Allard Hendriksen

Allard Hendriksen is a developer technology engineer at NVIDIA with an expertise in modern GPU memory subsystems. He has worked on projects accelerating key workloads for customers, is the lead developer of cuda::ptx, and has presented on optimizing bandwidth and minimizing latency at GTC and other venues. He holds a PhD in applied mathematics from Leiden University and the CWI in Amsterdam.
Avatar photo

Posts by Allard Hendriksen

Data Center / Cloud

Accelerating Data Processing with NVIDIA Multi-Instance GPU and NUMA Node Localization

NVIDIA flagship data center GPUs in the NVIDIA Ampere, NVIDIA Hopper, and NVIDIA Blackwell families all feature non-uniform memory access (NUMA) behaviors, but... 12 MIN READ