Consider the join r œA
(a) Sort-merge
(b) Hash-based
(c) Index-nested loops
Briefly (in one paragraph per technique) explain which technique is reasonably good in the above case. (In case of at least one technique you have to distinguish between two cases.) The explanations should be short and to the point.
Solution:
Sort-merge join is a reasonable technique. In the last phase of sorting, we can perform the join by comparing the attribute A of the front tuples of r with the attribute of the front tuple of s and select those where the condition A
Hash-based join is useless in this case, since the join condition does not involve equality.
Index-nested loops join will work reasonably well provided that the index on B in s is clustered. For unclustered index, this technique will result in one I/O per matching tuple, which is worse than plain scan.
You might also like to view...
An exception's default error message can be retrieved by using the __________ method.
a. getMessage() b. getErrorMessage() c. getDefaultMessage() d. getDefaultErrorMessage()
________ law gives the author intellectual property ownership of the property and the right to choose who uses it
Fill in the blank(s) with correct word
How do you bring up the Layer Style dialog box?
What will be an ideal response?
Describe at least three different types of expert review methods.
What will be an ideal response?