| 7:45AM-8:30AM |
Breakfast/Registration
|
|
| 8:30AM-10:30AM |
Tutorial 1: Datacenter Racks
|
|
| |
How AI workloads shape rack system architecture
|
Frank Helms, AMD |
| |
Scaling fabric technologies
|
Darrin Vallis, AMD |
| |
Liquid cooling with Google Characteristics
|
Jorge Padilla, Google |
| |
Rearchitected power systems
|
Harsha Bojja, Microsoft |
| 10:30AM-10:55AM |
Coffee Break (1/2 hr)
|
|
| 10:55AM-12:25PM |
Tutorial 1: Datacenter Racks (cont)
|
|
| |
Case study: Nvidia GB200NVL72
|
John Norton, NVIDIA |
| |
Case study: Meta’s Catalina (NVL72)
|
William Arnold and Matt Bowman, Meta |
| |
Case study: Google TPU Rack
|
Pankaj Makhija, Google |
| 12:25PM-1:30PM |
Lunch (1 hr 5 min)
|
|
| 1:30PM-3:30PM |
Tutorial 2: AI Kernel Programming
|
|
| |
Introduction
|
Fredrik Kjolstad, Stanford |
| |
Decoupling Performance from Correctness with User-Schedulable Kernel Languages
|
Andrew Adams, Adobe Research |
| |
Pallas: Using JAX to write custom kernels for GPUs and TPUs
|
Sharad Vikram, Google |
| 3:30PM-4:00PM |
Coffee Break (1/2 hr)
|
|
| 4:00PM-5:30PM |
Tutorial 2: Kernel Programming (cont)
|
|
| |
Domain specific languages for GPU kernels and automatic kernel authoring with LLMs
|
Tri Dao, Princeton, Together AI |
| |
Programming techniques for implementing ML models on GPUs
|
Zihao Jia, CMU |
| 5:30PM-7:30PM |
Reception
|
|