract] |
John W. Lockwood (CEO, Algo-Logic)
[Bio] |
|
| Tidalscale | Gary Smerdon (CEO, Tidalscale) | |
| 12:15 | Lunch | |
| 13:00 | Technical Paper Session D: Routing and Switches | Session chair: Ching-Hsiang Chu (Facebook) |
| Polarized routing: An efficient and versatile algorithm for large direct networks | Cristóbal Camarero, Carmen Martinez and Ramon Beivide (University of Cantabria) | |
| MBitTree: A fast and scalable packet classification for software switches | Jing Tan, Gaofeng Lv and Guanjie Qiao (NUDT) | |
| 14:00 | Silver Sponsor Talks | Session chair: Min Si (Facebook) |
| Ayar Labs | Shahab Ardalan
[Bio] |
|
| Lenovo | Lenovo | |
| 14:10 | Break | |
| 14:25 | Technical Paper Session E: Hot Topics | Session chair: Ryan Grant (Sandia National Laboratories) |
| The Future of Low Latency Memory | Thomas Coughlin (Coughlin Associates, Inc.) and Jim Handy (Objective Analysis) | |
| 14:45 | Closing Remarks | General Co-chair: Manjunath Gorentla Venkata |
Tutorials: Friday, August 20, 2021
| Time | Title | Presenters |
|---|---|---|
| 09:00 - 12:00 | Morning Sessions (in parallel) | |
| High Performance Machine Learning, Deep Learning, and Data Science
[Abstract] |
Dhabaleswar K. (DK) Panda, Arpan Jain, and Aamir Shafi (The Ohio State University)
[Bios] |
|
| Unified Communication X for Performance Portable Network Acceleration
[Abstract] |
Oscar Hernandez (Nvidia), Jeffrey Young (Georgia Institute of Technology), Matthew Baker (Oak Ridge National Laboratory), Gilad Shainer (Nvidia), Yossi Itigin (Nvidia)
[Bios] |
|
| 12:00 | Lunch | |
| 13:00 - 16:00 | Afternoon Session | |
| Introduction to Networking Technologies for High-Performance Computing
[Abstract] |
Dhabaleswar K. (DK) Panda and Hari Subramoni (The Ohio State University)
[Bios] |