Lecture 16:

Relational Algebra
Monday, May 10, 2010

Dan Suciu -- 444 Spring 2010 1

 Outline
Relational Algebra:
•  Chapters 5.1 and 5.2

Dan Suciu -- 444 Spring 2010 2

 The WHAT and the HOW
•  In SQL we write WHAT we want to get form
the data

•  The database system needs to figure out

HOW to get the data we want

•  The passage from WHAT to HOW goes

through the Relational Algebra

Data
Dan Independence
Suciu -- 444 Spring 2010 3
 SQL = WHAT
Product(pid, name, price)
Purchase(pid, cid, store)
Customer(cid, name, city)
SELECT DISTINCT x.name, z.name
FROM Product x, Purchase y, Customer z
WHERE x.pid = y.pid and y.cid = y.cid and
x.price > 100 and z.city = ‘Seattle’

It’s clear WHATDanwe want,

Suciu unclear
-- 444 Spring 2010 HOW to get it4
 Relational Algebra = HOW
Final answer
Product(pid, name, price) δ

Purchase(pid, cid, store) T4(name,name)

Customer(cid, name, city)

Π

x.name,z.name
T3(. . . )
T2( . . . .) σ

price>100 and city=‘Seattle’

T1(pid,name,price,pid,cid,store) cid=cid

Temporary tables pid=pid

T1, T2, . . . Customer
5
Product Purchase
 Relational Algebra = HOW
The order is now clearly specified:
Iterate over PRODUCT…

…join with PURCHASE…

…join with CUSTOMER…

…select tuples with Price>100 and
City=‘Seattle’…

…eliminate duplicates…

…and that’s the final answer !

Dan Suciu -- 444 Spring 2010 6

 Sets v.s. Bags
•  Sets: {a,b,c}, {a,d,e,f}, { }, . . .
•  Bags: {a, a, b, c}, {b, b, b, b, b}, . . .

Relational Algebra has two semantics:

•  Set semantics
•  Bag semantics

Dan Suciu -- 444 Spring 2010 7

 Extended Algebra Operators
•  Union ∪, intersection ∩, difference -
•  Selection σ

•  Projection Π

•  Join ⨝
•  Rename ρ

•  Duplicate elimination δ

•  Grouping and aggregation γ

•  Sorting τ

Dan Suciu -- 444 Spring 2010 8
 Relational Algebra (1/3)
The Basic Five operators:
•  Union: ∪
•  Difference: -
•  Selection: σ
•  Projection: Π
•  Join: ⨝

Dan Suciu -- 444 Spring 2010 9

 Relational Algebra (2/3)
Derived or auxiliary operators:
•  Renaming: ρ
•  Intersection, complement
•  Variations of joins
– natural, equi-join, theta join,
semi-join, cartesian product

Dan Suciu -- 444 Spring 2010 10

 Relational Algebra (3/3)
Extensions for bags:
•  Duplicate elimination: δ

•  Group by: γ

•  Sorting: τ

Dan Suciu -- 444 Spring 2010 11

Union and Difference

R1 ∪ R2

R1 – R2

What do they mean over bags ?

Dan Suciu -- 444 Spring 2010 12

 What about Intersection ?
•  Derived operator using minus

R1 ∩ R2 = R1 – (R1 – R2)

•  Derived using join (will explain later)

R1 ∩ R2 = R1 ⨝ R2

Dan Suciu -- 444 Spring 2010 13

 Selection
•  Returns all tuples which satisfy a
condition

σc(R)

•  Examples
–  σSalary > 40000 (Employee)
–  σname = “Smith” (Employee)
•  The condition c can be =, <, ≤, >, ≥, <>
Dan Suciu -- 444 Spring 2010 14
Employee
SSN Name Salary
1234545 John 200000
5423341 Smith 600000
4352342 Fred 500000

σSalary > 40000 (Employee)

SSN Name Salary

5423341 Smith 600000
4352342 Fred 500000
Dan Suciu -- 444 Spring 2010 15
 Projection
•  Eliminates columns

Π A1,…,An (R)

•  Example: project social-security number
and names:
–  Π SSN, Name (Employee)
–  Answer(SSN, Name)

Semantics differs over set or over bags

Dan Suciu -- 444 Spring 2010 16
 Employee
SSN Name Salary
1234545 John 200000
5423341 John 600000
4352342 John 200000
Π Name,Salary (Employee)

Name Salary Name Salary

John 20000 John 20000
John 60000 John 60000
John 20000
Bag semantics
Set semantics

Which is Dan
more efficient to implement ?

