Ambo University Woliso Campus

School of Technology and Informatics

Department of Information Technology

Course title: - Advanced Database

Module Name: - Data and Information Management Program: - Regular $ Weekend

Module code: - ITec-M2071 Target Group: - IT 2nd Reg $ 3rd
Wknd
Course Name: - Advanced Database Semester: - II
Course code: - ITec-2072 Content: - Course Handout
Academic year: - 2020/2012 ECTS credits: - 7
Status of the course: - Core Instructor Name: - Husen Adem
Instructor Email: - [email protected] Phone Number: - +251925100878

Course Objectives
At the end of this course you will be able to:
Understand the database query processing and optimization
Know the basics of transaction management
Understand database security
Use different recovery methods when there is a database failure Design a
distributed database system in homogenous and heterogeneous environments
Understand how to give privilege for the user of the database
 Advanced Database Management

AUWC, School of Technology and Informatics Page 2

 Advanced Database Management

Table of Contents
CHAPTER ONE..........................................................................................................................................4
QUERY PROCESSING AND OPTIMIZATION.......................................................................................4
Relational Algebra...................................................................................................................................4
TRANSLATING SQL QUERIES INTO RELATIONAL ALGEBRA...................................................4
QUERY PROCESSING........................................................................................................................11
Heuristic Query Tree Optimization.......................................................................................................14
Summary of Heuristics for Algebraic Optimization:.............................................................................17
Cost-based query optimization:.............................................................................................................18
Semantic Query Optimization:..............................................................................................................19
CHAPTER TWO.......................................................................................................................................20
DATABASE SECURITY AND AUTHORIZATION...............................................................................20
Introduction to Database Security Issues...............................................................................................20
Authentication.......................................................................................................................................20
Authorization/Privilege.........................................................................................................................20
Database Security and the DBA............................................................................................................21
Comparing DAC and MAC...................................................................................................................25
Introduction to Statistical Database Security.........................................................................................25
Types of Cryptosystems........................................................................................................................26
CHAPTER THREE...................................................................................................................................28
TRANSACTION PROCESSING CONCEPTS.........................................................................................28
Introduction to Transaction Processing..................................................................................................28
ACID properties of the transactions......................................................................................................29
SIMPLE MODEL OF A DATABASE..................................................................................................29
Transaction Atomicity and Durability...................................................................................................30
Serializability.........................................................................................................................................33
Transactions as SQL Statements............................................................................................................36
Summary...............................................................................................................................................38
CHAPTER FOUR.....................................................................................................................................40
CONCURRENCY CONTROL TECHNIQUES........................................................................................40
Introduction to Concurrency control techniques....................................................................................40
The Two-Phase Locking Protocol.........................................................................................................41
Deadlock Detection and Recovery.........................................................................................................43
Recovery from Deadlock.......................................................................................................................43
Timestamp-Based Protocols..................................................................................................................44
The Timestamp-Ordering Protocol........................................................................................................45
Purpose of Concurrency Control...........................................................................................................46
CHAPTER FIVE.......................................................................................................................................48
DATABASE RECOVERY TECHNIQUES..............................................................................................48
Recovery Outline and Categorization of Recovery Algorithms.............................................................48
Write-Ahead Logging, Steal/No-Steal, and Force/No-Force.................................................................49

AUWC, School of Technology and Informatics Page 1

 Advanced Database Management
Transaction Actions That Do Not Affect the Database..........................................................................50
Recovery Techniques Based on Immediate Update...............................................................................51
Shadow Paging......................................................................................................................................51
The ARIES Recovery Algorithm...........................................................................................................52
Recovery in Multidata base Systems.....................................................................................................53
Database Backup and Recovery from Catastrophic Failures.................................................................53
CHAPTER SIX.........................................................................................................................................55
DISTRIBUTED DATABASE SYSTEMS................................................................................................55
Distributed Database Concepts..............................................................................................................55
Parallel Versus Distributed Technology................................................................................................55
Additional Functions of Distributed Databases.....................................................................................57
Data Fragmentation...............................................................................................................................58
Data Replication and Allocation............................................................................................................60
Types of Distributed Database Systems.................................................................................................61
Federated Database Management Systems Issues..................................................................................61
Semantic Heterogeneity.........................................................................................................................62
Query Processing in Distributed Databases...........................................................................................63
Data Transfer Costs of Distributed Query Processing.......................................................................63
An Overview of Client-Server Architecture and Its Relationship to Distributed Databases..............64
CHAPTER EIGHT....................................................................................................................................66
WEB- BASED DATABASES..................................................................................................................66
Databases on the World Wide Web.......................................................................................................66
Providing Access to Databases on the World Wide Web......................................................................66
The Web Integration Option of INFORMIX.........................................................................................67
The ORACLE Web Server....................................................................................................................68
Open Problems with Web Databases.....................................................................................................69
CHAPTER NINE......................................................................................................................................71
DATA WAREHOUSING.........................................................................................................................71
What is Data Warehouse?......................................................................................................................71
Understanding a Data Warehouse..........................................................................................................72
Why a Data Warehouse is separated from Operational Databases?.......................................................72
Data Warehouse Applications...........................................................................................................72
Types of Data Warehouse..................................................................................................................72
OLTP vs. OLAP....................................................................................................................................73
Data Warehouse Architecture................................................................................................................74
What is an Aggregation?...................................................................................................................75
What is Data Mining?............................................................................................................................75

AUWC, School of Technology and Informatics Page 2

 Advanced Database Management

CHAPTER ONE
QUERY PROCESSING AND OPTIMIZATION
Relational Algebra
 The basic set of operations for the relational model is known as the relational algebra. These
operations enable a user to specify basic retrieval requests.
 The result of retrieval is a new relation, which may have been formed from one or more
relations.
 The algebra operations thus produce new relations, which can be further manipulated using
operations of the same algebra.
 A sequence of relational algebra operations forms a relational algebra expression, whose result
will also be a relation that represents the result of a database query (or retrieval request).

TRANSLATING SQL QUERIES INTO RELATIONAL ALGEBRA

Query block
 The basic unit that can be translated into the algebraic operators and optimized.
 A query block contains a single SELECT-FROM-WHERE expression, as well as GROUP BY
and HAVING clause if these are part of the block.
 Nested queries within a query are identified as separate query blocks.
 Aggregate operators in SQL must be included in the extended algebra
SELECT LNAME, FNAME FROM EMPLOYEE WHERE SALARY > (SELECT MAX (SALARY)
FROM EMPLOYEE WHERE DNO = 5);

SELECT LNAME, FNAME FROM EMPLOYEE SELECT MAX (SALARY) FROM

WHERE SALARY > C EMPLOYEE WHERE DNO = 5

π LNAME, FNAME (σ SALARY>C(EMPLOYEE)) ℱ MAX SALARY (σ DNO=5 (EMPLOYEE))

Relational Algebra consists of several groups of operations

1. Unary Relational Operations
 SELECT (symbol: σ (sigma))  RENAME (symbol: ρ (rho))
 PROJECT (symbol: π (pi))
2. Relational Algebra Operations from Set Theory
 UNION (∪), INTERSECTION (∩), DIFFERENCE (or MINUS, –)
 CARTESIAN PRODUCT (X)
3. Binary Relational Operations
 JOIN (several variations of JOIN exist)
 DIVISION

AUWC, School of Technology and Informatics Page 3

 Advanced Database Management
4. Additional Relational Operations
 OUTER JOINS, OUTER UNION
 AGGREGATE FUNCTIONS (These compute summary of information: for
example, SUM, COUNT, AVG, MIN, MAX)
1. Unary Relational Operations
A) SELECT Operation
 SELECT operation is used to select a subset of the tuples from a relation that satisfy a selection
condition.
 It is a filter that keeps only those tuples that satisfy a qualifying condition those satisfying the
condition are selected while others are discarded.
 Example: To select the EMPLOYEE tuples whose department number is four or those whose
salary is greater than $30,000 the following notation is used:
σ DNO=4(EMPLOYEE) Employee is table name
σ SALARY>30,000(EMPLOYEE)
 In general, the selection operation is denoted by σ<selection condition>(R) where the symbol σ
(Sigma) is used to denote the select operator and the selection condition is a Boolean expression
specified on the attributes of relation(R).

What is the SQL statement of the above unary relational operations?

SELECT * FROM EMPLOYEE WHERE DNO=”4” AND SALARY>30000;

SELECT Operation Properties

The SELECT operations σ<selection condition>(R) Produces a relation S that has the same schema as R.
 The SELECT operations σ is commutative i.e. σ<condition1> (σ<condition2>(R)) =
σ<condition2> (σ<condition1>(R)).
 A cascade SELECT operation may be applied in any order i.e. σ<condition1> (σ<condition2> (σ
<condition3>(R)) = σ<condition2> (σ<condition3> (σ <condition1>(R)))
 A cascade SELECT operation may be replaced by a single selection with conjunctions of all the
conditions. i.e. σ<condition1> (σ<condition2> (σ <condition3>(R)) = σ<condition1> AND
σ<condition2> AND σ <condition3>(R)))
What are the results of the following select and projection operation? And what is its SQL Statement?
 Results of SELECT and PROJECT operations
A. σ (DNO=4 AND SALARY>25000) OR (DNO=5 AND SALARY>30000) (EMPLOYEE)
B. π LNAME, FNAME, SALARY(EMPLOYEE) π SEX, SALARY(EMPLOYEE)
The SQL statement of the above relational algebra is the following.
A. SELECT * FROM EMPLOYEE WHERE DNO='4' AND SALARY>2500 OR
DNO='5' AND SALARY>30000;
B. SELECT FIRSTNAME, LASTNAME, SALARY FROM EMPLOYEE;
C. SELECT SEX, SALARY FROM EMPLOYEE;
The result of the above SQL statement is:

AUWC, School of Technology and Informatics Page 4

 Advanced Database Management

B) PROJECT Operation
 This operation selects certain columns from the table and discards the other columns.
 The PROJECT creates a vertical partitioning one with the needed columns (attributes
containing results of the operation and other containing the discarded Columns.
 Example: To list each employee’s first and last name and salary, the following is used:
π LAME, FNAME, SALARY (EMPLOYEE)
 The general form of the project operation is: π<attribute list>(R)
 π (pi) is the symbol used to represent the project operation<attribute list> is the desired list of
attributes from relation R.
 The project operation removes any duplicate tuples. This is because the result of the project
operation must be a set of tuples. Mathematical sets do not allow duplicate elements.
PROJECT Operation Properties
 The number of tuples in the results of projection π<attribute list>(R) is always less or equal to the
number of tuples in R. If the lists of attributes include a key of R, then the number of tuples is
equal to the number of tuples in R. π<list> (π<list2>(R)) = π<list>(R) as long as <list2> contains
the attributes in <list2>.
2. Relational Algebra Operations from Set Theory
A. UNION OPERATION
 The result of this operation, denoted by RUS is a relation that includes all tuples that are either in R or
in S or in both R and S. Duplicate tuples is eliminated.
 Example: To retrieve the social security numbers of all employees who either work in department 5
or directly supervise an employee who works in department 5, we can use the union operation as
follows:
DEP5_EMPS σ DNO=5(EMPLOYEE)
RESULT1 π SSN (DEP5_EMPS)
RESULT2 (SSN) π SUPERSSN (DEP5_EMPS)
RESULT RESULT1 U RESULT2
 The union operation produces the tuples that are in either RESULT1 or RESULT2 or both.
 The two operands must be “type compatible”.
TYPE COMPATABILITY

AUWC, School of Technology and Informatics Page 5

 Advanced Database Management
 Type Compatibility of operands is required for the binary set operation UNION ∪, (also for
INTERSECTION ∩, and SET DIFFERENCE -
 R1(A1, A2, ..., An) and R2(B1, B2, ..., Bn) are type compatible if:
 They have the same number of attributes, and
 The domains of corresponding attributes are type compatible (i.e. dom(Ai)=dom(Bi) for
i=1, 2, ..., n).
 The resulting relation for R1∪R2 (also for R1∩R2, or R1–R2) has the same attribute names as the
first operand relation R1 (by convention)
Instructor Student

Instructor U Student

SELECT * FROM student UNION ALL SELECT * FROM instructor;

B. INTERSECTION OPERATION
The result of this operation denoted by R n S is a relation that includes all tuples that are in both R and
S.
The two operands must be "type compatible “. Example: The result of the intersection operation (figure
below) includes only those who are both students and instructors.

Customers Customer

C. Set Difference (or MINUS) Operation

 The result of this operation, denoted by R - S, is a relation that includes all tuples that are in R
but not in S. The two operands must be "type compatible”.
 Note: both union and intersection are commutative operations; i.e. RUS=SUR and R n S = S n R
 Both union and intersection can be treated as n-ary operations applicable to any number of
relations as both are associative operations; that is RU(SUT)=(RUS)UT and (R n S) n =R n (S n
T)
 The minus operation is not commutative, that is in general. R—S? S--R
D. CARTESIAN (or cross product) Operation
 This operation is used to combine tuples from two relations in a combinational fashion.

