[chord] my plans to change merkle
James Robertson
jsr at MIT.EDU
Fri Aug 1 14:31:26 EDT 2003
On Fri, 1 Aug 2003, Benjie Chen wrote:
> dbm time time to insert into merkle tree
> trial 0 7.5 4.7
> trial 1 7.6 8.5
> trial 2 7.4 22.5
> trial 3 7.2 40.7
>
> okay, so we now know where the time goes. there are several possible causes
Hey Benjie, how about doing a test where you change database_get_keys to
stop iterating once it reaches the end of the range you are interested in?
I hope that would limit the blowup in insertion time.
James
More information about the chord
mailing list