> Are you sure that is a good idea?  Surely the disk seeks would destroy
> you on index load?

I'm not sure what you mean there.  We can create a total HFile block
key index that has pointers to within blocks, eg, they'll store a
block index int, and a position (within the block) int.  Where would
the index load occur as this method should allow faster key lookup?

On Sun, Jun 5, 2011 at 11:37 PM, Ryan Rawson <[email protected]> wrote:
> On Sun, Jun 5, 2011 at 11:33 PM, Jason Rutherglen
> <[email protected]> wrote:
>> Ok, the block index is only storing the first key of each block?
>> Hmm... I think we can store a pointer to an exact position in the
>> block, or at least allow that (for the FST implementation).
>
> Are you sure that is a good idea?  Surely the disk seeks would destroy
> you on index load?
>

Reply via email to