AUWC, School of Technology and Informatics Page 6

 Advanced Database Management
 In general, the result of R(A1, A2,….An) x S(B1, B2,….,Bm) is a relation Q with degree n+m
attributes Q(A1, A2,….An, B1, B2,….,Bm), in that order.
 The resulting relation Q has one tuple for each combination of tuples—one from R and one from
S. Hence, if R has n R tuples (denoted as |R|= n R), and S has nS tuples, then |RXS| will have n R*nS
tuples.
 The two operands do not have to be “type compatible”.
Example FEMALE_EMPS σ SEX = ‘F’(EMPLOYEE)
EMPNAMES π FNAME, LNAME,
SSN(FEMALE_EMPS)
EMP_DEPENDENTS EMPNAMES X DEPENDENT
3. Binary Relational Operations
A. JOIN Operation
 The sequence of Cartesian product followed by select is used quite commonly to identify and
select related tuples from two relations, a special operation, called JOIN.
 It is denoted by a
 This operation is very important for any relational database with more than a single relation,
because it allows us to process relationships among relations.
 The general form of a join operation on two relations R (A1, A2, …, An) and S (B1, B2,… Bn) is
R <join condition> S. Where R and S can be any relations that result from general
relational algebra expressions.
 EXAMPLE: - Suppose that we want to retrieve the name of the manager of each
Department. To get the managers name we need to combine each DEPARTMENT tuple with
the EMPLOYEE tuple whose SSN value matches the MGRSSN value in the department
tuple. We do this by using the join operation. DEPT_MGR DEPARTMENT
EMPLOYEE MGRSSN=SSN
B. EQUIJOIN Operation
 The most common use of join involves join conditions with equality comparisons only.
 Such a join, where the only comparison operator used is =, is called an EQUIJOIN.
 In the result of an EQUIJOIN we always have one or more pairs of attributes (whose names need
not be identical) that have identical values in every tuple.
 The JOIN seen in the previous example was EQUIJOIN.
C. NATURAL JOIN Operation
 Because one of each pair of attributes with identical values is superfluous, a new operation called
natural join—denoted by *—was created to get rid of the second (superfluous) attribute in an
EQUIJOIN condition.
 The standard definition of natural join requires that the two join attributes, or each pair of
corresponding join attributes, have the same name in both relations.
 If this is not the case, a renaming operation is applied first.
EXAMPLE: To apply a natural join on the DNUMBER attributes of DEPARTMENT and
DEPT_LOCATIONS, it is sufficient to write: DEPT_LOCS DEPARTMENT *
DEPT_LOCATIONS

AUWC, School of Technology and Informatics Page 7

 Advanced Database Management
 The set of operations including select, project, union(u), set difference (-), and Cartesian
product(x) is called a complete set because any other relational algebra expression can be
expressed by a combination of these five operations. Example R n S =(RUS)-((R-S) U(S-R))
4. Additional Relational Operations
A. Aggregate Functions and Grouping
 A type of request that cannot be expressed in the basic relational algebra is to specify
mathematical aggregate functions on collections of values from the database.
 Examples of such functions include retrieving the average or total salary of all employees or the
total number of employee tuples.
 These functions are used in simple statistical queries that summarize information from the
database tuples.
 Common functions applied to collections of numeric values include SUM, AVERAGE,
MAXIMUM, and MINIMUM.
 The COUNT function is used for counting tuples or values.
Use of the Functional operator F
 F MAX Salary(Employee) retrieves the maximum salary value from the Employee relation
 F MIN Salary(Employee) retrieves the minimum Salary value from the Employee relation
 F SUM Salary(Employee) retrieves the sum of the Salary from the Employee relation
 DNO F COUNT SSN, AVERAGE Salary(Employee) groups employees by DNO (department number) and
computes the count of employees and average salary per department.[ Note: count just counts the
number of rows, without removing duplicates]
What is Query Processing?
 The Steps required transforming high level SQL query into a correct and “efficient” strategy for
execution and retrieval. The activities involved in parsing, validating, optimizing, and
executing a query.
What is the aim of query processing?
 To transform a query written in a high-level language, typically SQL, into a correct and
efficient execution strategy expressed in a low-level language (implementing the relational
algebra), and to execute the strategy to retrieve the required data.
What is Query Optimization?
 The activity of choosing a single “efficient” execution strategy (from hundreds) as determined
by database catalog statistics for processing a query. An important aspect of query processing is
query optimization.
What is the aim of query Optimization?
 As there are many equivalent transformations of the same high-level query, the aim of query
optimization is to choose the one that minimizes resource usage.
 Generally, we try to reduce the total execution time of the query, which is the sum of the
execution times of all individual operations that make up the query.
Examples for query Optimization: Identify all managers who work in a London branch
SQL: - SELECT * FROM Staff s, Branch b WHERE s.branchNo = b.branchNo AND s.position =
‘Manager’ AND b.city = ‘london’;
Results in these equivalent relational algebra statements
(1) s(position=‘Manager’)^(city=‘London’)^(Staff.branchNo=Branch.branchNo) (Staff X Branch)
(2) s(position=‘Manager’)^(city=‘London’) (Staff Staff.branchNo = Branch.branchNo Branch)

AUWC, School of Technology and Informatics Page 8

 Advanced Database Management

(3) [s(position=‘Manager’) (Staff)] Staff.branchNo = Branch.branchNo [s(city=‘London’) (Branch)]

Assume:
 1000 tuples in Staff.  All temporary results are written
 50 Managers back to disk (memory is small)
 50 tuples in Branch.  Tuples are accessed one at a
 5 London branches time (not in blocks)
 No indexes or sort keys
Query 1 (Bad)

 Requires (1000+50) disk accesses to read from Staff and Branch relations
 Creates temporary relation of Cartesian Product (1000*50) tuples
 Requires (1000*50) disk access to read in temporary relation and test predicate
Total Work = (1000+50) + 2*(1000*50) = 101,050 I/O operations
Query 2 (Better)

 Again requires (1000+50) disk accesses to read from Staff and Branch
 Joins Staff and Branch on branchNo with 1000 tuples (1 employee : 1 branch )
 Requires (1000) disk access to read in joined relation and check predicate
Total Work = (1000+50) + 2*(1000) = 3050 I/O operations
3300% Improvement over Query 1
Query 3 (Best)

 Read Staff relation to determine ‘Managers’ (1000 reads)

 Create 50 tuple relation(50 writes)
 Read Branch relation to determine ‘London’ branches (50 reads)
 Create 5 tuple relation (5 writes)
 Join reduced relations and check predicate (50 + 5 reads)
Total Work = 1000 + 2*(50) + 5 + (50 + 5) = 1160 I/O operations
8700% Improvement over Query 1
Two main Techniques for Query Optimization
1. Heuristic Rules: -Rules for ordering the operations in query optimization.
2. Systematical estimation: -It estimates cost of different execution strategies and chooses the
execution plan with lowest execution cost.
Steps in Processing High-Level Query

AUWC, School of Technology and Informatics Page 9

 Advanced Database Management

Scanning, Parsing, Validating

 Scanner: The scanner identifies the language tokens such as SQL Keywords, attribute names,
and relation names in the text of the query.
 Parser: The parser checks the query syntax to determine whether it is formulated according to
the syntax rules of the query language.
 Validation: The query must be validated by checking that all attributes and relation names are
valid and semantically meaningful names in the schema of the particular database being queried.
QUERY PROCESSING
 Query Optimization: The process of choosing a suitable execution strategy for processing a
query. This module has the task of producing an execution plan.
 Query Code Generator: It generates the code to execute the plan.
 Runtime Database Processor: It has the task of running the query code whether in compiled or
interpreted mode. If a runtime error results an error message is generated by the runtime
database processor.
Query Processing Steps

Processing can be divided into Decomposition, Optimization, Execution, and Code generation main
categories.
1. Query Decomposition
 It is the process of transforming a high-level query into a relational algebra query, and to check
that the query is syntactically and semantically correct.
 It Consists of parsing and validation
Typical stages in query decomposition are:

AUWC, School of Technology and Informatics Page 10

 Advanced Database Management
1. Analysis:
 Lexical and syntactical analysis of the query (correctness) based on attributes, data type...
 Query tree will be built for the query containing
 leaf node for base relations,
 one or many non-leaf nodes for relations produced by relational algebra operations and
 Root node for the result of the query.
 Sequence of operation is from the leaves to the root. (SELECT * FROM Catalog c, Author a
Where a.authorid = c.authorid AND c.price>200 AND a.country= ‘ USA’ )
2. Normalization:
 Convert the query into a normalized form.
 The predicate WHERE will be converted to Conjunctive (∨) or Disjunctive (∧) Normal form.
3. Semantic Analysis
 To reject normalized queries that is not correctly formulated or contradictory.
 Incorrect if components do not contribute to generate result.
 Contradictory if the predicate cannot be satisfied by any tuple.
 Say for example, (Catalog = “BS”  Catalog= “CS”) since a given book can only be classified in
either of the category at a time
4. Simplification
 to detect redundant qualifications,
 eliminate common sub-expressions, and
 Transform the query to a semantically equivalent but more easily and effectively computed
form.
 For example, if a user doesn’t have the necessary access to all of the objects of the query, it
should be rejected.
2. Query Optimization
 Everyone wants the performance of their database to be optimal.
 In particular, there is often a requirement for a specific query or object that is query based, to run
faster.
 Problem of query optimization is to find the sequence of steps that produces the answer to user
request in the most efficient manner, given the database structure.
 The performance of a query is affected by
 the tables or queries that underlies the query and
 The complexity of the query.
 Given a request for data manipulation or retrieval, an optimizer will choose an optimal plan for
evaluating the request from among the manifold alternative strategies. i.e. there are many ways
(access paths) for accessing desired file/record. Hence, DBMS is responsible to pick the best
execution strategy based on various considerations (Least amount of I/O and CPU resources.
Transformational rules for relational Algebra
1. Cascade of SELECTON: conjunctive SELCTION operations can cascade into individual
selection operations and vice versa.

2. Commutativity of SELECTION operations

AUWC, School of Technology and Informatics Page 11

 Advanced Database Management
3. Cascade of PROJECTION: in the sequence of PROJECTION Operations, only last in the
sequence is required.

4. Commutativity of SELECTION with PROJECTION and vice versa

a. If the predicate C1 involves only the attributes in the projection list(L1), then the
selection and projection operations commute

5. Commutativity of THETA JOIN/ Cartesian product RXS is equivalent to SXR. Also holds for
Equi-join and Natural join

6. Commutativity of SELECTION with THETA JOIN

a. If the predicate c1 involves only attributes of one of the relations (R) being joined, then
the selection and join operations commute.

b. If the predicate is in the forms of c1^ c2 and c1 involves only attributes of R and c2
involves only attributes of S, then the selection and theta join operations commute.

7. Commutativity of PROJECTION and THETA JOIN

If the projection list of the form L1 L2, where L1 involves only attributes of R and L2 involves
only attributes of S being joined, and the predicate c involve only attributes in the projection list,
then the SELECTION and JOIN operations commute.

8. Commutativity of the set operations: UNION and INTERSECTION but not STE DIFFERENCE

9. Associativity of THETA JOIN, CARTESIAN PRODUCT, UNION and INTERSECTION

10. Commuting SELECTION with SET OPERATIONS

11. Commuting SELECTION with UNION

Heuristic approach will be implemented by using the above transformation rules in the following
sequence or steps. Sequence for applying Transformation Rules
1. Use Rule 1—Cascade Selection
2. Use
Rule 2. Commutatively of SELECTION
Rule 4. Commuting SELECTION with PROJECTION
Rule 6 Commuting SELECTION with JOIN and CARTESIAN
Rule 10 Commuting SELECTION with SET OPERATIONS

AUWC, School of Technology and Informatics Page 12

 Advanced Database Management
3. Use: Rule 9 Associativity of binary operations (JOIN, CARTESIAN, UNION and
INTERSECTION). Rearrange nodes by making the most restrictive operations to be performed
first (moving it as far down the tree as possible)
4. Perform Cartesian Operations with the subsequent selection operation
5. Use Rule 3 Cascade of PROJECTION
Rule 4 Commuting PROJECTION with SELECTION
Rule 7 Commuting PROJECTION with JOIN and CARTESIAN
Rule 11 Commuting PROJECTION with UNION
Heuristic Query Tree Optimization
 It has some rules which utilize equivalence expressions to transform the initial tree into final,
optimized query tree.
 Process for heuristics optimization
1. The parser of a high-level query generates an initial internal representation;
2. Apply heuristics rules to optimize the internal representation.
3. A query execution plan is generated to execute groups of operations based on the access
paths available on the files involved in the query.
A) The main heuristic is to apply first the operations that reduce the size of intermediate
results.
E.g. Apply SELECT and PROJECT operations before applying the JOIN or other binary
operations.
 The main idea behind is to reduce intermediate results. This includes performing
 SELECT operation to reduce the number of tuples &
 PROJECT operation to reduce number of attributes.
Query tree:
 A tree data structure that corresponds to a relational algebra expression. It represents the input
relations of the query as leaf nodes of the tree, and represents the relational algebra operations as
internal nodes.
 Example: For every project located in ‘Stafford’, retrieve the project number, the controlling
department number and the department manager’s last name, address and birthdate.
 SQL query: Q2: SELECT P. NUMBER, P.DNUM,E.LNAME,E.ADDRESS, E.BDATE FROM
PROJECT AS P,DEPARTMENT AS D, EMPLOYEE AS E WHERE P.DNUM=D.DNUMBER
AND D.MGRSSN=E.SSN AND P.PLOCATION=‘STAFFORD’;
 Relation algebra: P PNUMBER, DNUM, LNAME, ADDRESS, BDATE (((s PLOCATION=‘STAFFORD’ (PROJECT))
DNUM=DNUMBER (DEPARTMENT)) MGRSSN=SSN (EMPLOYEE))
 The same query could correspond too many different relational algebra expressions and hence
many different query trees.

