1. Completing the Physical-Query-Plan and Chapter 16 Summary (16.7-16.8)
CS257 Spring 2009
Professor Tsau Lin
Student: Suntorn Sae-Eung
Donavon Norwood

2. Chapter 16: Query Compiler
16.7 Completing the Physical-Query-Plan
Choosing a Selection Method
Choosing a Join Method
Pipelining Versus Materialization
Pipelining Unary Operations
Pipelining Binary Operations
Notation for Physical Query Plan
Ordering the Physical Operations
16.8 Summary of Chapter 16

3. 16.7 Completing the Physical-Query-Plan
A query previously has been
parsed and preprocessed (16.1)
converted to logical query plans (16.3)
Estimated the costs of operations (16.4)
determined costs by cost-based plan selection (16.5)
weighed costs of join operations

4. Completing the Physical-Query-Plan (continued)
Three issues related to turning the logical plan into a complete physical plan are:
Selection of algorithm to do the operations of the query plan
Decisions after intermediate results are materialized and pipelined
Notation for physical-query-plan operators

5. 16.7.1 Choosing a Selection Method
Choose algorithms for each selection operators
e.g. use Indexes or table scan?
If so, how many indexed are used
Filter or not?

6. Example of Selection Method

7. Choosing a Join Method
Determine costs associated with each join algorithms
One-pass join
Sort-join

8. 16.7.3 Pipelining Versus Materialization
Materialization (naïve way) – order the operations appropriately, and store the result of each operations on disk until it is needed by another operations
Pipelining (more efficient way) – Interleave the execution of several operations, the tuples produced by one operation are passed directly to the operations the used it, without ever storing the intermediate tuples on disk.

9. 16.7.4 Pipelining Unary Operations
Unary = selection and projection
Best candidates for pipelining
Pipelining Unary Operations are implemented by iterators (15.1.6)

10. 16.7.5 Pipelining Binary Operations
The results of binary operations can also be pipelined.
See the example for more understanding

11. Pipelining Binary Operations (continued)
Example 16.36

12. 16.7.6 Notation for Physical Query Plans
There are various operators are found in physical query plan.
Operator for leaves
Physical Operators for Selection
Physical Sorts Operators
Other Relational-Algebra Operations
However, each DBMS will use its own internal notation for physical query plan.

13. 16.7.7 Ordering of Physical Operations
The physical query plan is represented as a tree, which implies about order of operations.
Three rules summarize the ordering of events implicit in a physical-query-plan tree:

14. Ordering of Physical Operations (continued)
Break the tree into subtrees at each edge that represents materialization. The subtrees will be executed one-at-a-time.
Order the execution of the subtrees in a bottom-up, left-to-right manner.
Execute all nodes of each subtree using a network of iterators. Thus, all the nodes in one subtree are executed simultaneously.

15. 16.8 Summary of Chapter 16 Compilation of Queries The parser
View expansion
Semantic Checking
Conversion to a Logical Query Plan
Algebraic Transformations
Choosing a Logical Query Plan

16. Summary of Chapter 16 (continued)
Estimating Sizes of Relations
Histograms
Cost-Based Optimization
Plan-Enumeration Strategies
Left-Deep Join Trees
Physical Plans for Selection
Pipelining Versus Materialization

17. Questions & Answers

18. For your attention

19. Reference
[1] H. Garcia-Molina, J. Ullman, and J. Widom, “Database System: The Complete Book,” second edition: p , Prentice Hall, New Jersy, 2008