Hawk's upgrade consists of 24 HPE Apollo 6500 Gen10 Plus systems with 192 NVIDIA A100 GPUs based on the NVIDIA Ampere architecture. It offers 120 petaflops of AI performance and will provide a dramatic increase in the capabilities that HLRS provides its users for applications of machine learning, deep learning, high-performance data analytics, and artificial intelligence (AI).
With the integration of GPUs into Hawk's existing CPU infrastructure, computer scientists at HLRS also look forward to working with the center's users to develop new hybrid computing workflows that integrate traditional simulation methods with AI approaches.
Sep 20, 2021
Information for Users
Hawk
AI & Data Analytics
Systems & Infrastructure
See all news
Hawk Expansion Will Accelerate Research Combining Simulation and Artificial Intelligence
Technical Details about Hawk
