Hacker GitHub

← All repos

deepseek-ai/3FS

View on GitHub ↗ ★ 10,408 C++

A high-performance distributed file system designed to address the challenges of AI training and inference workloads.

15 mentions on Hacker News by 14 people: 1 story, 14 comments · first 2025-02-28 · latest 2026-10-20

Mentions per year 15 mentions · busiest year 2025 (12)
20082026
  1. …[0] also open sourced [1] [0] https://github.com/deepseek-ai/DeepInfer/blob/main/doc/infer... [1] https://github.com/deepseek-ai/3fs
  2. …GiB per session. Save/restore is exactly what DeepSeek does using its 3FS distributed filesystem: https://github.com/deepseek-ai/3fs#3-kvcache With this much cheaper setup backed by disks, they can offer much better caching experience: >…
    comment by throwdbaaway · 2026-04-23 · in An update on recent Claude Code quality reports
  3. …If Anthropic got nothing here, then their inference cost can be much higher. DeepSeek also got https://github.com/deepseek-ai/3FS that makes cached reads a lot cheaper with way longer TTL. If Anthropic didn't need to invent and uses some…
  4. …down? - What's the difference in performance between using the kernel module versus using FUSE? [1] https://github.com/deepseek-ai/3FS/blob/main/docs/design_not...
  5. DeepSeek team, which is also an HFT shop, also implemented their DFS - https://github.com/deepseek-ai/3FS
  6. Deepseek has their own and they’re relatively small https://github.com/deepseek-ai/3FS It’s specialized knowledge, hard to do “correctly” (read posix here) but obtainable and implementable by a small team if you pick your battles right.…
  7. Cool. Also https://github.com/deepseek-ai/3FS by DeepSeek which came out of High-Flyer a Chinese HFT firm.
  8. …iCloud at Apple, and is _something_ at Snowflake, among a few others. Recently DeepSeek used it for https://github.com/deepseek-ai/3FS "The Fire-Flyer File System (3FS) is a high-performance distributed file system designed to address the…
    comment by fidotron · 2025-05-07 · in Foundation DB Record Layer SQL API
  9. …FUSE client for all the metadata operations that the native client would have needed to implement. https://github.com/deepseek-ai/3FS/blob/ee9a5cee0a85c64f4797...
  10. …is cool and all, but it's also stupid and their scale-out strategy is non-existent. This is why https://github.com/deepseek-ai/3FS came out but of course nobody had figured it out because they still think LLM's is like, chatbots, or…
    comment by tucnak · 2025-04-05 · in The Llama 4 herd
  11. https://github.com/deepseek-ai/3FS?tab=readme-ov-file#3-kvca... KVCache is a technique used to optimize the LLM inference process. It avoids redundant computations by caching the key and value vectors of previous tokens in the decoder…
  12. Deepseek is already using SSDs for their KV cache: https://github.com/deepseek-ai/3FS
  13. …AI but also distributed computing: 1. smallpond: https://github.com/deepseek-ai/smallpond 2. 3fs: https://github.com/deepseek-ai/3FS 3. deepep: https://github.com/deepseek-ai/DeepEP
  14. …In terms of fast FUSE - also my first question, appears to be`io_uring` + FUSE :) https://github.com/deepseek-ai/3FS/blob/main/src/lib/api/Usr...
    comment by grohan · 2025-02-28 · in Fire-Flyer File System (3FS)