TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Seaweed-FS: simple distributed file system written in Go

67 点作者 thinxer将近 10 年前

6 条评论

sargun将近 10 年前
Self-plug. This is based off the Haystack paper: <a href="https:&#x2F;&#x2F;www.usenix.org&#x2F;legacy&#x2F;event&#x2F;osdi10&#x2F;tech&#x2F;full_papers&#x2F;Beaver.pdf" rel="nofollow">https:&#x2F;&#x2F;www.usenix.org&#x2F;legacy&#x2F;event&#x2F;osdi10&#x2F;tech&#x2F;full_papers&#x2F;...</a> from Facebook. Relevant blog-post is here: <a href="https:&#x2F;&#x2F;code.facebook.com&#x2F;posts&#x2F;685565858139515&#x2F;needle-in-a-haystack-efficient-storage-of-billions-of-photos&#x2F;" rel="nofollow">https:&#x2F;&#x2F;code.facebook.com&#x2F;posts&#x2F;685565858139515&#x2F;needle-in-a-...</a><p>I gave a talk about Haystack at PWL: <a href="http:&#x2F;&#x2F;www.meetup.com&#x2F;papers-we-love-too&#x2F;events&#x2F;220795812&#x2F;" rel="nofollow">http:&#x2F;&#x2F;www.meetup.com&#x2F;papers-we-love-too&#x2F;events&#x2F;220795812&#x2F;</a> with the relevant video here: <a href="https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=EuNumdi1Do0" rel="nofollow">https:&#x2F;&#x2F;www.youtube.com&#x2F;watch?v=EuNumdi1Do0</a>
评论 #10032287 未加载
评论 #10031726 未加载
notacoward将近 10 年前
Already posted on HN (albeit quite a while ago when it was still called Weed-FS) and still not actually a file system.
评论 #10031733 未加载
pram将近 10 年前
Looks very interesting, thanks. Kind of hilarious he didn&#x27;t make a native Go API though..<p>This is the only one AFAICT: <a href="https:&#x2F;&#x2F;github.com&#x2F;ginuerzh&#x2F;weedo" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;ginuerzh&#x2F;weedo</a>
评论 #10032799 未加载
e12e将近 10 年前
Looks interesting. I&#x27;m curious about why there&#x27;s a 32bit cookie added to the volume,filename -- it seems like a rather weak protection -- and as such, it becomes an unnecessary complication? As have been shown with facebook, relying on (permanent) secret urls to grant&#x2F;deny access is a bad idea.<p>So, why not just use volume,id, and then deploy a proxy that handless access based on tokens in front -- if access control is wanted? (Not all uses of files will need&#x2F;want access control).<p>I suppose one reason for a &quot;cookie&quot; would be cache invalidation in case of volume,id reuse.
fit2rule将近 10 年前
I&#x27;m thinking that IPFS (<a href="http:&#x2F;&#x2F;ipfs.io&#x2F;" rel="nofollow">http:&#x2F;&#x2F;ipfs.io&#x2F;</a>) has more of an opportunity to fill the distributed-filesystem role de jeur, alas .. Seaweed-FS seems flakey and not quite ready for primetime. Poor conclusion?
millstone将近 10 年前
Honest question: why is this called a file system and not a file server? It looks closer to Apache than, say, ext3.