AUWC, School of Technology and Informatics Page 13

 Advanced Database Management

Query graph:
 A graph data structure that corresponds to a relational calculus expression. It does not indicate an
order on which operations to perform first. Nodes represent Relations. Ovals represent constant
nodes. Edges represent Join & Selection conditions. Attributes to be retrieved from relations
represented in square brackets.
 Drawback: - Does not indicate an order on which operations are performed first.
 There is only a single graph corresponding to each query.

Example 2 of Heuristic Optimization of Query Trees:

 The same query could correspond too many different relational algebra expressions and hence
many different query trees.
 The task of heuristic optimization of query trees is to find a final query tree that is efficient to
execute.
 For Example : Q:-SELECT LNAME FROM EMPLOYEE, WORKS_ON, PROJECT WHERE
PNAME= ‘AQUARIUS’ AND PNUMBER=PNO AND ESSN=SSN AND BDATE > ‘1957-12-
31’

AUWC, School of Technology and Informatics Page 14

 Advanced Database Management

Fig 1: Initial (canonical) query tree for SQL query Q

Fig2: move select down the tree using cascade & commutatively rule of select operation

Fig3: rearrange of leaf nodes, using commutatively & associativity of binary operations.

AUWC, School of Technology and Informatics Page 15

 Advanced Database Management

Fig 4: converting select & Cartesian product into join

Fig 5: break-move of project using cascade & commuting rules of project operations.
Query Execution Plans
 An execution plan for a relational algebra query consists of a combination of the relational
algebra query tree and information about the access methods to be used for each relation as well
as the methods to be used in computing the relational operators stored in the tree.
 Materialized evaluation: the result of an operation is stored as a temporary relation.
 Pipelined evaluation: as the result of an operator is produced, it is forwarded to the next operator
in sequence.
Summary of Heuristics for Algebraic Optimization:
1) The main heuristic is to apply first the operations that reduce the size of intermediate results.
2) Perform select operations as early as possible to reduce the number of tuples and perform project
operations as early as possible to reduce the number of attributes. (This is done by moving select
and project operations as far down the tree as possible.)
3) The select and join operations that are most restrictive should be executed before other similar
operations. (This is done by reordering the leaf nodes of the tree among themselves and adjusting
the rest of the tree appropriately.)

AUWC, School of Technology and Informatics Page 16

 Advanced Database Management

Cost-based query optimization:

 The optimizer examines alternative access paths and operator algorithms and chooses the
execution plan with lowest estimate cost.
 The query cost is calculated based on the estimated usage of resources such as I/O, CPU and
memory needed.
 Application developers could specify hints to the ORACLE query optimizer.
 The idea is that an application developer might know more information about the data.
 Issues of cost-based query optimization are.
 Cost function
 Number of execution strategies to be considered
 Cost Components for Query Execution
 Access cost to secondary  Computation cost
storage  Memory usage cost
 Storage cost  Communication cost
 Note: Different database systems may focus on different cost components.
 Catalog Information Used in Cost Functions
 Information about the size of a  record size (R),
file  number of blocks (b)
 number of records (tuples) (r),  blocking factor (bfr)
 Information about indexes and indexing attributes of a file
 Number of levels (x) of each multilevel  Number of distinct values (d) of an
index attribute
 Number of first-level index blocks (bI1)  Selectivity (sl) of an attribute
 Selection cardinality (s) of an attribute. (s = sl * r)
1. Access Cost of Secondary Storage
 Data is going to be accessed from secondary storage, as a query will need some part of the data
stored in the database. The disk access cost can again be analyzed in terms of:
 Searching  Writing, data blocks used to store
 Reading, and some portion of a relation.
 Remark: The disk access cost will vary depending on
 The file organization used and the access method implemented for the file organization.
 Whether the data is stored contiguously or in scattered manner, will affect the disk access
cost.
2. Storage Cost
 While processing a query, as any query would be composed of many database operations, there
could be one or more intermediate results before reaching the final output.
 These intermediate results should be stored in primary memory for further processing.
 The bigger the intermediate relation, the larger the memory requirement, which will have impact
on the limited available space.
3. Computation Cost
 Query is composed of many operations. The operations could be database operations like reading
and writing to a disk, or mathematical and other operations like:
 Searching  Merging
 Sorting  Computation on field values

AUWC, School of Technology and Informatics Page 17

 Advanced Database Management
4. Communication Cost
 In most database systems the database resides in one station and various queries originate from
different terminals.
 This will have impact on the performance of the system adding cost for query processing.
 Thus, the cost of transporting data between the database site and the terminal from where the
query originate should be analyzed.
3. Query Execution Plans
 An execution plan for a relational algebra query consists of a combination of the relational
algebra query tree and information about the access methods to be used for each relation as well
as the methods to be used in computing the relational operators stored in the tree.
Semantic Query Optimization:
 Uses constraints specified on the database schema in order to modify one query into another
query that is more efficient to execute. Consider the following SQL query, SELECT E.
LNAME, M. LNAME FROM EMPLOYEE E M WHERE E. SUPERSSN=M.SSN AND E.
SALARY>M.SALARY
Explanation:
 Suppose that we had a constraint on the database schema that stated that no employee can earn
more than his or her direct supervisor.
 If the semantic query optimizer checks for the existence of this constraint, it need not execute the
query at all because it knows that the result of the query will be empty. Techniques known as
theorem proving can be used for this purpose.
Exercise
I. Select the alternative answer from the given letter option
1. Among the list below one is not cost based optimization

A) Cost function C) Computational cost

B) Number of execution strategies to be D) None of above
considered

2. The basic set of operations for the relational model is known as the _________________

A) Relational algebra. C) Relational query

B) Relational models D) Relational theory

3. Nested queries within a query are identified as separate ______________________.

A) query blocks C) query invention

B) query summary D) query control

4. _____________ operation is used to select a subset of the tuples from a relation that satisfy a
selection condition.
A) Delete C) Update
B) Select D) Project
5. Which one is Binary Relational Operations?
A) Selection C) Join
B) Deletion D) Projection

AUWC, School of Technology and Informatics Page 18

 Advanced Database Management
6. Which one is Relational Algebra Operations from Set Theory?
A) Select C) Rename
B) Project D) Cartesian product
7. Which one is unary relational algebra?

A) Cartesian product C) Division

B) Rename D) Projection

8. Which one is not catalog information used in cost function used to exhibit relational algebra?

A) Record size C) Cost function

B) Number of blocks D) Blocking factor

9. Which one is not part of information about indexes and indexing attributes of a file?

A) Number of levels C) Selection cardinality

B) Selectivity D) Number of blocks

10. The disk access cost can again be analyzed in terms of:

A) Searching C) Writing
B) Reading D) All can be answer

II. Write short answer for the following question

1. To increase your understanding try to generalize relational algebra in terms of database!
2. Illustrate query processing and explain each step undoubtable?
3. How you define heuristic query optimizations?

AUWC, School of Technology and Informatics Page 19

 Advanced Database Management

CHAPTER TWO
DATABASE SECURITY AND AUTHORIZATION
Introduction to Database Security Issues
 In today's society, some information is extremely important that needs to be protected.
 For example, disclosure or modification of military information could cause danger to
national security.
 A good database security management system has to handle the possible database threats.
 Threat may be any situation or event, whether intentional (planned) or accidental, that may
adversely affect a system and consequently the organization.
 Threats to databases: It may result in degradation of some/all security goals like;
 Loss of Integrity
 Only authorized users should be allowed to modify data.
 For example, students may be allowed to see their grades, but not allowed to modify
them.
 Loss of Availability-if DB is not available for those users/ to which they have a legal right to
uses the data.
 Authorized users should not be denied access.
 For example, an instructor who wishes to change a grade should be allowed to do so.
 Loss of Confidentiality
 Information should not be disclosed to unauthorized users.
 For example, a student should not be allowed to examine other students' grades.
Authentication
 All users of the database will have different access levels and permission for different data
objects.
 Authentication is the process of checking whether the user is the one with the privilege for the
access level. Thus, the system will check whether the user with a specific username and
password is trying to use the resource.
Authorization/Privilege
 Authorization refers to the process that determines the mode in which a particular (previously
authenticated) client is allowed to access a specific resource controlled by a server.
 Any database access request will have the following three major components.
1. Requested Operation: what kind of operation is requested by a specific query?
2. Requested Object: on which resource or data of the database is the operation sought to
be applied?
3. Requesting User: who is the user requesting the operation on the specified object?
Forms of user authorization
There are different forms of user authorization on the resource of the database. These include:
1. Read Authorization: the user with this privilege is allowed only to read the content of the
data object.
2. Insert Authorization: the user with this privilege is allowed only to insert new records or
items to the data object.
3. Update Authorization: users with this privilege are allowed to modify content of attributes
but are not authorized to delete the records.
4. Delete Authorization: users with this privilege are only allowed to delete a record and not
anything else.

AUWC, School of Technology and Informatics Page 20

 Advanced Database Management
Note: Different users, depending on the power of the user, can have one or the combination of the above
forms of authorization on different data objects.
Database Security and the DBA
 The database administrator (DBA) is the central authority for managing a database system.
 The DBA’s responsibilities include
 Account creation
 granting privileges to users who need to use the system
 Privilege revocation
 classifying users and data in accordance with the policy of the organization
Access Protection, User Accounts, and Databases Audits
 Whenever a person or group of persons need to access a database system, the individual or group
must first apply for a user account.
 The DBA will then create a new account id and password for the user if he/she believes there is a
legitimate need to access the database.
 The user must log in to the DBMS by entering account id and password whenever database access
is needed.
 The database system must also keep track of all operations on the database that are applied by a
certain user throughout each login session.
 If any tampering with the database is assumed, a database audit is performed
 A database audit consists of reviewing the log to examine all accesses and operations
applied to the database during a certain time period.
 A database log that is used mainly for security purposes is sometimes called an audit trail.
 To protect databases against the possible threats two kinds of countermeasures can be
implemented:
1. Access control, and 2.Encryption
Access Control (AC)
1. Discretionary Access Control (DAC)
 The typical method of enforcing discretionary access control in a database system is based on the
granting and revoking privileges.
 The granting and revoking of privileges for discretionary privileges known as the access matrix
model where
 The rows of a matrix M represents subjects (users, accounts, programs)
 The columns represent objects (relations, records, columns, views, operations).
 Each position M(i,j) in the matrix represents the types of privileges (read, write, update)
that subject i holds on object j.
 To control the granting and revoking of relation privileges, each relation R in a database is
assigned an owner account, which is typically the account that was used when the relation was
created in the first place.
 The owner of a relation is given all privileges on that relation.
 The owner account holder can pass privileges on any of the owned relation to other users
by granting privileges to their accounts.
Privileges Using Views
 The mechanism of views is an important discretionary authorization mechanism in its own right.

