Zhongquan Zhou

Shanghai, China

Home
Categories
Tags
Repository
Books
Links

Board

Welcome to my personal blog! Feel free to explore and share your thoughts!

Contact

Email zhouzhq2021@sjtu.edu.cn
GitHub github.com/zhouzhq2021
Location Shanghai, China

Categories

AI Infrastructure7
Cloud Computing and Big Data4

Tags

AI Infra Engineering Series7
CUDA Performance1
Capacity Planning1
Ceph1
Distributed Training1
FlashAttention1
GPU Memory1
Hadoop1
Kubernetes1
LLM Inference1
LLM Serving1
OpenStack1

Tag Cloud

AI Infra Engineering Series CUDA Performance Capacity Planning Ceph Distributed Training FlashAttention GPU Memory Hadoop Kubernetes LLM Inference LLM Serving OpenStack

Archive

July 20267
June 20241
May 20242
April 20241

Recent Posts

AI Infrastructure

70B 大模型推理服务的容量规划与性能设计

2026-07-30
AI Infrastructure

Softmax 数值稳定性与 IO-Aware Attention：从在线归一化到 FlashAttention

2026-07-28
AI Infrastructure

分布式训练的通信模型：从集合通信到多维并行

2026-07-26
AI Infrastructure

LLM 推理引擎的内存管理与调度：PagedAttention、Prefix Cache 与 Continuous Batching

2026-07-24
AI Infrastructure

GPU 性能分析方法：Roofline 模型、CUDA 内存层次与 Nsight

2026-07-22

Books

Personal collection of books recommended to everyone [Douban]

在读
想读
已读

正在加载...

© 2026 Zhongquan Zhou

Theme by cofess base on pure.