[p2p-hackers] Simple lightweight DHT

Michael Parker mgp at ucla.edu
Tue Jan 11 17:32:37 UTC 2005


I was about to ask a similar question... You're a seasoned peer-to-peer 
developer -- for those of us who are developers just starting out, and 
perhaps trying to invent our own new, novel topologies and systems, what 
are the hardest things to 'get right' in a peer-to-peer system?

I've read the paper "Designing a DHT for Low Latency and High 
Throughput" [1] by the Chord group at MIT. It seems to sum up pretty 
well what their challenges were, and what practices they found were best.

I know there are some other people out on this mailing list who could 
answer this too (Clarke, Freedman... if you feel that your name should 
be on this list, just respond). Sorry to try and drag you into the 
spotlight, but you definitely have a captive audience ;)

- Michael Parker

[1] http://citeseer.ist.psu.edu/dabek04designing.html


Zooko O'Whielacronx wrote:

> On 2005, Jan 10, at 20:00, Sean C. Rhea wrote:
>
>> I wrote the Bamboo router and got it working in about a week, 
>> although I had the experience and code base from writing Tapestry 
>> (another DHT) before that, so I had a bit of a head start.
>>
>> It took me another year to get Bamboo to perform as well as it does 
>> today.  I believe the Chord people had a similar experience.
>
>
> What a fascinating story!  What things did you have to learn and 
> invent during the course of that year to improve the performance of 
> Bamboo?
>
> Regards,
>
> Zooko
>
> _______________________________________________
> p2p-hackers mailing list
> p2p-hackers at zgp.org
> http://zgp.org/mailman/listinfo/p2p-hackers
> _______________________________________________
> Here is a web page listing P2P Conferences:
> http://www.neurogrid.net/twiki/bin/view/Main/PeerToPeerConferences
>



More information about the P2p-hackers mailing list