AUWC, School of Technology and Informatics Page 21

 Advanced Database Management
 For example, If the owner A of a relation R wants another account B to be able to
retrieve only some fields of R, then A can create a view V of R that includes only those
attributes and then grant SELECT on V to B.
Revoking Privileges
 In some cases, it is desirable to grant a privilege to a user temporarily.
 For example, the owner of a relation may want to grant the SELECT privilege to a user for a
specific task and then revoke that privilege once the task is completed. Hence, a mechanism for
revoking privileges is needed.
 In SQL, a REVOKE command is included for the purpose of canceling privileges.
Propagation of Privileges using the GRANT OPTION
 Whenever the owner A of a relation R grants a privilege on R to another account B, privilege
can be given to B with or without the GRANT OPTION.
 If the GRANT OPTION is given, this means that B can also grant that privilege on R to other
accounts.
 Suppose that B is given the GRANT OPTION by A and that B then grants the privilege on R
to a third account C, also with GRANT OPTION.
 In this way, privileges on R can propagate to other accounts without the knowledge of the
owner of R.
 If the owner account A now revokes the privilege granted to B, all the privileges that B
propagated based on that privilege should automatically be revoked by the system.
Example 1
 Suppose that the DBA creates four accounts: A1, A2, A3, A4 and wants only A1 to be able to
create relations. Then the DBA must issue the following GRANT command in SQL: -
GRANT CREATE TABLE TO A1;
Example 2
 Suppose that A1 creates the two base relations EMPLOYEE and DEPARTMENT. A1 is then
owner of these two relations and hence A1 has all the relation privileges on each of them.
 Suppose that A1 wants to grant A2 the privilege to insert and delete rows in both of these
relations, but A1 does not want A2 to be able to propagate these privileges to additional accounts:
 GRANT INSERT, DELETE ON EMPLOYEE, DEPARTMENT TO A2;
Example 3
 Suppose that A1 wants to allow A3 to retrieve information from either of the table (Department
or Employee) and also to be able to propagate the SELECT privilege to other accounts. A1 can
issue the command:
 GRANT SELECT ON EMPLOYEE, DEPARTMENT TO A3 WITH GRANT
OPTION;
 A3 can grant the SELECT privilege on the EMPLOYEE relation to A4 by issuing:
 GRANT SELECT ON EMPLOYEE TO A4;
 Notice that A4 can’t propagate the SELECT privilege because GRANT OPTION was not given
to A4
Example 4
Suppose that A1 decides to revoke the SELECT privilege on the EMPLOYEE relation from A3;
A1 can issue:
 REVOKE SELECT ON EMPLOYEE FROM A3;

AUWC, School of Technology and Informatics Page 22

 Advanced Database Management
 The DBMS must now automatically revoke the SELECT privilege on EMPLOYEE from A4, too,
because A3 granted that privilege to A4 and A3 does not have the privilege any more.
Example 5
 Suppose that A1 wants to give back to A3 a limited capability to SELECT from the EMPLOYEE
relation and wants to allow A3 to be able to propagate the privilege. The limitation is to retrieve
only the NAME, BDATE, and ADDRESS attributes and only for the tuples with DNO=5.
 A1 then create the view:
 CREATE VIEW A3 EMPLOYEE AS SELECT NAME, BDATE, ADDRESS
FROM EMPLOYEE WHERE DNO = 5;
 After the view is created, A1 can grant SELECT on the view A3 EMPLOYEE to A3 as follows:
 GRANT SELECT ON A3 EMPLOYEE TO A3 WITH GRANT OPTION;
Example 6
Finally, suppose that A1 wants to allow A4 to update only the SALARY attribute of
EMPLOYEE; A1 can issue:
 GRANT UPDATE ON EMPLOYEE (SALARY) TO A4;
2 Mandatory Access Control (MAC)
 DAC techniques are an all-or-nothing method: A user either has or does not have a certain privilege.
 In many applications, additional security policy is needed that classifies data and users based on
security classes.
 Typical security classes are top secret (TS), secret (S), confidential (C), and unclassified (U),
where TS is the highest level and U the lowest: TS ≥ S ≥ C ≥ U
 The commonly used model for multilevel security, known as the Bell-LaPadula model,
classifies each subject (user, account, program) and object (relation, tuple, column, view,
operation) into one of the security classifications, TS, S, C, or U:
 Clearance (classification) of a subject S as class(S) and to the classification of an
object O as class (O).

 Two restrictions are enforced on data access based on the subject/object classifications:

 A subject S is not allowed read access to an object O unless class(S) ≥ class (O).
 A subject S is not allowed to write an object O unless class(S) ≤ class (O).
 To incorporate multilevel security notions into the relational database model, it is common to
consider attribute values and rows as data objects. Hence, each attribute A is associated with a
classification attribute C in the schema.
 In addition, in some models, a tuple classification attribute TC is added to the relation attributes
to provide a classification for each tuple as a whole.
 Hence, a multilevel relation schema R with n attributes would be represented as
 R(A1,C1,A2,C2, …, An,Cn,TC) where each Ci represents the classification attribute
associated with attribute Ai.
 The value of the TC attribute in each tuple t – which is the highest of all attribute classification
values within t – provides a general classification for the tuple itself.
 Whereas, each Ci provides a finer security classification for each attribute value within the tuple.
 A multilevel relation will appear to contain different data to subjects (users) with different
clearance levels.

AUWC, School of Technology and Informatics Page 23

 Advanced Database Management
 In some cases, it is possible to store a single tuple in the relation at a higher classification
level and produce the corresponding tuples at a lower-level classification through a
process known as filtering.
 In other cases, it is necessary to store two or more tuples at different classification levels
with the same value for the apparent key.
 This leads to the concept of poly instantiation where several tuples can have the same apparent
key value but have different attribute values for users at different classification levels.
 Example. Consider query SELECT * FROM employee

a. The original employee table,

b. After filtering employee table for classification C users,
c. After filtering employee table for classification U users
d. Poly instantiation of the smith row for C users who want to modify some value

 A user with a security clearance S would see the same relation shown above (a) since all row
classification are less than or equal to S as shown in (a).
 However, a user with security clearance C would not allow to see values for salary of Brown and
job performance of Smith, since they have higher classification as shown in (b)
 For a user with security clearance U, filtering introduces null values for attributes values whose
security classification is higher than the user’s security clearance as shown in (c)
 A user with security clearance C may request for update on the values of job performance of
smith to ‘Excellent’ and the view will allow him to do so. However, the user shouldn't be allowed
to overwrite the existing value at the higher classification level.
 Solution: to create Ploy Station for smith row at the lower classification level C as
shown in (d)

AUWC, School of Technology and Informatics Page 24

 Advanced Database Management

Comparing DAC and MAC

 DAC policies are characterized by a high degree of flexibility, which makes them suitable for a large
variety of application domains.
 The main drawback of DAC models is their weakness to malicious attacks, such as Trojan horses
embedded in application programs.
 By contrast, mandatory policies ensure a high degree of protection in a way; they prevent any illegal
flow of information.
 Mandatory policies have the drawback of being too rigid and they are only applicable in limited
environments.
 In many practical situations, discretionary policies are preferred because they offer a better trade-off
between security and applicability.
3. Role-Based Access Control
 Its basic notion is that permissions are associated with roles, and users are assigned to appropriate
roles.
 Roles can be created using the CREATE ROLE and DESTROY ROLE commands.
 The GRANT and REVOKE commands discussed under DAC can then be used to assign
and revoke privileges from roles.
 RBAC appears to be a feasible alternative to discretionary and mandatory access controls;
 It ensures that only authorized users are given access to certain data or resources.
 Many DBMSs have allowed the concept of roles, where privileges can be assigned to roles.
 Role hierarchy in RBAC is a natural way of organizing roles to reflect the organization’s lines of
authority and responsibility:\My DB File\Role.ppt
Introduction to Statistical Database Security
 Statistical databases are used mainly to produce statistics on various populations.
 The database may contain confidential data on individuals, which should be protected from user
access.
 Users are permitted to retrieve statistical information on the populations, such as averages,
sums, counts, maximums, minimums, and standard deviations.
 A population is a set of rows of a relation (table) that satisfy some selection condition.
 Statistical queries involve applying statistical functions to a population of rows.
 For example, we may want to retrieve the number of individuals in a population or the average
income in the population.
 However, statistical users are not allowed to retrieve individual data, such as the income
of a specific person.
 Statistical database security techniques must disallow the retrieval of individual data.
This can be achieved by elimination of queries that retrieve attribute values and by allowing only
queries that involve statistical aggregate functions such as, SUM, MIN, MAX, and Such queries are
sometimes called statistical queries.
 It is DBMS’s responsibility to ensure confidentiality of information about individuals, while still
providing useful statistical summaries of data about those individuals to users.
 Provision of privacy protection of users in a statistical database is paramount.
 In some cases, it is possible to infer the values of individual rows from sequence statistical queries.
 This is particularly true when the conditions result in a population consisting of a small
number of rows.

AUWC, School of Technology and Informatics Page 25

 Advanced Database Management
Encryption
 Authorization may not be sufficient to protect data in database systems, especially when there
is a situation where data should be moved from one location to the other using network facilities.
 Encryption is used to protect information stored at a particular site or transmitted between sites
from being accessed by unauthorized users.
 Encryption is the encoding of the data by a special algorithm that renders the data unreadable
by any program without the decryption key.
 It is not possible for encrypted data to be read unless the reader knows how to decipher/decrypt
the encrypted data.
 If a database system holds particularly sensitive data, it may be believed necessary to encode it
as an insurance against possible external threats or attempts to access it.
 The DBMS can access data after decoding it, although there is degradation in performance
because of the time taken to decode it.
 Encryption also protects data transmitted over communication lines.
 To transmit data securely over insecure networks requires the use of a Cryptosystem, which
includes:
1. An encryption key to encrypt the data (plaintext)
2. An encryption algorithm that, with the encryption key, transforms the plaintext into
cipher text
3. A decryption key to decrypt the cipher text
4. A decryption algorithm that, with the decryption key, transforms the cipher text back
into plaintext
 Data encryption standard is an approach which does both a substitution of characters and a
rearrangement of their order based on an encryption key.
Types of Cryptosystems
 Cryptosystems can be categorized into two:
1. Symmetric encryption – uses the same key for both encryption and decryption and relies on
safe communication lines for exchanging the key.
2. Asymmetric encryption – uses different keys for encryption and decryption.
 Generally,
 Symmetric algorithms are much faster to execute on a computer than those that are
asymmetric.
 Asymmetric algorithms are more secure than symmetric algorithms.
Public Key Encryption algorithm: Asymmetric encryption
 This algorithm operates with modular arithmetic – mod n, where n is the product of two large prime
numbers.
 Two keys, d and e, are used for decryption and encryption.
 n is chosen as a large integer that is a product of two large distinct prime numbers, p and q.
 The encryption key e is a randomly chosen number between 1 and n that is relatively prime
to (p-1) x (q-1).
 The plaintext m is encrypted as C= me mod n.
 However, the decryption key d is carefully chosen so that C d mod n = m.
 The decryption key d can be computed from the condition that d x e -1 is divisible by (p-1)x(q-1).
 Thus, the legitimate receiver who knows d simply computes Cd mod n = m and recovers m.

AUWC, School of Technology and Informatics Page 26

 Advanced Database Management
Simple Example: Asymmetric encryption
1. Select primes p=11, q=3.
2. n = pq = 11*3 = 33
3. Find phi winch Is Gien by, phi = (p-1)(q-1) = 10*2 = 20
4. Choose e=3 ( 1<e<phi)
5. Check for gcd(e, phi) = gcd(e, (p-1)(q-1)) = gcd(3, 20) = 1
6. Compute d (1<d<phi) such that d *e -1 is divisible by phi
Simple testing (d = 2, 3 ...) gives d = 7
7. Check: ed-1 = 3*7 - 1 = 20, which is divisible by phi (20).
Given
Public key = (n, e) = (33, 3)
Private key = (n, d) = (33, 7)
 Now say we want to encrypt the message m = 7
 c = me mod n = 73 mod 33 = 343 mod 33 = 13
 Hence the ciphertext c = 13
 To check decryption, we compute
 m = cd mod n = 137 mod 33 =62,748,517 mod 33 = 7
Exercise

Select the best answer from the given option

1.

AUWC, School of Technology and Informatics Page 27

 Advanced Database Management

CHAPTER THREE
TRANSACTION PROCESSING CONCEPTS
Introduction to Transaction Processing
 Single-User System:
 At most one user at a time can use the database management system. Eg. Personal computer
system.
 Multi-user System:
 Many users can access the DBMS concurrently. Eg. Airline reservation, Bank and the like
system are operated by many users who submit transaction concurrently to the system.
 This is achieved by multi programming, which allows the computer to execute multiple
programs /processes at the same time.
Concurrency
 Interleaved processing:
 Concurrent execution of processes is interleaved in a single CPU using for example, round robin
algorithm
 Advantages:
 keeps the CPU busy when the process requires I/O by switching to execute another process rather
than remaining idle during I/O time and hence this will increase system throughput (average no.
of transactions completed within a given time)
 Prevents long process from delaying other processes (minimize unpredictable delay in the
response time).
 Parallel processing:
 If Processes are concurrently executed in multiple CPUs.

