Re: DHT expiration questions

--- In [email protected], "Raphael Manfredi" <Raphael.Manfredi@...> wrote:
>
> --- In [email protected], "pwang708" <wang.peng@> wrote:
> > After processing the STORE requests, we see that about
> > 227649 publishes did not have republishes, 9586 have 1 republishes,
> > 1976 have 2 republishes...
> 
> Could you focus only on STORE requests made for keys that fall within
> your k-ball?
> 
> Out of these 227649 publishes, I'd love to see how many would be ruled
> out.  This can change the picture significantly.
> 

I did not keep my neighborhood infomation. But I computed the common prefix length between my node and the keys in the store requests. 99.6% have common prefix longer or equal to 13 bits. 93% have common prefix longer or equal to 14 bits. The picture won't change much.


> Also, your study focuses on the traffic a given node receives, but you
> need also to consider the publishing cost for the publishing node. 

It was a sample of STOREs happening in the network. 

Best,
Peng