All Categories ¹⁶

Paper ²⁶

[Paper Note] H2O Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models

[Paper Note] Sparse Indexing Large Scale, Inline Deduplication Using Sampling and Locality

[Paper Note] IMPRESS an Importance-Informed Multi-Tier Prefix KV Storage System for Large Language Model Inference

[Paper Note] Attentionstore Cost-Effective Attention Reuse Across Multi-Turn Conversations in Large Language Model Serving

[Paper Note] Orca a Distributed Serving System for Transformer-Based Generative Models

Concurrency ³

【译】无锁算法导论

【译】内存屏障：软件黑客的硬件视角

【译】一种面向多核架构的无锁、缓存高效的共享环形缓冲区

Go ³

深入剖析 Go 语言运行时：IO 轮询器

【译】Go 语言数据竞争检测器

【译】Go 语言内存模型：2022-06-06 版

Distributed System ²

Redis Cluster 的设计

【论文阅读】f4 Facebook’s Warm BLOB Storage System

Linux ²

【译】拨开 Linux 平均负载的谜团

Tar 格式详解

Storage ²

KV Cache on SSD: Taking Twitter's Fatcache as an Example.

【论文阅读】f4 Facebook’s Warm BLOB Storage System

Vim ²

用 Vim 搭建开发环境

Vimspector: 最强 Vim 调试插件

Allocator ¹

致敬经典：K&R Allocator 内存分配器

C ¹

C/C++ 的严格别名规则

Cache ¹

KV Cache on SSD: Taking Twitter's Fatcache as an Example.

Command Line ¹

使用 Yadm 管理点文件

Computer Architecture ¹

【译】如何设计正确运行多进程程序的多核计算机

Lua ¹

Lua 语言简明教程

Network ¹

网络层：控制面板

OS ¹

UNIX 进程关系

瞎折腾 ¹

搭建家庭私有云