RE: Cost of hash join

Hi Naveen  

>cost= (access cost of A * number of hash partitions of

>B) + access cost of B.

>

>Why is the access cost of A multiplied by no. of hash

>partitions?
 

In my opinion this is not a good one (perhaps for this reason it's no more in the 10g documentation...).

In fact the cost is something like:  

cost = access cost A + access cost B + cost join.  

If the hash table is "small", the cost of the join itself will be close to 0.

If the hash table is "large", the cost is strongly dependent on the amount of "available" PGA memory.  

>Is "A" read for each hash partition of B?
 

Each table is read only once.    

HTH Chris

--
http://www.freelists.org/webpage/oracle-l