 A Transaction
 Logical unit of database processing that includes one or more access operations (read -
retrieval, write - insert or update, delete). Examples include ATM transactions, credit
card approvals, flight reservations, hotel check-in, phone calls, supermarket scanning,
academic registration and billing.
 Collections of operations that form a single logical unit of work are called transactions.
 A transaction is a unit of program execution that accesses and possibly updates various
data items. Usually, a transaction is initiated by a user program written in a high-level

AUWC, School of Technology and Informatics Page 28

 Advanced Database Management
data-manipulation language (typically SQL), or programming language (for example, C+
+, or Java), with embedded database accesses in JDBC or ODBC.
 Transaction boundaries
 Any single transaction in an application program is bounded with Begin and End
statements. An application program may contain several transactions separated by the
Begin and End transaction boundaries.
 This collection of steps must appear to the user as a single, indivisible unit. Since a
transaction is indivisible, it either executes in its entirety or not at all. Thus, if a
transaction begins to execute but fails for whatever reason, any changes to the database
that the transaction may have made must be undone. This requirement holds regardless
of whether the transaction itself failed. For example,
 if it divided by zero,  The computer itself stopped operating.
 the operating system crashed, or
ACID properties of the transactions
1. Atomicity. Either all operations of the transaction are reflected properly in the database, or none
are. This “all-or-none” property is referred to as atomicity.
2. Consistency. Execution of a transaction in isolation (that is, with no other transaction executing
concurrently) preserves the consistency of the database.
A transaction must preserve database consistency—if a transaction is run atomically in isolation
starting from a consistent database, the database must again be consistent at the end of the
transaction.
3. Isolation. Even though multiple transactions may execute concurrently, the system guarantees
that, for every pair of transactions Ti and Tj, it appears to Ti that either Tj finished execution
before Ti started or Tj started execution after Ti finished. Thus, each transaction is unaware of
other transactions executing concurrently in the system.
The database system must take special actions to ensure that transactions operate properly
without interference from concurrently executing database statements. This property is referred
to as isolation.
4. Durability. After a transaction completes successfully, the changes it has made to the database
persist, even if there are system failures.
SIMPLE MODEL OF A DATABASE
 A database is a collection of named data items.
 Because SQL is a powerful and complex language, we focus on when data are moved from disk
to main memory and from main memory to disk.
 Granularity of data a field, a record, or a whole disk block that measure the size of the data item
 Basic operations that a transaction can perform are read and write. Transactions access data
using two operations.
 read item(X): Reads a database item named X into a program variable. To simplify our
notation, we assume that the program variable is also named X.
 write item(X): Writes the value of program variable X into the database item named X.
 Basic unit of data transfer from the disk to the computer main memory is one block.
 read item(X) command includes the following steps:
 Find the address of the disk block that contains item X.

AUWC, School of Technology and Informatics Page 29

 Advanced Database Management
 Copy that disk block into a buffer in main memory (if that disk block is not
already in some main memory buffer).
 Copy item X from the buffer to the program variable named X.
 write item(X) command includes the following steps:
 Find the address of the disk block that contains item X.
 Copy that disk block into a buffer in main memory (if that disk block is not already in
some main memory buffer).
 Copy item X from the program variable named X into its correct location in the buffer.
 Store the updated block from the buffer back to disk ( either immediately or later).
 The DBMS maintains a number of buffers in the main memory that holds database disk blocks
which contains the database items being processed.
 When this buffer is occupied and
 if there is a need for additional database block to be copied to the main memory.
 Some buffer management policy is used to choose for replacement but if the chosen buffer has
been modified, it must be written back to disk before it is used.
Transaction Atomicity and Durability
 A transaction may not always complete its execution successfully. Such a transaction is termed
aborted.
 If we are to ensure the atomicity property, an aborted transaction must have no effect on the
state of the database. Thus, any changes that the aborted transaction made to the database must
be undone.
 Once the changes caused by an aborted transaction have been undone, we say that the transaction
has been rolled back. It is part of the responsibility of the recovery scheme to manage
transaction aborts. This is done typically by maintaining a log.
 Each database modification made by a transaction is first recorded in the log. We record the
identifier of the transaction performing the modification, the identifier of the data item being
modified, and both the old value (prior to modification) and the new value (after modification)
of the data item. Only then is the database itself modified.
 Maintaining a log provides the possibility of redoing a modification to ensure atomicity and
durability as well as the possibility of undoing a modification to ensure atomicity in case of a
failure during transaction execution.
 A transaction that completes its execution successfully is said to be committed. A committed
transaction that has performed updates transforms the database into a new consistent state, which
must persist even if there is a system failure.
 Once a transaction has committed, we cannot undo its effects by aborting it.
 The only way to undo the effects of a committed transaction is to execute a compensating
transaction (pay costs). For instance, if a transaction added $20 to an account, the compensating
transaction would subtract $20 from the account.
Transaction States
A transaction must be in one of the following states:
 Active, the initial state; the transaction stays in this state while it is executing.
 Partially committed, after the final statement has been executed.
 Failed, after the discovery that normal execution can no longer proceed.
 Aborted, after the transaction has been rolled back and the database has been restored to its state
prior to the start of the transaction.

AUWC, School of Technology and Informatics Page 30

 Advanced Database Management
 Committed, after successful completion.
 A transaction has committed only if it has entered the committed state.
 A transaction has aborted only if it has entered the aborted state.
 A transaction is said to have terminated if it has either committed or aborted.

Figure 1: - State diagram of a transaction

 A transaction starts in the active state. When it finishes its final statement, it enters the partially
committed state. At this point, the transaction has completed its execution, but it is still possible
that it may have to be aborted, since the actual output may still be temporarily residing in main
memory, and thus a hardware failure may preclude its successful completion.
 The database system then writes out enough information to disk that, even in the event of a failure,
the updates performed by the transaction can be re-created when the system restarts after the failure.
When the last of this information is written out, the transaction enters the committed state.
A transaction enters the failed state after the system determines that the transaction can no longer
proceed with its normal execution (for example, because of hardware or logical errors). Such a
transaction must be rolled back. Then, it enters the aborted state. At this point, the system has two
options:
1. It can restart the transaction, but only if the transaction was aborted as a result of some
hardware or software error that was not created through the internal logic of the transaction. A
restarted transaction is considered to be a new transaction.
2. It can kill the transaction. It usually does so because of some internal logical error that can be
corrected only by rewriting the application program, or because the input was bad, or because
the desired data were not found in the database.
Transaction Isolation
 Transaction-processing systems usually allow multiple transactions to run concurrently.
Allowing multiple transactions to update data concurrently causes several complications with
consistency of the data. There are two good reasons for allowing concurrency:
1. Improved throughput and resource utilization. To increases the throughput of the system—
that is, the number of transactions executed in a given amount of time. Correspondingly, the
processor (CPU) and disk utilization also increase; in other words, the processor and disk
spend less time idle, or not performing any useful work.
2. Reduced waiting time
 If transactions run serially, a short transaction may have to wait for a preceding long
transaction to complete, which can lead to unpredictable delays in running a
transaction.
 If the transactions are operating on different parts of the database, it is better to let them
run concurrently, sharing the CPU cycles and disk accesses among them.

AUWC, School of Technology and Informatics Page 31

 Advanced Database Management
 Concurrent execution reduces the unpredictable delays in running transactions.
Moreover, it also reduces the average response time: the average time for a transaction
to be completed after it has been submitted.
 The database system must control the interaction among the concurrent transactions to prevent
them from destroying the consistency of the database. It does so through a variety of mechanisms
called concurrency-control schemes.
Example: - Let T1 and T2 be two transactions that transfer funds from one account to another.
A. Transaction T1 transfers 50 birr from B. Transaction T2 transfers 10 percent of
account A to account B. It is defined as: the balance from account A to account
B. It is defined as:
T1:read(A);
A:= A − 50; T2:read(A);
write(A); Temp: = A *0.1;
read(B); A: = A − temp;
B:= B + 50; write(A);
write(B) read(B);
B:= B + temp;
write(B)
 Suppose the current values of accounts A and B are 1000 birr and 2000 birr, respectively.
Suppose two transactions are executed one at a time in the order T1 followed by T2.
 This execution sequence appears in Figure 2. In the figure, the sequence of instruction steps is in
chronological order from top to bottom. The final values of accounts A and B, after the
execution in Figure 2 takes place, are 855 birr and 2145 birr, respectively. Thus, the total
amount of money in accounts A and B—that is, the sum A + B—is preserved after the execution
of both transactions. T1
8 read(A)
T2
1 read(A)
2 A: = A − 50 9 temp: = A ∗ 0.1
3 write(A) 10 A: = A − temp
4 read(B) 11 write(A)
5 B: = B + 50 12 read(B)
6 write(B) 13 B: = B + temp
7 commit 14 write(B)
15 commit

Figure 2 Schedule 1a serial schedule in which T1 is followed by T2.

 Schedule 1—a serial schedule in which T1 is followed by T2. Similarly, if the transactions are
executed one at a time in the order T2 followed by T1, then the corresponding execution
sequence is that of Figure 3. Again, as expected, the sum A + B is preserved, and the final values
of accounts A and B are 850 birr and 2150 birr, respectively.
9 read(A) 1 read(A)
T1 T2
10 A: = A − 50 2 temp: = A ∗ 0.1
read(A)
11 write(A) 3 A: = A − temp
A:= A − 50 4 write(A)
12 read(B)
write(A)
13 B: = B + 50 5 read(B)
read(B)
B:=
14Bwrite(B)
+ 50 6 B: = B + temp
write(B)
15 commit 7 write(B)
8 commit

AUWC, School of Technology and Informatics Page 32

 Advanced Database Management
Figure 3 Schedule 2 a serial schedule in which T2 is followed by T1.
 The execution sequences just described are called schedules. They represent the chronological
order in which instructions are executed in the system. These schedules are serial.
 Each serial schedule consists of a sequence of instructions from various transactions, where the
instructions belonging to one single transaction appear together in that schedule.
 When the database system executes several transactions concurrently, the corresponding schedule
no longer needs to be serial. If two transactions are running concurrently, the operating system
may execute one transaction for a little while, then perform a context switch, execute the second
transaction for some time, and then switch back to the first transaction for some time, and so on.
With multiple transactions, the CPU time is shared among all the transactions.
 Not all concurrent executions result in a correct state. Consider the schedule of Figure 5. After
the execution of this schedule, we arrive at a state where the final values of accounts A and B are
950 birr and 2100 birr, respectively. This final state is an inconsistent state, since we have
gained 50 birr in the process of the concurrent execution. Indeed, the sum A + B is not preserved
by the execution of the two transactions.
 It is the job of the database system to ensure that any schedule that is executed will leave the
database in a consistent state. The concurrency-control component of the database system
carries out this task. 4 read(A)
1 read(A)
5 temp: = A ∗ 0.1
T1 2 A: = A −50 T2 6 A: = A − temp
3 Write(A)
7 write(A)
8 read(B)
12 read(B)
9 B: = B + 50
13 B: = B + temp
10 write(B)
14 write(B)
11 commit
15 commit

Figure 4 Schedule 3 a concurrent schedule equivalent to schedule 1.

 We can ensure consistency of the database under concurrent execution by making sure that any
schedule that is executed has the same effect as a schedule that could have occurred without any
concurrent execution. That is, the schedule should, in some sense, be equivalent to a serial
schedule. Such schedules are called serializable schedules.
3 read(A)
T1 1 read(A) T2
4 temp: = A ∗ 0.1
2 A: = A −50 5 A: = A − temp
8 Write(A) 6 write(A)
9 read(B) 7 read(B)
10 B: = B + 50 13 B: = B + temp
11 write(B) 14 write(B)
12 commit 15 commit

Figure 5 Schedule 4—a concurrent schedule resulting in an inconsistent state.

Serializability
 Serial schedules are serializable, but if steps of multiple transactions are interleaved, it is harder
to determine whether a schedule is serializable. Since transactions are programs, it is difficult to
determine exactly
 What operations a transaction performs and
 How operations of various transactions interact.

AUWC, School of Technology and Informatics Page 33

 Advanced Database Management
 For this reason, we shall consider only on two types of operations that a transaction can perform
on a data item: read and write.
 We assume that, between a read (Q) instruction and a write (Q) instruction on a data item Q, a
transaction may perform an arbitrary sequence of operations on the copy of Q that is residing in
the local buffer of the transaction.

 Let us consider a schedule S in which there are two consecutive instructions, I and J, of
transactions Ti and Tj, respectively (i = j).
 If I and J refer to different data items, then we can swap (exchange) I and J without affecting
the results of any instruction in the schedule.
 If I and J refer to the same data item Q, then the order of the two steps may matter.
1. I = read (Q), J = read (Q). The order of I and J does not matter.
2. I = read (Q), J = write (Q). If I come before J, then Ti does not read the value of Q
that is written by Tj in instruction J. If J comes before I, then Ti reads the value of
Q that is written by Tj. Thus, the order of I and J matters.
T1 1 read(A) T2 3 read(A)
2 write(A) 4 write(A)
5 read(B) 7 read(B)
6 write(B) 8 write(B)

Figure 6 Schedule 3 showing only the read and write instructions.