Suciu -- 444 Spring 2010 17
 Cartesian Product
•  Each tuple in R1 with each tuple in R2

R1 × R2

•  Very rare in practice; mainly used to

express joins

Dan Suciu -- 444 Spring 2010 18

Employee
Dependent

Name
SSN
EmpSSN
DepName

John
999999999
999999999
Emily

Tony
777777777
777777777
Joe

Employee ✕ Dependent

Name
SSN
EmpSSN
DepName

John
999999999
999999999
Emily

John
999999999
777777777
Joe

Tony
777777777
999999999
Emily

Tony
777777777
777777777
Joe

Dan Suciu -- 444 Spring 2010 19
 Renaming
•  Changes the schema, not the instance

ρ B1,…,Bn (R)

•  Example:
–  ρN, S(Employee)  Answer(N, S)

Dan Suciu -- 444 Spring 2010 20

 Natural Join

R1 ⨝ R2

•  Meaning: R1⨝ R2 = ΠA(σ(R1 × R2))

•  Where:
–  The selection σ checks equality of all
common attributes
–  The projection eliminates the duplicate
common attributes

Dan Suciu -- 444 Spring 2010 21

 Natural Join
R
A B S
B C
X Y Z U
X Z V W
Y Z Z V
Z V

A B C
R ⨝ S =
X Z U
ΠABC(σR.B=S.B(R × S))
X Z V
Y Z U
Y Z V
Z V W
Dan Suciu -- 444 Spring 2010 22
 Natural Join
•  Given the schemas R(A, B, C, D), S(A, C,
E), what is the schema of R ⨝ S ?

•  Given R(A, B, C), S(D, E), what is R ⨝ S ?

•  Given R(A, B), S(A, B), what is R ⨝ S ?

Dan Suciu -- 444 Spring 2010 23

 Theta Join
•  A join that involves a predicate

R1 ⨝θ R2 = σ θ (R1 × R2)

•  Here θ can be any condition

Dan Suciu -- 444 Spring 2010 24

 Eq-join
•  A theta join where θ is an equality

R1 ⨝A=B R2 = σA=B (R1 × R2)

•  This is by far the most used variant of

join in practice

Dan Suciu -- 444 Spring 2010 25

 So Which Join Is It ?
•  When we write R ⨝ S we usually mean
an eq-join, but we often omit the
equality predicate when it is clear from
the context

Dan Suciu -- 444 Spring 2010 26

 Semijoin

R ⋉C S = Π A1,…,An (R ⨝C S)

•  Where A1, …, An are the attributes in R

Dan Suciu -- 444 Spring 2010 27

 Semijoins in Distributed
Databases
Dependent
Employee
EmpSSN DepName Age Stuff
SSN Name Stuff ... ... ... ....
... ... .... network

Employee ⨝SSN=EmpSSN (σ age>71 (Dependent))

Task: compute the query with minimum amount of data transfer

28
 Semijoins in Distributed
Databases
Dependent
Employee
EmpSSN DepName Age Stuff
SSN Name Stuff ... ... ... ....
... ... .... network

Employee ⨝SSN=EmpSSN (σ age>71 (Dependent))

T(SSN) = Π SSN σ age>71 (Dependents)

29
 Semijoins in Distributed
Databases
Dependent
Employee
EmpSSN DepName Age Stuff
SSN Name Stuff ... ... ... ....
... ... .... network

Employee ⨝SSN=EmpSSN (σ age>71 (Dependent))

T(SSN) = Π SSN σ age>71 (Dependents)

R = Employee ⨝SSN=EmpSSN T
= Employee ⋉SSN=EmpSSN (σ age>71 (Dependents))
30
 Semijoins in Distributed
Databases
Dependent
Employee
EmpSSN DepName Age Stuff
SSN Name Stuff ... ... ... ....
... ... .... network

Employee ⨝SSN=EmpSSN (σ age>71 (Dependent))

T(SSN) = Π SSN σ age>71 (Dependents)

R = Employee ⋉SSN=EmpSSN T

Answer = R ⨝SSN=EmpSSN Dependents

31
 Joins R US
•  The join operation in all its variants (eq-
join, natural join, semi-join, outer-join) is
at the heart of relational database
systems

•  WHY ?

Dan Suciu -- 444 Spring 2010 32

 Operators on Bags
•  Duplicate elimination δ

δ(R) = select distinct * from R

•  Grouping γ

γA,sum(B) = select A,sum(B) from R group by A

•  Sorting τ
Dan Suciu -- 444 Spring 2010 33
Complex RA Expressions
γ u.name, count(*)

x.ssn=y.buyer-ssn

y.pid=u.pid

y.seller-ssn=z.ssn

Π ssn Π pid
σname=fred σname=gizmo
Person x Purchase y Person z Product u
Dan Suciu -- 444 Spring 2010 34
 RA = Dataflow Program
•  Several operations, plus strictly
specified order

•  In RDBMS the dataflow graph is always

a tree

•  Novel applications (s.a. PIG), dataflow

graph may be a DAG
Dan Suciu -- 444 Spring 2010 35
 Limitations of RA
•  Cannot compute “transitive closure”
Name1 Name2 Relationship
Fred Mary Father
Mary Joe Cousin
Mary Bill Spouse
Nancy Lou Sister

•  Find all direct and indirect relatives of Fred

•  Cannot express in RA !!! Need to write Java program
•  Remember the Bacon number ? Needs TC too !

Dan Suciu -- 444 Spring 2010 36