Distributed Hash Table Protocol (2008) | Hacker News Reader

Distributed Hash Table Protocol (2008)

bittorrent.org

Since we're talking about DHT's, does anybody know what happened to OpenDHT? That had a lot of buzz and activity at one point in time, then just kinda disappeared.

And on the same note... is there a generally available, sort of "free for all" DHT system out there now? If not, what stops one from existing?

Just curious, as it seems like DHTs have lots of uses, but it seems redundant for every app that needs one, to re-implement their own DHT network.

I'm working on both! http://github.com/urdht

Hey cool, I'll take a look... maybe I can help a little.

You could write one pretty simply on top of something like Ringpop:

https://github.com/uber/ringpop

There's also a Go implementation of Ringpop coming right around the corner too.

There's BEP 44 which enables use of the bittorrent DHT for non-bittorrent applications.

http://bittorrent.org/beps/bep_0044.html

You mean an open library? Or do you actually mean a unified DHT network, with many different applications all using the same network?

Or do you actually mean a unified DHT network, with many different applications all using the same network?

Yes, this.

router.bitcomet.com router.bittorrent.com

the latter runs: https://github.com/bittorrent/bootstrap-dht

It may not be the implementation you referred to, but https://github.com/savoirfairelinux/opendht

exist and is used by the Ring distributed communication platform along with SIP (ring.cx) to implement the account less, P2P and PKI encrypted communication.

(disclaimer, I have previously worked on Ring.cx)

Anyone know of a good python or java implementation of Kademila DHT that this is based on?

Edit:

Found a list:

https://en.wikipedia.org/wiki/Kademlia#Implementations

Would still like to hear your impressions if you have used any of them.

I've played around with openkad [1], but never got far enough to create anything meaningful with it. The authors implemented multiple distance metrics for more efficient lookup and distribution; these are described in [2].

[1] https://code.google.com/p/openkad/

[2] http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6...

TomP2P is really great!

TomP2P is mature enough to run in an closed environment (easy network conditions). We are currently working on making it run (relaying, NAT, UPNP) stable in the Internet.

https://www.youtube.com/watch?v=1eew8BngN7Y#t=3m14s might be interesting to some people.

Er, this 2008 BEP is widely known and referenced and was updated in 2013.

Doesn't seem worth a post.

On other hand, it would be worthwhile if a reader had a paper for an alternative with better attributes. I wonder what the state-of-the-art is in these things.

Are there any experiments with incentivized distributed hash tables? eg. nodes could be paid bitcoin or something for servicing the network.

> Are there any experiments with incentivized distributed hash tables? eg. nodes could be paid bitcoin or something for servicing the network.

http://filecoin.io/ is supposed to be this, but it hasn't been fully implemented yet. http://ipfs.io/, which is the generalized DHT framework without the incentives, is fairly functional at this point.

There's also MaidSafe-Routing / Safecoin, which is production-ready.

https://www.youtube.com/watch?v=YFV908uoLPY

https://github.com/maidsafe-archive/MaidSafe-Routing/wiki

I'm the author of TomP2P. We are currently working on integrating bitcoins for registration to make Sybil attacks much harder: https://github.com/tomp2p/TomP2P/tree/bitcoin

Enigma

http://enigma.media.mit.edu/

http://enigma.media.mit.edu/enigma_full.pdf

Reply on news.ycombinator.com