T1 T2
1 read(A) 3 read(A)
2 write(A)) 5 write(A)
4 read(B) 7 read(B)
6 write(B) 8 write(B)

Figure 7 Schedule 5 schedule 3 after swapping of a pair of instructions.

3. I = write (Q), J = read (Q). The order of I and J matters for reasons similar to those
of the previous case.
4. I = write (Q), J = write (Q). Since both instructions are write operations, the order of
these instructions does not affect either Ti or Tj.
However, the value obtained by the next read (Q) instruction of S is affected, since
the result of only the latter of the two write instructions is preserved in the database.
If there is no other write (Q) instruction after I and J in S, then the order of I and J
directly affects the final value of Q in the database state that results from schedule S.
We say that I and J conflict if they are operations by different transactions on the same data item, and at
least one of these instructions is a write operation.
To illustrate the concept of conflicting instructions, we consider schedule 3. The write (A) instruction of
T1 conflicts with the read (A) instruction of T2. However, the write (A) instruction of T2 does not
conflict with the read (B) instruction of T1, because the two instructions access different data items.

T1 1 read(A) T2 5 read(A)
2 write(A) 6 write(A)
3 read(B) 7 read(B)
4 write(B) 8 write(B)

Figure 8 Schedule 6 a serial schedule that is equivalent to schedule 3.

1 read(Q)
3 write(Q)
AUWC, School of Technology and Informatics Page 34
 Advanced Database Management
T3 T4
2 write(Q)

Figure 9 Schedules 7.
 Let I and J be consecutive instructions of a schedule S. If I and J are instructions of different
transactions and I and J do not conflict, then we can swap the order of I and J to produce a new
schedule S.
 S is equivalent to S, since all instructions appear in the same order in both schedules except for I
and J, whose order does not matter.
 Since the write (A) instruction of T2 in schedule 3 does not conflict with the read (B)
instruction of T1, we can swap these instructions to generate an equivalent schedule, schedule
5, in Figure 7. Regardless of the initial system state, schedules 3 and 5 both produce the same
final system state.
 We continue to swap non conflicting instructions:
 Swap the read (B) instruction of T1 with the read (A) instruction of T2.
 Swap the write (B) instruction of T1 with the write (A) instruction of T2.
 Swap the write (B) instruction of T1 with the read (A) instruction of T2.
 The final result of these swaps, schedule 6 of Figure 8, is a serial schedule.
Transaction Isolation and Atomicity
 Effect of transaction failures during concurrent execution.
T1 T5
1 read(A) 3 read(A)
2 write(A) 4 commit
5 read(B)

Example Schedule 9, a non-recoverable schedule.

 If a transaction Ti fails, for whatever reason, we need to undo the effect of this transaction to
ensure the atomicity property of the transaction.
 In a system that allows concurrent execution, the atomicity property requires that any
transaction Tj that is dependent on Ti (that is, Tj has read data written by Ti) is also
aborted.
 To achieve this, we need to place restrictions on the type of schedules permitted in the
system. In the following two subsections, we address the issue of what schedules are
acceptable from the viewpoint of recovery from transaction failure.
Recoverable Schedules
 A recoverable schedule is one where, for each pair of transactions Ti and Tj such that Tj reads a
data item previously written by Ti, the commit operation of Ti appears before the commit
operation of Tj.
Cascade less Schedules
 Even if a schedule is recoverable, to recover correctly from the failure of a transaction Ti, we
may have to roll back several transactions. Such situations occur if transactions have read data
written by Ti.
 Example transaction T8 writes a value of A that is read by transaction T9. Transaction T9 writes
a value of A that is read by transaction T10.
 Suppose that, at this point, T8 fails. T8 must be rolled back. Since T9 is dependent on
T8, T9 must be rolled back. Since T10 is dependent on T9, T10 must be rolled back.

AUWC, School of Technology and Informatics Page 35

 Advanced Database Management
 This phenomenon, in which a single transaction failure leads to a series of transaction rollbacks,
is called cascading rollback. 1 read(A) T9 4 read(A) T10
2 read(B) 6 read(A)
5 write(A)
T8 3 write(A) T10
7 abort

Example Schedules 10.

 Cascading rollback is undesirable, since it leads to the undoing of a significant amount of work.
 It is desirable to restrict the schedules to those where cascading rollbacks cannot occur. Such
schedules are called cascade less schedules.
 Formally, a cascade less schedule is one where, for each pair of transactions Ti and Tj
such that Tj reads a data item previously written by Ti, the commit operation of Ti
appears before the read operation of Tj. It is easy to verify that every cascade less
schedule is also recoverable.
Transaction Isolation Levels
 Serializability is allows programmers to ignore issues related to concurrency when they code
transactions.
 The SQL standard also allows a transaction to specify that it may be executed in such a way that
it becomes non-serializable with respect to other transactions. For instance, a transaction may
operate at the isolation level of read uncommitted, which permits the transaction to read a data
item even if it was written by a transaction that has not been committed.
The isolation levels specified by the SQL standard are as follows:
 Serializable usually ensures serializable execution.
 Repeatable read allows only committed data to be read and further requires that, between two
reads of a data item by a transaction, no other transaction is allowed to update it
 Read committed allows only committed data to be read, but does not require repeatable reads. For
instance, between two reads of a data item by the transaction, another transaction may have
updated the data item and committed.
 Read uncommitted allows uncommitted data to be read. It is the lowest isolation level allowed
by SQL.
All the isolation levels above additionally disallow dirty writes, that is, they disallow writes to a data item
that has already been written by another transaction that has not yet committed or aborted.
Locking
 Instead of locking the entire database, a transaction could, lock only those data items that it
accesses. Under such a policy, the transaction must hold locks long enough to ensure
Serializability, but for a period short enough not to harm performance excessively.
 There are two kinds of locks: shared and exclusive.
1. Shared locks are used for data that the transaction reads and
2. Exclusive locks are used for those it writes.
 Many transactions can hold shared locks on the same data item at the same time, but a
transaction is allowed an exclusive lock on a data item only if no other transaction holds any lock
(regardless of whether shared or exclusive) on the data item. This use of two modes of locks
along with two-phase locking allows concurrent reading of data while still ensuring
Serializability.

AUWC, School of Technology and Informatics Page 36

 Advanced Database Management

Transactions as SQL Statements

 In SQL, insert statements create new data and delete statements delete data.
 These two statements are, write operations, since they change the database, but their interactions
with the actions of other transactions are different.
 Example, consider the following SQL query on our university database that finds all
instructors who earn more than 90,000 birrs.
Select ID, name from instructor where salary > 90000;
 Using our sample instructor relation, we find that only Ein-stein and Brandt satisfy the
condition. Now assume that around the same time we are running our query, another user
inserts a new instructor named “James” whose salary is $100,000.
Insert into instructor values (’11111’, ’James’, ’Marketing’, 100000);
 The result of our query will be different depending on whether this insert comes before or after
our query is run. In a concurrent execution of these transactions, it is intuitively clear that they
conflict, but this is a conflict not captured by our simple model. This situation is referred to as the
phantom phenomenon, because a conflict may exist on “phantom” data.
 But in an SQL statement, the specific data items (tuples) referenced may be determined by a
where clause predicate. So, the same transaction, if run more than once, might reference
different data items each time it is run if the values in the database change between runs.
 One way of dealing with the above problem is to recognize that it is not sufficient for concurrency
control to consider only the tuples that are accessed by a transaction; the information used to find
the tuples that are accessed by the transaction must also be considered for the purpose of
concurrency control.
 The information used to find tuples could be updated by an insertion or deletion, or in the case
of an index, even by an update to a search-key attribute.
 For example, if locking is used for concurrency control, the data structures that track the tuples in
a relation, as well as index structures, must be appropriately locked. However, such locking can
lead to poor concurrency in some situations; index-locking protocols which maximize
concurrency, while ensuring Serializability in spite of inserts, deletes, and predicates in
queries.
 Let us consider again the query:
Select ID, name from instructor where salary> 90000;
 And the following SQL update:
Update instructor set salary = salary *0.9 where name = ’Wu’;
 If our query reads the entire instructor relation, then it reads the tuple with Wu’s data and
conflicts with the update. However, if an index were available that allowed our query direct
access to those tuples with salary > 90000, then our query would not have accessed Wu’s data at
all because Wu’s salary is initially 90,000 birr in our example instructor relation, and reduces to
81,000 birr after the update.
 However, using the above approach, it would appear that the existence of a conflict depends on a
low-level query processing decision by the system that is unrelated to a user-level view of the
meaning of the two SQL statements! An alternative approach to concurrency control treats an
insert, delete or update as conflicting with a predicate on a relation, if it could affect the set of
tuples selected by a predicate.

AUWC, School of Technology and Informatics Page 37

 Advanced Database Management
 In our example query above, the predicate is “salary > 90000”, and an update of Wu’s salary
from 90,000 birr to a value greater than 90,000 birr, or an update of Einstein’s salary from a
value greater than 90,000 birr to a value less than or equal to 90,000 birr, would conflict with
this predicate. Locking based on this idea is called predicate locking; however predicate locking
is expensive, and not used in practice.
Summary
 A transaction is a unit of program execution that accesses and updates various data items.
 Transactions are required to have the ACID properties: atomicity, consistency, isolation, and
durability.
 Atomicity ensures that either all the effects of a transaction are reflected in the
database, or none are; a failure cannot leave the database in a state where a
transaction is partially executed.
 Consistency ensures that, if the database is initially consistent, the execution of the
transaction (by itself) leaves the database in a consistent state.
 Isolation ensures that concurrently executing transactions are isolated from one
another, so that each has the impression that no other transaction is executing
concurrently with it.
 Durability ensures that, once a transaction has been committed, that transaction’s
updates do not get lost, even if there is a system failure.
 Concurrent execution of transactions improves throughput of transactions and system utilization,
and also reduces waiting time of transactions.
 The various types of storage in a computer are volatile storage, nonvolatile storage, and stable
storage. Data in volatile storage, such as in RAM, are lost when the computer crashes. Data in
nonvolatile storage, such as disk, are not lost when the computer crashes, but may occasionally be
lost because of failures such as disk crashes. Data in stable storage are never lost.
 Stable storage that must be accessible online is approximated with mirrored disks, or other forms
of RAID, which provide redundant data storage. Offline or archival, stable storage may consist of
multiple tape copies of data stored in physically secure locations.
 When several transactions execute concurrently on the database, the consistency of data may no
longer be preserved. It is therefore necessary for the system to control the interaction among the
concurrent transactions.
 Since a transaction is a unit that preserves consistency, a serial execution of transactions
guarantees that consistency is preserved.
 A schedule captures the key actions of transactions that affect concurrent execution, such as
read and write operations, while abstracting away internal details of the execution of the
transaction.
 We require that any schedule produced by concurrent processing of a set of transactions will
have an effect equivalent to a schedule produced when these transactions are run serially in
some order.
 A system that guarantees this property is said to ensure serializability.

AUWC, School of Technology and Informatics Page 38

 Advanced Database Management

 There are several different notions of equivalence leading to the concepts of conflict
serializability and view serializability.
 Serializability of schedules generated by concurrently executing transactions can be ensured
through one of a variety of mechanisms called concurrency-control policies.
 We can test a given schedule for conflict serializability by constructing a precedence graph for the
schedule, and by searching for absence of cycles in the graph. However, there are more efficient
concurrency-control policies for ensuring serializability.
 Schedules must be recoverable, to make sure that if transaction a sees the effects of transaction b,
and b then aborts, then an also gets aborted.
 Schedules should preferably be cascade less, so that the abort of a transaction does not result in
cascading aborts of other transactions. Cascadelesness s is ensured by allowing transactions to
only read committed data.
 The concurrency-control–management component of the database is responsible for handling the
concurrency-control policies.

AUWC, School of Technology and Informatics Page 39

 Advanced Database Management

CHAPTER FOUR
CONCURRENCY CONTROL TECHNIQUES
Introduction to Concurrency control techniques
 When several transactions execute concurrently in the database, however, the isolation property
may no longer be preserved. To ensure that it is, the system must control the interaction among
the concurrent transactions; this control is achieved by the mechanisms called concurrency-
control schemes.
 There are a variety of concurrency-control schemes. The most frequently used schemes are two-
phase locking and snapshot isolation.
Lock-Based Protocols
 One way to ensure isolation is to require that data items be accessed in a mutually exclusive
manner; that is, while one transaction is accessing a data item, no other transaction can
modify that data item.
 The most common method used to implement this requirement is to allow a transaction to access
a data item only if it is currently holding a lock on that item.

