Blog
DeepSeek 3FS Performance Journal
A series analyzing the filesystem (3FS) DeepSeek uses for LLM training and inference
- Introduces 3FS, its claims and high‑level design
- Reviews DeepSeek’s published graphs against theoretical limits
- Builds and evaluates network and storage microbenchmarks ~ asking how does it perform in practice
Highlighted Posts
- Describes the process of building an end‑to‑end pipeline for scraping/decoding League of Legends replays. Datasets: 700K+ and 700K+ (S12).
- Intercepts OpenGL pipelines and re‑renders them with NVIDIA ray tracing (and its many various difficulties)
- Measuring cloud storage and network performance for LLM training pipelines
2025
2022
| Dec 05, 2022 | Raysterizer |
|---|
2021
| Sep 06, 2021 | Machine Learning with League of Legends - Minimap Detection (Part 2) |
|---|---|
| Sep 05, 2021 | Machine Learning with League of Legends - Minimap Detection (Part 1) |
2020
2019
| Jul 21, 2019 | Running opengl under WSL |
|---|---|
| Feb 28, 2019 | Practical Reverse Engineering Sample D |
2018
| Dec 08, 2018 | DXR-RTX Path Tracer Project HLSL Shader Code |
|---|---|
| Dec 07, 2018 | DXR-RTX Path Tracer Project Host Code |
| Dec 06, 2018 | DXR-RTX Path Tracer Project User Guide |
| Dec 06, 2018 | DXR-RTX Path Tracer Project Introduction |
| Aug 17, 2018 | Object Oriented C Library - Introduction |