Live data from Hacker News

Baidu File System – A distributed file system for real-time applications

github.com

41–50 of 169 posts

Re: Baidu File System – A distributed file system for real-time applications

#42
post #25
post #21

Earlier quoted context omitted.

I imagined that. I am more like "let me tell you exactly what I want" rather than relying on behavior defined elsewhere. It would not pass review where I work.

Not to be rude, but I'm glad I don't work where you work. How would that be consistent with your own classes? "Oh no, you can't just use a 'Tree' object, you need to explicitly set that there are no leaves yet, no branches yet, no squirrels yet, etc… etc…" Do you .clear() your vectors before you use them? This sounds like newbies that do: #define TRUE (1 == 1)

That define is very clever: it always lands on it's feet! :)

Re: Baidu File System – A distributed file system for real-time applications

#44
post #21
post #19

Earlier quoted context omitted.

The value constructor for strings is the empty string.

I imagined that. I am more like "let me tell you exactly what I want" rather than relying on behavior defined elsewhere. It would not pass review where I work.

But to a C++ programmer, that code tells exactly what is going on... There's no hidden logic, no non-standard or compiler dependent features.

Re: Baidu File System – A distributed file system for real-time applications

#45
post #28

Earlier quoted context omitted.

Do you even know how it is when not-your-own-language is the dominant one for everything in computers?

Sssshhh don't point out the hypocrisy.

Is it hypocrisy? English is not my native language but I consider it the default CS language. There must be a way for us to share knowledge, and that turned out to be English.

Re: Baidu File System – A distributed file system for real-time applications

#46
Based on the design[1], it has a leader / follower pattern (although you should have multiple leaders with Raft consensus to avoid having a single point of failure), where the leader is called "nameserver" and decides where to put each piece of data and metadata among a set of chunk servers and metadata servers.

That design is very reminiscent of CephFS's cluster monitors, metadata servers, object storage devices.

[1]: https://github.com/baidu/bfs/blob/master/docs/design.md, https://github.com/baidu/bfs/blob/master/docs/BFS_design.md

Re: Baidu File System – A distributed file system for real-time applications

#47

Can a distributed storage expert comment in what ways this differs from hadoop?

first of all, impl in C++ (JVM/GC is pain in the ass) - clear arch (only master and dataserver) - very concise config file and easy to deploy - most important, 10k nodes scalability without federation design of namespace

Re: Baidu File System – A distributed file system for real-time applications

#48

Based on the design[1], it has a leader / follower pattern (although you should have multiple leaders with Raft consensus to avoid having a single point of failure), where the leader is called "nameserver" and decides where to put each piece of data and metadata among a set of chunk servers and metadata servers. That design is very reminiscent of CephFS's cluster monitors, metadata servers, object storage devices. [1…

I thought Raft was always-single-leader? Followers can happily become leaders through elections?

Re: Baidu File System – A distributed file system for real-time applications

#49

Why is there no English documentation ?

Why is there rarely any non-english-language documentation for most codebases nowadays? The world doesn't revolved around english-speaking countries.

Because English is the lingua franca of the software industry, and developers are usually expected to know English, no matter where they're from.
Post reply on HN