Re: [Gzz-commits] manuscripts/storm article.rst

[hemppah]

Quoting Benja Fallenstein <address@hidden>:

> address@hidden wrote:
> > Lainaus Benja Fallenstein <address@hidden>:
> >>But how is it relevant to Storm? I mean, you surely aren't saying that 
> >>we should include everything that is interesting, no matter whether it 
> >>is relevant or not? :-)
> > 
> > I think it's as relevant as other p2p techniques to Storm :).
> 
> E.g. DHTs are relevant because using them, you can implement a StormPool 
> with blocks from all servers/peers on the public Internet-- and you 
> cannot without using a DHT or other decentralized indexing scheme. So 
> you *can* implement a vital part of the Storm design with them, and you 
> *can't* without them-- that clearly makes them relevant IMO :-)

Hm..okay, this is Gzz/Storm specific, so I trust your word (since you know them
better than me) :).

About DHTs and a Storm Pool: is the fact that in DHTs data is identified by keys
the key factor for StormPool suitableness, or are there other factors as well ?

Btw, this is quite essential point to my thesis, when comparing suitabilities of
differenct p2p schemes ;). 


> 
>  > But I don't say
> > we should include everything in p2p field; just the things which are most
> 
> > common/interesting are worth to mention in short overview section on p2p 
> > research field (?).
> 
> No, I think we should include only those things that are directly 
> related to Storm, i.e. file sharing/distributed indexing things.

Ok, I'll remove/modidy text ;).


-Hermanni





-------------------------------------------------
This mail sent through IMP: http://horde.org/imp/