bittorent doubt( about DHT)

Status
Not open for further replies.

greenhorn

Enclave Plus
Juggernaut
where do they get the peers in torrent whose tracker is dead ? I tried t download a torrent which was tracked by demonoid. the demonoid trackers been dead for a while. , but still i get some peers , and it starts downloading.

I imagine its the work of DHT , but how exactly does DHT kick in? I mean now that the tracker is dead , how, and specifically , from where does the torrent client find out the peers
 
itz indeed work of dht...
dht allows the torrent to work widout a tracker....what it means is that your client will be able to find peers even when the tracker is down, or doesn't even exist anymore...
it allows the swarm to continue as normal widout a tracker...
how it actually works is not known by me either...:P
may be when the tracker is offline(or rather even when itz online) the dht operates like the gnutella n/w and helps finding peers...not sure though
may be this is of help...DHT
 
from my understanding , DHT works only with the peers you know, and even if a peer is added via the tracker , the others can know via peer exchange. but what happens when you dont know the peers . Apparently it still works, but i dont know how...
 
greenhorn said:
where do they get the peers in torrent whose tracker is dead ? I tried t download a torrent which was tracked by demonoid. the demonoid trackers been dead for a while. , but still i get some peers , and it starts downloading.
I imagine its the work of DHT , but how exactly does DHT kick in? I mean now that the tracker is dead , how, and specifically , from where does the torrent client find out the peers

Was there just one tracker???
From what i know, if the tracker is down u cant download it...DHT and all connect thorugh known peers...
Do correct me if im wrong..
 
in this case , there was just one demonoid tracker, and its down ( no connexion because target machine refused it)

and yet i got so many peers and its downloading. so was wondering...
 
I had to do a bit of research on DHT earlier,so here's a simple explanation of its working (as its in my head atm) :)

The main idea of DHT is to provide decentralization and fault tolerance.So,In a DHT network,as its name suggest, there is huge distributed hash table,which stores many data items,client info and routing info etc. Each client stores a part of the table ,other clients may lookup data from our table.Thus the other client can know which client has what file.Thus it can find new peers which has required file.

So when at start we know no clients in DHT,there are several methods to find a client in DHT,Then its starts bootstrapping,where peers' exchange data,so it get to know more peers.

There are a lot of diff implementation and types of DHT,as you can see in wiki.

In simple terms,its similar to gnutella in some ways,but its more scalable,no query flooding,network stability is much better.
 
Status
Not open for further replies.