Slideshow:
- Important fact:
- The join (⋈) operation is an essential operation in query processing
Therefore, it is very important that:
- The join (⋈) operation
have a
flexible
(= wide variety) of
implementation algorithms:
- We need
cost efficient
⋈ algorithm:
with low disk I/O cost
but
high memory requirement
E.g.: the one-pass join (⋈) algorithm --- click here
- And memory efficient ⋈ algorithm with higher disk I/O cost
- We need
cost efficient
⋈ algorithm:
with low disk I/O cost
but
high memory requirement
- Nested-loop algorithms:
- Some operators
(notably:
the cartesian product (×)
and the join (⋈))
have implementations
that uses
very little memory buffers
(As little as 2 buffers)
- These low memory requirement
algorithms
will read some
input relation(s)
multiple times
I.e.: the their disk IO cost will be higher when the algorithm uses lower number of buffers
(Another example of the Law of Conservation of Misery...)
- These low memory requirement
algorithms are called
nested-loop algorithms
(Because they use nested while-loops to process the tuples)
- These low memory requirement
algorithms are called
nested-loop algorithms
- Some operators
(notably:
the cartesian product (×)
and the join (⋈))
have implementations
that uses
very little memory buffers