Huan Xu·Feb 8, 2025NUMA architecture quick refresherThis blog’s goal is to offer a quick lookup reference to essential NUMA concepts, from allocation policies to hardware interconnects, to…
Huan Xu·Jan 19, 2024Sparse GEMM and Tensor Core’s Structured SparsityThe world of scientific computing and deep neural networks is abuzz with the term sparse general matrix multiplication (spGEMM). But what…
Huan Xu·Jul 8, 2023A Trip to Kernels: Understanding PyTorch’s Internal ArchitectureIf you’re here, you know that PyTorch is one of the most popular libraries among deep learning practitioners. It is highly efficient, and…A response icon3A response icon3
Huan Xu·Mar 29, 2023Stream OpenAI with FastAPI and Consuming it with React.jsProblem StatementA response icon3A response icon3
InAWS TipbyHuan Xu·Mar 29, 2023Warm a Vercel-hosted Next.js Website with Cloudflare WorkersWhen I hosted a Next.js-based ChatGPT clone on Vercel using the open-source project chatbot-ui for my parents in China, I noticed a…
