[OpenDHT-Users] subsequent gets

Kannaiyan Natesan gbpnkans at gmail.com
Mon Aug 21 16:41:18 PDT 2006


Hi Sean,

It is a good option.
I'm not sure the 20 byte hash will be able to fullfill the
requirements of all the clients.
It is good to store multiple chunk of data on a single hash.

Don't bother about the load you give to OpenDHT. It is upto them to
maintain the balance of the servers.

Think of a scheme, where you can split a file and store in a single
hash and retrieve the whole file from the single hash. It is an
example. Not sure whether you can use openDHT for filesharing
purposes.

It need not be file, if you want to store any info on the DHT
associated with you, you can use single hash and end with that.

Regards,
Kannaiyan


On 8/22/06, Sean Rhea <sean.c.rhea at gmail.com> wrote:
> On 8/21/06, Dominik Berndl <berndld at cip.ifi.lmu.de> wrote:
> > Hi everyone,
> > for my application i store a lot of vlaues (lets say 200) under the same key.
> > If i submit a get via xmlrpc with this key the dht returns only two values
> > per get, so i have to do about 100 subsequent gets.
> > The average time for one get is about 500ms, so the whole procedure takes
> > something about 50 seconds.
> > Is there any possibility in opendht to get more than only two values per get?
>
> They must be large values if you're only getting two values back per
> get request.
>
> I think the idea is that OpenDHT isn't meant to be used that way.
> Storing so many large values under the same key leads to load
> imbalance.  It would be better if you could find a way to store them
> under different keys.
>
> Sean
> --
> Looking for a PGP signature?
> See http://srhea.net/gmail-explain.txt
>
> _______________________________________________
> OpenDHT-Users mailing list
> OpenDHT-Users at opendht.org
> http://opendht.org/cgi-bin/mailman/listinfo/opendht-users
>



More information about the OpenDHT-Users mailing list