Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> comp.databases.theory -> System R and

System R and

From: <accpactec_at_hotmail.com>
Date: 16 Oct 2006 12:20:53 -0700
Message-ID: <1161026453.769455.284230@k70g2000cwa.googlegroups.com>


Given

Branch(bid, accNo, budget, status)
Account(accNo, code, balance)

and the query:

Select        B.bid, Count(*)

>From Branch B, Account A
Where B.accNo=A.accNo

Group by B.bid

All indexes below are clustered. If we have to just guess (without calculation)

  1. Suppose that no indexes are available. What is the plan with the lowest estimated cost.
  2. If there is a hash index on A.accNo, what is the plan with lowest estimated cost.
  3. If there is a hash index on B.accNo, what is the plan with lowest estimated cost.
  4. If there is a hash index on B.accNo and A.accNo, what is the plan with lowest estimated cost.
Received on Mon Oct 16 2006 - 14:20:53 CDT

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US