搜索、推荐、广告、用增等工业界实践文章收集（来源：知乎、Datafuntalk、技术公众号）

Visualize linux scheduling data in Trace Event format

Linux, Android and Chromium Performance Tools built using the Microsoft Performance Toolkit. Cross-platform .NET Core + WPA GUI

Oracle Linux container images

A blazingly fast multi-language serialization framework powered by JIT and zero-copy.

The Ignominous Profiler

universal serialization engine

flamegraphdiff

A C++ library for debugging purposes that can print any variable, even user-defined types.

Unofficial mirror of mob development branch

An implementation of a deep learning recommendation model (DLRM)

Less than 100 Kilobytes. Works for Android 5.1 and above

Omnitrace: Application Profiling, Tracing, and Analysis

eGPU on Apple Silicon, Trail for Fun! We're doing this for fun and just for taking challenge

A high-performance, extensible Python AOT compiler.

Display and control your Android device

An intel PT trace converter from `perf.data` to Fuchsia trace format.

Dynolog is a telemetry daemon for performance monitoring and tracing. It exports metrics from different components in the system like the linux kernel, CPU, disks, Intel PT, GPUs etc. Dynolog also …

common in-memory tensor structure

A lightweight parameter server interface

GWPSan: Sampling-Based Sanitizer Framework

High-Performance C++ Fundamental Library

纯c++的全平台llm加速库，支持python调用，chatglm-6B级模型单卡可达10000+token / s，支持glm, llama, moss基座，手机端流畅运行

Haclog(Happy Aync C log) is an extremely fast plain C logging library

A revised version of NanoLog which writes human readable log file, and is easier to use.

Measures the latency between CPU cores

A Moonscript dialect compiles to Lua.