1. 1 Overview of Indexing Chapter 8 – Part II. 1. Introduction to indexing 2. First glimpse at indices and workloads

2. 2 Understanding the Workload  For each query in workload:  Which relations does it access?  Which attributes are retrieved?  Which attributes are involved in selection/join conditions?  How selective are these conditions likely to be?  For each update in workload:  Which attributes are involved in selection/join conditions?  How selective are these conditions likely to be?  The type of update ( INSERT/DELETE/UPDATE ), and the attributes that are affected.

3. 3 Choice of Indexes  What indexes should we create?  Which relations should have indexes?  What field(s) should be the search key?  Should we build several indexes?  For each index, what kind of an index should it be?  Clustered vs. unclustered? Hash vs. tree? Clustering must be used sparingly and only when justified by frequent queries that benefit from clustering. At most one index can be clustered. (Why?) Consider utilizing index-only evaluation. (e.g., avg(age))

4. 4 Choice of Indexes: One Approach  Consider most important queries in turn.  Consider best plan using current indexes, and see if a better plan possible with additional index. If so, create it.  Consider impact on updates in workload! Trade-off: Indexes can make queries go faster, updates slower. Require disk space, too.  Obviously, we must understand how DBMS evaluates queries and creates query evaluation plans

5. 5 Choice of Indexes: Simple Approach For now, we discuss simple 1-table queries.

6. 6 Index Selection Guidelines  Attributes in WHERE clause are candidates for index keys.  Exact match condition suggests hash index.  Range query suggests tree index.  Clustering is especially useful for range queries  Clustering can also help equality queries if there are many duplicates.

7. 7 Index Selection Guidelines  Multi-attribute search keys considered when WHERE clause contains several conditions.  Order of attributes is important for range queries.  Such indexes can sometimes enable index-only strategies for important queries.  Question : For index-only strategies, is clustering important ?

8. 8 Index Selection Guidelines  Try to choose indexes that benefit as many queries as possible.  Since only one index can be clustered per relation, choose it based on important queries that would benefit the most from clustering.

9. 9 Examples of Clustered Indexes  B+ tree index on E.dno?  B+ tree index on E.age ?  Trade-offs :  How selective is the condition? (all > 40?) or (only some > 40)  Is the index clustered? SELECT E.dno FROM Emp E WHERE E.age>40

10. 10 Examples of Clustered Indexes  Consider the GROUP BY query.  Index on E.age ? E.dno ?  Issues :  Use Index on E.age ?  If many tuples have E.age > 10, using E.age index and sorting the retrieved tuples may be costly.  Use Index on E.dno ?  Clustered E.dno index may be better!  What about without WHERE condition? SELECT E.dno, COUNT (*) FROM Emp E WHERE E.age>10 GROUP BY E.dno

11. 11 Examples of Clustered Indexes  B+ tree index on E.hobby?  NOTE: is equality query.  NOTE : may contain many duplicates.  Clustered or Unclustered index ?  CONCLUDE : Clustering on E.hobby helps!  QUESTION: what if index is unclustered ?  CONCLUDE: may prefer to do a full scan. SELECT E.dno FROM Emp E WHERE E.hobby=Stamps

12. 12 Indexes with Composite Search Keys sue1375 bob cal joe12 10 20 8011 12 nameagesal 12,20 12,10 11,80 13,75 20,12 10,12 75,13 80,11 11 12 13 10 20 75 80 Data records sorted by name Data entries in index sorted by Data entries sorted by Composite Search Keys : Search on combination of fields (sal and age).

13. 13 Equality and Composite Search Keys  Equality query: Every field value is equal to a constant value.  Examples :  age=20  sal =75  age=20 and sal =75  sal =75 and age=20 sue1375 bob cal joe12 10 20 8011 12 nameagesal 12,20 12,10 11,80 13,75 20,12 10,12 75,13 80,11 11 12 13 10 20 75 80 Data records sorted by name Data entries in index sorted by Data entries sorted by

14. 14 Composite Search Keys  If retrieve Emp records with age =30 AND sal =4000  Index on would be better than an index on age or an index on sal.  Choice of index key orthogonal to clustering

15. 15 Ranges and Composite Search Keys  Range query: Some field value is not a constant but a range.  Examples :  age=12 and sal > 10  age =12 sue1375 bob cal joe12 10 20 8011 12 nameagesal 12,20 12,10 11,80 13,75 20,12 10,12 75,13 80,11 11 12 13 10 20 75 80 Data records sorted by name Data entries in index sorted by Data entries sorted by Examples of composite key indexes using lexicographic order.

16. 16 Composite Search Keys  If condition is: 20< age <30 AND 3000< sal <5000:  Clustered tree index on or is best.  If condition is: age =30 AND 3000< sal <5000:  Clustered index much better than index!  Composite indexes are larger, updated more often.

17. 17 Index-Only Plans  Answer a query without retrieving actual tuples …  Is that possible ?  If index with suitable information is available.  Why is it a good idea ?

18. 18 Index-Only Plans  A number of queries can be answered without retrieving any tuples from one or more of the relations involved if a suitable index is available. SELECT D.mgr FROM Dept D, Emp E WHERE D.dno=E.dno SELECT D.mgr, E.eid FROM Dept D, Emp E WHERE D.dno=E.dno SELECT E.dno, COUNT (*) FROM Emp E GROUP BY E.dno SELECT E.dno, MIN (E.sal) FROM Emp E GROUP BY E.dno SELECT AVG (E.sal) FROM Emp E WHERE E.age=25 AND E.sal BETWEEN 3000 AND 5000 Tree index! Tree index! or Tree!

19. 19 Index-Only Plans SELECT E.dno, COUNT (*) FROM Emp E GROUP BY E.dno SELECT E.dno, MIN (E.sal) FROM Emp E GROUP BY E.dno SELECT AVG (E.sal) FROM Emp E WHERE E.age=25 AND E.sal BETWEEN 3000 AND 5000 ? or ? Tree index! + Does index-only evaluation make sense?

20. 20 Index-Only Plans : Multi-Key Index PROS: + The chance for index-only evaluation is increased. CONS: - Index size larger. - Update response for any field.

21. 21 Index-Only Plans  Tree index on, or on :  Which is better? SELECT E.dno, COUNT (*) FROM Emp E WHERE E.age=30 GROUP BY E.dno

22. 22 Index-Only Plans  Tree index on, or on :  Which is better? SELECT E.dno, COUNT (*) FROM Emp E WHERE E.age=30 GROUP BY E.dno SELECT E.dno, COUNT (*) FROM Emp E WHERE E.age>30 GROUP BY E.dno What if we consider the second query?

23. 23 Index-Only Plans : Multiple Relations SELECT DISTINCT ( D.mgr ) FROM Dept D, Emp E WHERE D.dno=E.dno Or Or

24. 24 Index-Only Plans SELECT D.mgr FROM Dept D, Emp E WHERE D.dno=E.dno SELECT D.mgr, E.eid FROM Dept D, Emp E WHERE D.dno=E.dno Or Or Tree index!

25. 25 Summary  Understanding nature of workload for application, and the performance goals is essential to developing a good design.  What are the important queries and updates?  What attributes/relations are involved?

26. 26 More Summary  Indexes must be chosen to speed up important queries  Index maintenance overhead on updates to key fields.  Choose indexes that can help many queries, if possible.  Build indexes to support index-only strategies.  Clustering is an important decision; only one index on a given relation can be clustered!  Order of fields in composite index key can be important.