Chapter 18 Query Processing and Optimization

Chapter Outline u Introduction. u Using Heuristics in Query Optimization –Query Trees and Query Graphs –Transformation Rules for RA expressions –Heuristic Optimization

Introduction u With declarative languages such as SQL, user specifies what data is required rather than how it is to be retrieved. u Relieves user of knowing what constitutes good execution strategy. u Also gives DBMS more control over system performance. u Two main techniques for query optimization: – heuristic rules that order operations in a query. – comparing different strategies based on relative costs, and selecting one that minimizes resource usage.

Introduction Query Processing (QP) : Activities involved in retrieving data from the database. u Aims of QP: – transform query written in high-level language (e.g. SQL), into correct and efficient execution strategy expressed in low-level language (implementing RA); – execute the strategy to retrieve required data. Query Optimization (QO): Activity of choosing an efficient execution strategy for processing query. u As there are many equivalent transformations of same high-level query, aim of QO is to choose one that minimizes resource usage. u Generally, reduce total execution time of query. u May also reduce response time of query. u Problem computationally intractable with large number of relations, so strategy adopted is reduced to finding near optimum solution.

Introduction

Query Trees and Query Graphs

Transformation Rules for RA Operations u Cascade of Selection: Conjunctive selection operations can cascade into individual selection operations  p  q  r (R) =  p (  q (  r (R))) u Commutativity of selection.  p (  q (R)) =  q (  p (R)) u Cascade of Projection: In a sequence of projection operations, only the last in the sequence is required.   L   M …   N (R) =   L (R) u Commutativity of  and  : If the selection condition c involves only those attributes A 1, … A n in the projection list, then the two operations can commute:   Ai, …, Am (  c (R)) =  c (   Ai, …, Am (R))

Transformation Rules for RA Operations u Commutativity of  (and  ). R  p S = S  p R R  S = S  R u Associativity of  (and  ).  and natural join ( * )are always associative: (R * S)  T = R * (S * T) (R  S)  T = R  (S  T) If join condition q involves attributes only from S and T, then  is associative as follows: (R  p S)  q  r T = R  p  r (S  q T)

Transformation Rules for RA Operations u Commutativity of  and  (or  ). If all the attributes in the selection condition c involves only the attributes of one of the relations being joined – Say R – then the two operations can be commuted as follows:  c (R  r S) = (  c (R))  r S  c (R  S) = (  c (R))  S Alternatively if the condition c can be written as (c 1 and c 2 ), where condition c 1 involves only the attributes of the relations R and condition c 2 involves only the attributes of the relations S then the two operations can be commuted as follows:  c1 (R  r S) = (  c1 (R))  r (  c2 ( S))  c1 (R  S) = (  c1 (R))  (  c2 ( S))

Transformation Rules for RA Operations u Commutativity of  and  (or  ). If projection list is of form L = L 1  L 2, where L 1 only involves attributes of R, and L 2 only involves attributes of S, provided join condition only contains attributes of L, projection and theta-join operations commute as:   L1  L2 (R  r S) = (   L1 (R))  r (   L2 (S)) If join condition contains additional attributes not in L, say attributes M = M 1  M 2 where M 1 only involves attributes of R, and M 2 only involves attributes of S, a final projection operation is required:   L1   L2 ( R  r S ) =   L1  L2 ( (   L1   M1 ( R ))  r (   L2  M2 (S))) u Commutativity of  and  (but not - ). R  S = S  R R  S = S  R u Associativity of  and  (but not - ). (R  S)  T = S  (R  T) (R  S)  T = S  (R  T)

Transformation Rules for RA Operations Commutativity of  and set operations ( , , and - ).  p (R  S) =  p (S)   p (R)  p (R  S) =  p (S)   p (R)  p (R - S) =  p (S) -  p (R) u Commutativity of  and .   L (R  S) =   L (S)    L (R)

Heuristical Processing Strategies u Perform selection operations as early as possible. – Keep predicates on same relation together. u Combine Cartesian product with subsequent selection whose predicate represents join condition into a join operation. u Use associativity of binary operations to rearrange leaf nodes so leaf nodes with most restrictive selection operations executed first. u Perform projection as early as possible. – Keep projection attributes on same relation together. u Compute common expressions once. – If common expression appears more than once, and result not too large, store result and reuse it when required. – Useful when querying views, as same expression is used to construct view each time.

Use of Transformation Rules Q: Find the last name of employees born after 1975 who work on a project named ‘Aquarius’. SELECTLNAME, FROMEMPLOYEE, WORKS_ON, PROJECT, WHEREPNAME=‘Aquarius’ AND ESSN= SSN AND PNUMBER=PNO AND BDATE > ‘ ’

Use of Transformation Rules

Example 18.3 Use of Transformation Rules