Locks

 There are various modes in which a data item may be locked. Two modes of locks mostly focused
are:
1. Shared. If a transaction Ti has obtained a shared-mode lock (denoted by S) on item Q, then
Ti can read, but cannot write, Q.
2. Exclusive. If a transaction Ti has obtained an exclusive-mode lock (denoted by X) on item
Q, then Ti can both read and write Q.
 Every transaction request a lock in an appropriate mode on data item Q, depending on the types
of operations that it will perform on Q. The transaction makes the request to the concurrency-
control manager.
 The transaction can proceed with the operation only after the concurrency-control manager
grants the lock to the transaction. The use of these two lock modes allows multiple transactions
to read a data item but limits write access to just one transaction at a time.
 Examples: -Let A and B represent arbitrary lock modes. Suppose that a transaction Ti requests a
lock of mode A on item Q on which transaction Tj (Ti =Tj) currently holds a lock of mode B. If
transaction Ti can be granted a lock on Q immediately, in spite of the presence of the mode B
lock, then we say mode A is compatible with mode B. Such a function can be represented
conveniently by a matrix.

Granting of Locks
 When a transaction requests a lock on a data item in a particular mode, and no other transaction
has a lock on the same data item in a conflicting mode, the lock can be granted. However, care
must be taken to avoid the following scenario.
 Suppose a transaction T2 has a shared-mode lock on a data item, and transaction T1
requests an exclusive-mode lock on the data item. Clearly, T1 has to wait for T2 to release the
shared-mode lock.

AUWC, School of Technology and Informatics Page 40

 Advanced Database Management
 Meanwhile, a transaction T3 may request a shared-mode lock on the same data item. The
lock request is compatible with the lock granted to T2, so T3 may be granted the shared-mode
lock.
 At this point T2 may release the lock, but still T1 has to wait for T3 to finish. But again, there
may be a new transaction T4 that requests a shared-mode lock on the same data item, and is
granted the lock before T3 releases it.
 In fact, it is possible that there is a sequence of transactions that each requests a shared-mode
lock on the data item, and each transaction releases the lock a short while after it is granted,
but T1 never gets the exclusive-mode lock on the data item. The transaction T1 may never
make progress, and is said to be starved.
 We can avoid starvation of transactions by granting locks in the following manner: When a
transaction Ti requests a lock on a data item Q in a particular mode M, the concurrency-control
manager grants the lock provided that:
1. There is no other transaction holding a lock on Q in a mode that conflicts with M.
2. There is no other transaction that is waiting for a lock on Q and that made its lock request
before Ti. Thus, a lock request will never get blocked by a lock request that is made later.

The Two-Phase Locking Protocol

 One protocol that ensures serializability is the two-phase locking protocol. 2-phase locking
protocol is one in which there are 2 phases that a transaction goes through. The first is the
growing phase in which it is acquiring locks, the second is one in which it is releasing locks.
This protocol requires that each transaction issue lock and unlock requests in two phases:
1. Growing phase. A transaction may obtain locks, but may not release any lock.
2. Shrinking phase. A transaction may release locks, but may not obtain any new locks.
 Initially, a transaction is in the growing phase. Once the transaction releases a lock, it enters the
shrinking phase, and it can issue no more lock requests.
 For example, transactions T3 and T4 are two phases. On the other hand, transactions T1 and
T2 are not two phase. Note that the unlock instructions do not need to appear at the end of the
transaction.
 For example, in the case of transaction T3, we could move the unlock (B) instruction to just
after the lock-X (A) instruction, and still retain the two-phase locking property.
 We can show that the two-phase locking protocol ensures conflict serializability. Consider any
transaction. The point in the schedule where the transaction has obtained its final lock (the end of
its growing phase) is called the lock point of the transaction.
 Now, transactions can be ordered according to their lock points— this ordering is, in fact, a
serializability ordering for the transactions.
 Two-phase locking does not ensure freedom from deadlock. Observe that transactions T3 and
T4 are two phase, but, in schedule 2 they are deadlocked.
 Cascading rollback may occur under two-phase locking.
 Cascading rollbacks can be avoided by a modification of two-phase locking called the strict
two-phase locking protocol.
 This protocol requires not only that locking be two phase, but also that all exclusive-mode
locks taken by a transaction be held until that transaction commits.

AUWC, School of Technology and Informatics Page 41

 Advanced Database Management
 This requirement ensures that any data written by an uncommitted transaction are locked
in exclusive mode until the transaction commits, preventing any other transaction from
reading the data.
 Another variant of two-phase locking is the rigorous two-phase locking protocol, which
requires that all locks be held until the transaction commits.
 Deadlock Handling
 A system is in a deadlock state if every transaction in the set is waiting for another transaction in
the set.
 There exists a set of waiting transactions {T0, T1,..., Tn} such that T0 is waiting for a data item
that T1 holds, and T1 is waiting for a data item that T2 holds, and ... ,and Tn−1 is waiting for a
data item that Tn holds, and Tn is waiting for a data item that T0 holds. None of the transactions
can make progress in such a situation.
 To ensure this problem the system rolling back some of the transactions create such problem.
 There are two principal methods for dealing with the deadlock problem.
1. Use a deadlock prevention protocol to ensure that the system will never enter a deadlock state.
2. Allow the system to enter a deadlock state, and then try to recover by using a deadlock detection
and deadlock recovery scheme.
 As we shall see, both methods may result in transaction rollback. Prevention is commonly used if the
probability that the system would enter a deadlock state is relatively high; otherwise, detection and
recovery are more efficient.
 Deadlock Prevention
 There are two approaches to deadlock prevention.
1. Ensures that no cyclic waits can occur by ordering the requests for locks, or requiring all
locks to be acquired together.
2. Performs transaction rollback instead of waiting for a lock, whenever the wait could
potentially result in a deadlock.
 The simplest scheme under the first approach requires that each transaction locks all its data
items before it begins execution. Moreover, either all are locked in one step or none are locked.
There are two main disadvantages to this protocol:
1. It is often hard to predict, before the transaction begins, what data items need to be locked;
2. Data-item utilization may be very low, since many of the data items may be locked but unused for
a long time.
 Another approach for preventing deadlocks is to impose an ordering of all data items, and to
require that a transaction lock data items only in a sequence consistent with the ordering.
 The second approach for preventing deadlocks is to use preemption and transaction rollbacks.
 In preemption, when a transaction Tj requests a lock that transaction Ti holds, the lock
granted to Ti may be preempted by rolling back of Ti, and granting of the lock to Tj.
 To control the preemption, we assign a unique timestamp, based on a counter or on the
system clock, to each transaction when it begins.
 The system uses these timestamps only to decide whether a transaction should wait or
roll back.
 Locking is still used for concurrency control.
 If a transaction is rolled back, it retains its old timestamp when restarted.
 Two different deadlock-prevention schemes using timestamps have been proposed:
1. The wait–die scheme is a non-preemptive technique.

AUWC, School of Technology and Informatics Page 42

 Advanced Database Management
 When transaction Ti requests a data item currently held by Tj, Ti is allowed
to wait only if it has a timestamp smaller than that of Tj (that is, Ti is older
than Tj). Otherwise, Ti is rolled back (dies).
 For example, suppose that transactions T14, T15, and T16 have timestamps
5, 10, and 15, respectively.
i. If T14 requests a data item held by T15, then T14 will wait.
ii. If T24 requests a data item held by T15, then T16 will be rolled back.
2. The wound–wait scheme is a preemptive technique.
 It is a counterpart to the wait–die scheme.
 When transaction Ti requests a data item currently held by Tj, Ti is allowed to wait
only if it has a timestamp larger than that of Tj (that is, Ti is younger than Tj).
Otherwise, Tj is rolled back (Tj is wounded by Ti).
 Returning to our example, with transactions T14, T15, and T16,
i. If T14 requests a data item held by T15, then the data item will be preempted
from T15, and T15 will be rolled back.
ii. If T16 requests a data item held by T15, then T16 will wait.
 The major problem with both of these schemes is that unnecessary rollbacks may occur.
 Another simple approach to deadlock prevention is based on lock timeouts.
 In this approach, a transaction that has requested a lock waits for at most a specified
amount of time.
 If the lock has not been granted within that time, the transaction is said to time out, and it
rolls itself back and restarts.
 If there was in fact a deadlock, one or more transactions involved in the deadlock will
time out and roll back, allowing the others to proceed. This scheme falls somewhere
between deadlock prevention, where a deadlock will never occur, and deadlock
detection and recovery.
 The timeout scheme is particularly easy to implement, and works well if transactions are
short and if long waits are likely to be due to deadlocks. However, in general it is hard to
decide how long a transaction must wait before timing out. Too long a wait results in
unnecessary delays once a deadlock has occurred. Too short a wait results in transaction
rollback even when there is no deadlock, leading to wasted resources.
 Starvation is also a possibility with this scheme. Hence, the timeout-based scheme has
limited applicability.

Deadlock Detection and Recovery

 If a system does not employ some protocol that ensures deadlock freedom, then a detection and
recovery scheme must be used.
 An algorithm that examines the state of the system is invoked periodically to determine whether a
deadlock has occurred.
 If one has, then the system must attempt to recover from the deadlock. To do so, the system
must:
 Maintain information about the current allocation of data items to transactions, as well as
any outstanding data item requests.
 Provide an algorithm that uses this information to determine whether the system has entered
a deadlock state.

AUWC, School of Technology and Informatics Page 43

 Advanced Database Management
 Recover from the deadlock when the detection algorithm determines that a deadlock exists.

Recovery from Deadlock

 When a detection algorithm determines that a deadlock exists, the system must recover from the
deadlock.
 The most common solution is to roll back one or more transactions to break the deadlock. Three
actions need to be taken:
1. Selection of a victim: - Given a set of deadlocked transactions, we must determine which
transaction to roll back to break the deadlock. We should roll back those transactions that will
incur the minimum cost. Many factors may determine the cost of a rollback, including:
A. How long the transaction has computed, and how much longer the transaction will
compute before it completes its designated task.
B. How many data items the transaction has used?
C. How many more data items the transaction needs for it to complete.
D. How many transactions will be involved in the rollback?
2. Rollback: - Once we have decided that a particular transaction must be rolled back, we must
determine how far this transaction should be rolled back. The simplest solution is a total
rollback: Abort the transaction and then restart it.
3. Starvation. In a system where the selection of victims is based primarily on cost factors, it
may happen that the same transaction is always picked as a victim. As a result, this
transaction never completes its designated task, thus there is starvation. We must ensure that
a transaction can be picked as a victim only a (small) finite number of times. The most
common solution is to include the number of rollbacks in the cost factor.

Timestamp-Based Protocols
 The locking protocols that we have described thus far determine the order between every pair of
conflicting transactions at execution time by the first lock that both members of the pair request
that involves incompatible modes.
 Another method for determining the serializability order is to select an ordering among
transactions in advance. The most common method for doing so is to use a timestamp-ordering
scheme.
Timestamps
 With each transaction Ti in the system, we associate a unique fixed timestamp, denoted by
TS(Ti). This timestamp is assigned by the database system before the transaction Ti starts
execution.
 If a transaction Ti has been assigned timestamp TS(Ti), and a new transaction Tj enters the
system, then TS(Ti) < TS(Tj).
 There are two simple methods for implementing this scheme:
1. Use the value of the system clock as the timestamp; that is, a transaction’s timestamp is
equal to the value of the clock when the transaction enters the system.
2. Use a logical counter that is incremented after a new timestamp has been assigned; that is,
a transaction’s timestamp is equal to the value of the counter when the transaction enters
the system.

AUWC, School of Technology and Informatics Page 44

 Advanced Database Management
 The timestamps of the transactions determine the serializability order. Thus, if TS(Ti) < TS(Tj),
then the system must ensure that the produced schedule is equivalent to a serial schedule in which
transaction Ti appears before transaction Tj.
 To implement this scheme, we associate with each data item Q two timestamp values:
1. W-timestamp(Q) denotes the largest timestamp of any transaction that executed write(Q) successfully.
2. R-timestamp(Q) denotes the largest timestamp of any transaction that executed read(Q) successfully.
 These timestamps are updated whenever a new read(Q) or write(Q) instruction is executed.

The Timestamp-Ordering Protocol

 The timestamp-ordering protocol ensures that any conflicting read and write operations are
executed in timestamp order. This protocol operates as follows:

1. Suppose that transaction Ti issues read(Q).

A. If TS(Ti) < W-timestamp(Q), then Ti needs to read a value of Q that was already
overwritten. Hence, the read operation is rejected, and Ti is rolled back.
B. If TS(Ti) ≥ W-timestamp(Q), then the read operation is executed, and R-
timestamp(Q) is set to the maximum of R-timestamp(Q) and TS(Ti).

