lyan
CodeFun
  • Home
  • Archives
  • About
  • Links
EN 中文 日本語
  1. Home
  2. Archives

How uv Works: Python Environments, Dependency Resolution, and Caching

lyan 2026-07-15

A practical explanation of uv's internal workflow, from Python selection and PubGrub resolution to lockfiles, caching, virtual environments, and uv run.

Python uv Developer Tools Dependency Management
Read More

Cross-Attention in Multimodal AI: When One Stream Needs to Read Another

lyan 2026-07-05

A practical guide to cross-attention, from Q/K/V to Stable Diffusion’s text-conditioned U-Net and cross-modal dimension design.

AI Transformer Cross-Attention Multimodal Stable Diffusion
Read More

LFT and Topology Exports in InfiniBand: Diagnose the Fabric Without Expanding the Attack Surface

lyan 2026-06-26

A practical, evidence-based framework for collecting and handling InfiniBand LFT, topology, and UFM snapshot data without exposing the management plane.

Network NVIDIA RDMA InfiniBand HPC Security UFM
Read More

The Decision Engine Behind Fused Attention: How Transformer Engine Orchestrates cuDNN on Blackwell

lyan 2026-06-07

A source-grounded tour of Transformer Engine’s fused-attention dispatch: cuDNN Graph API, versioned backend eligibility, Blackwell paths, kernel-pipelining context, and deterministic training constraints.

GPU cuDNN Blackwell Transformer Engine Fused Attention Computing
Read More

Megatron-Bridge in Practice: A Production 101 Guide to the Nemotron, Megatron-Core, and Transformer Engine Stack

lyan 2026-05-25

A technical guide to the architecture, repository layout, parallelism-aware conversion mechanics, and an import→fine-tune→export workflow for NVIDIA’s Megatron stack.

NVIDIA Training MLOps Megatron Nemotron LLM
Read More
  • 1
  • Next

Search

Top Posts

  • Step into the world of ARM Server
  • Running Debian ARM64 on QEMU with UEFI

Hot Posts

  • A Comprehensive Guide to Multi-Node, Multi-GPU NVIDIA GPU Fabric Deployment
  • How uv Works: Python Environments, Dependency Resolution, and Caching
  • Cross-Attention in Multimodal AI: When One Stream Needs to Read Another
  • LFT and Topology Exports in InfiniBand: Diagnose the Fabric Without Expanding the Attack Surface
  • The Decision Engine Behind Fused Attention: How Transformer Engine Orchestrates cuDNN on Blackwell
  • Megatron-Bridge in Practice: A Production 101 Guide to the Nemotron, Megatron-Core, and Transformer Engine Stack
  • FMHA 101: From Transformer Attention to a CUDA Flash Attention Kernel
  • Enroot + Pyxis + SPANK: A Practical Architecture and Operations Guide for Slurm Containers
  • HPC-X, MPI, PMIx & NCCL: A Full Dissection of the GPU Cluster Communication Stack

Recent Posts

  • How uv Works: Python Environments, Dependency Resolution, and Caching
  • Cross-Attention in Multimodal AI: When One Stream Needs to Read Another
  • LFT and Topology Exports in InfiniBand: Diagnose the Fabric Without Expanding the Attack Surface
  • The Decision Engine Behind Fused Attention: How Transformer Engine Orchestrates cuDNN on Blackwell
  • Megatron-Bridge in Practice: A Production 101 Guide to the Nemotron, Megatron-Core, and Transformer Engine Stack
  • FMHA 101: From Transformer Attention to a CUDA Flash Attention Kernel
  • Enroot + Pyxis + SPANK: A Practical Architecture and Operations Guide for Slurm Containers
  • HPC-X, MPI, PMIx & NCCL: A Full Dissection of the GPU Cluster Communication Stack
  • Deep Dive into Server Memory Architecture: From DRAM Granules to NUMA Modes

Tag Cloud

Linux (61) XEN (29) Life (27) Memory (24) Virtualization (23) Diary (21) C/C++ (19) QEMU (17) test (15) CPU (14) NVIDIA (13) InfiniBand (11) AI (11) Interest (11) Algorithm (11) VisualStudio (11) HPC (10) GPU (10) KVM (10) Transformer (9) eBPF (8) MFC (8) Debian (7) Engine (6) MLOps (6) Security (6) TorchDynamo (6) PyTorch (6) CUDA (6) programming (6) Kernel (6) Network (6)
lyan
CodeFun

©2026 xryan.net.

Code all fun things in the world by Leon.

  • Links
  • About Us
  • Feedback
QR Code
Scan to follow