2. Suppose that transaction Ti issues write(Q).

A. If TS(Ti) < R-timestamp(Q), then the value of Q that Ti is producing was needed
previously, and the system assumed that that value would never be produced. Hence,
the system rejects the write operation and rolls Ti back.
B. If TS(Ti) < W-timestamp(Q), then Ti is attempting to write an obsolete value of Q.
Hence, the system rejects this write operation and rolls Ti back.
C. Otherwise, the system executes the write operation and sets W-timestamp(Q) to
TS(Ti).
 If a transaction Ti is rolled back by the concurrency-control scheme as result of issuance of either
a read or writes operation, the system assigns it a new timestamp and restarts it.
 To illustrate this protocol, we consider transactions T25 and T26. Transaction T25 displays the
contents of accounts A and B:
T25: read(B);
read(A);
display (A + B).
 Transaction T26 transfers $50 from account B to account A, and then displays the contents of
both:
T26: read(B);
B := B − 50;
write(B);
read(A);
A := A + 50;
write(A);
display (A + B).
Validation-Based Protocols
 The validation protocol requires that each transaction Ti executes in two or three different
phases in its lifetime, depending on whether it is a read-only or an update transaction. The
phases are, in order:

AUWC, School of Technology and Informatics Page 45

 Advanced Database Management
1. Read phase. The system executes transaction Ti. It reads the values of the various data items
and stores them in variables local to Ti. It performs all write operations on temporary local
variables, without updates of the actual database.
2. Validation phase. The validation test is applied to transaction Ti. This determines whether Ti
is allowed to proceed to the write phase without causing a violation of serializability. If a
transaction fails the validation test, the system aborts the transaction.
3. Write phase. If the validation test succeeds for transaction Ti, the temporary local variables
that hold the results of any write operations performed by Ti are copied to the database.
Read-only transactions omit this phase.
 To perform the validation test, we need to know when the various phases of transactions took
place. We shall, therefore, associate three different timestamps with each transaction Ti:
1. Start (Ti), the time when Ti started its execution.
2. Validation (Ti), the time when Ti finished its read phase and started its validation phase.
3. Finish (Ti), the time when Ti finished its write phase.
 The validation test for transaction Ti requires that, for all transactions Tk with TS(Tk) < TS(Ti),
one of the following two conditions must hold:
1. Finish (Tk) < Start (Ti). Since Tk completes its execution before Ti started, the serializability
order is indeed maintained.
2. The set of data items written by Tk does not intersect with the set of data items read by Ti,
and Tk completes its write phase before Ti starts its validation phase (Start(Ti) < Finish(Tk)
< Validation(Ti)). This condition ensures that the writes of Tk and Ti do not overlap. Since
the writes of Tk do not affect the read of Ti, and since Ti cannot affect the read of Tk, the
serializability order is indeed maintained.
T26
T25 2. Read(B)
1. Read(B)
3. B: =B-50 read(A)
5. Read(A)
4. A: =A+50
6. <Validate>
8. <Validate>
7. Display(A+B)
9. Write(B)
10. Write(A)

Figure Schedule 6, a schedule produced by using validation.

 As an illustration, consider again transactions T25 and T26. Suppose that TS(T25) < TS(T26).
Then, the validation phase succeeds in the schedule 6. Note that the writes to the actual variables
are performed only after the validation phase of T26. Thus, T25 reads the old values of B and A,
and this schedule is serializable.
 The validation scheme automatically guards against cascading rollbacks, since the actual writes
take place only after the transaction issuing the write has committed. However, there is a
possibility of starvation of long transactions, due to a sequence of conflicting short transactions
that cause repeated restarts of the long transaction.
 To avoid starvation, conflicting transactions must be temporarily blocked, to enable the long
transaction to finish.
 This validation scheme is called the optimistic concurrency-control scheme since transactions
execute optimistically, assuming they will be able to finish execution and validate at the end. In
contrast, locking and timestamp ordering are pessimistic in that they force a wait or a rollback

AUWC, School of Technology and Informatics Page 46

 Advanced Database Management
whenever a conflict is detected, even though there is a chance that the schedule may be conflict
serializable.

Purpose of Concurrency Control

 To enforce Isolation (through mutual exclusion) among conflicting transactions.
 To preserve database consistency through consistency preserving execution of
transactions.
 To resolve read-write and write-write conflicts.

Concurrency Control Techniques

 Locking  Multi-version
 Timestamp  Lock Granularity
 Optimistic
 Locking and timestamp are conservative/ traditional approach

Starvation
 Starvation occurs when a particular transaction consistently waits or restarted and
never gets a chance to proceed further while other transaction continues normally
 This may occur, if the waiting method for item locking:
 Gave priority for some transaction over others
 Problem in Victim selection algorithm- it is possible that the same
transaction may consistently be selected as victim and rolled-back .example
In Wound-Wait
 Solution

 FIFO
 Allow for transaction that wait for a longer time
 Give higher priority for transaction that have been aborted for many time

AUWC, School of Technology and Informatics Page 47

 Advanced Database Management
CHAPTER FIVE
DATABASE RECOVERY TECHNIQUES
Recovery Outline and Categorization of Recovery Algorithms
 Recovery from transaction failures usually means that the database is restored to the most
recent consistent state just before the time of failure.
 To do this, the system must keep information about the changes that were applied to
data items by the various transactions. This information is typically kept in the system
log.
 A typical strategy for recovery may be summarized informally as follows:
1. If there is extensive damage to a wide portion of the database due to catastrophic
failure, such as a disk crash, the recovery method restores a past copy of the database
that was backed up to archival storage and reconstructs a more current state by reapplying
or redoing the operations of committed transactions from the backed up log, up to the
time of failure.
2. When the database on disk is not physically damaged, and a non-catastrophic failure
has occurred, the recovery strategy is to identify any changes that may cause an
inconsistency in the database.
 For non-catastrophic failure, the recovery protocol does not need a complete
archival copy of the database. Rather, the entries kept in the online system log
on disk are analyzed to determine the appropriate actions for recovery.
 Conceptually, we can distinguish two main techniques for recovery from non-catastrophic
transaction failures:
1. Deferred update and
2. Immediate update.

Deferred Update
 The deferred update techniques do not physically update the database on disk until after a
transaction reaches its commit point; then the updates are recorded in the database.
 Before reaching commit, all transaction updates are recorded in the local transaction
workspace or in the main memory buffers that the DBMS maintains.
 Before commit, the updates are recorded persistently in the log, and then after commit,
the updates are written to the database on disk.
 If a transaction fails before reaching its commit point, it will not have changed the database in
any way, so UNDO is not needed.
 It may be necessary to REDO the effect of the operations of a committed transaction from the
log, because their effect may not yet have been recorded in the database on disk. Hence, deferred
update is also known as the NO-UNDO/REDO algorithm.

Immediate Update
 In the immediate update techniques, the database may be updated by some operations of a
transaction before the transaction reaches its commit point.
 However, these operations must also be recorded in the log on disk by force-writing
before they are applied to the database on disk, making recovery still possible.

AUWC School of Technology and InformaticsPage 48

 Advanced Database Management
 If a transaction fails after recording some changes in the database on disk but before reaching its
commit point, the effect of its operations on the database must be undone; that is, the
transaction must be rolled back.
 In the general case of immediate update, both undo and redo may be required during recovery.
This technique, known as the UNDO/REDO algorithm, requires both operations during
recovery, and is used most often in practice.
 A variation of the algorithm where all updates are required to be recorded in the database on
disk before a transaction commits requires undo only, so it is known as the UNDO/NO-REDO
algorithm.
Caching (Buffering) of Disk Blocks
 Typically, multiple disk pages that include the data items to be updated are cached into main
memory buffers and then updated in memory before being written back to disk.
 The caching of disk pages is traditionally an operating system function, but because of
its importance to the efficiency of recovery procedures, it is handled by the DBMS by
calling low-level operating systems routines.
 When the DBMS requests action on some item,
1. First it checks the cache directory to determine whether the disk page containing the
item is in the DBMS cache.
2. Second if it is not, the item must be located on disk, and the appropriate disk pages are
copied into the cache.
 It may be necessary to replace (or flush) some of the cache buffers to make space available for
the new item.
 Some page replacement strategy similar to these used in operating systems, such as least recently
used (LRU) or first-in-first out (FIFO), or a new strategy that is DBMS-specific can be used to
select the buffers for replacement, such as DBMIN or Least-Likely-to-Use.
 Two main strategies can be employed when flushing a modified buffer back to disk.
1. In-place updating, writes the buffer to the same original disk location, thus
overwriting the old value of any changed data items on disk.
2. Shadowing, writes an updated buffer at a different disk location, so multiple
versions of data items can be maintained.
 In general, the old value of the data item before updating is called the before image (BFIM),
and the new value after updating is called the after image (AFIM).
 If shadowing is used, both the BFIM and the AFIM can be kept on disk; hence, it is not strictly
necessary to maintain a log for recovering.

Write-Ahead Logging, Steal/No-Steal, and Force/No-Force

 When in-place updating is used, it is necessary to use a log for recovery.
 In this case, the recovery mechanism must ensure that the BFIM of the data item is recorded in
the appropriate log entry and that the log entry is flushed to disk before the BFIM is
overwritten with the AFIM in the database on disk. This process is generally known as write-
ahead logging.
 There are two types of log entry information included for a write command:
1. The information needed for UNDO: The UNDO-type log entries include the old value
(BFIM) of the item since this is needed to undo the effect of the operation from the log (by
setting the item value in the database back to its BFIM).

AUWC School of Technology and InformaticsPage 49

 Advanced Database Management
2. The information needed for REDO. A REDO-type log entry includes the new value
(AFIM) of the item written by the operation since this is needed to redo the effect of the
operation from the log (by setting the item value in the database on disk to its AFIM).
 Standard DBMS recovery terminology includes the terms steal/no-steal and force/no-force,
which specify the rules that govern when a page from the database can be written to disk from
the cache:
1. If a cache buffer page updated by a transaction cannot be written to disk before the
transaction commits, the recovery method is called a no-steal approach.
 On the other hand, if the recovery protocol allows writing an updated buffer
before the transaction commits, it is called steal approach.
2. If all pages updated by a transaction are immediately written to disk before the
transaction commits, it is called a force approach. Otherwise, it is called no-force.
 The force rule means that REDO will never be needed during recovery, since
any committed transaction will have all its updates on disk before it is
committed.
 The deferred update (NO-UNDO) recovery scheme follows a no-steal approach. However,
typical database systems employ a steal/no-force strategy.
 The advantage of steal is that it avoids the need for a very large buffer space to store all
updated pages in memory.
 The advantage of no-force is that an updated page of a committed transaction may still
be in the buffer when another transaction needs to update it, thus eliminating the I/O
cost to write that page multiple times to disk, and possibly to have to read it again from
disk.
 This may provide a substantial saving in the number of disk I/O operations
when a specific page is updated heavily by multiple transactions.

Transaction Actions That Do Not Affect the Database

 In general, a transaction will have actions that do not affect the database, such as generating
and printing messages or reports from information retrieved from the database.
 If a transaction fails before completion, we may not want the user to get these reports, since the
transaction has failed to complete.
 If such erroneous reports are produced, part of the recovery process would have to
inform the user that these reports are wrong, since the user may take an action based
on these reports that affects the database. Hence, such reports should be generated
only after the transaction reaches its commit point.
 A common method of dealing with such actions is to issue the commands that generate the
reports but keep them as batch jobs, which are executed only after the transaction reaches its
commit point. If the transaction fails, the batch jobs are canceled.

NO-UNDO/REDO Recovery Based on Deferred Update

 The idea behind deferred update is to defer or postpone any actual updates to the database on
disk until the transaction completes its execution successfully and reaches its commit point.
 During transaction execution, the updates are recorded only in the log and in the cache buffers.
 We can state a typical deferred update protocol as follows:
1. A transaction cannot change the database on disk until it reaches its commit point.

AUWC School of Technology and InformaticsPage 50

 Advanced Database Management
2. A transaction does not reach its commit point until all its REDO-type log entries
are recorded in the log and the log buffer is force-written to disk.
 Notice that step 2 of this protocol is a restatement of the write-ahead logging (WAL) protocol.
 Because the database is never updated on disk until after the transaction commits,
there is never a need to UNDO any operations.
 REDO is needed in case the system fails after a transaction commits but before all its
changes are recorded in the database on disk.
 In this case, the transaction operations are redone from the log entries
during recovery.
 For multiuser systems with concurrency control, the concurrency control and recovery processes
are interrelated.
 Consider a system in which concurrency control uses strict two-phase locking, so the
locks on items remain in effect until the transaction reaches its commit point. After that,
the locks can be released. This ensures strict and serializable schedules.
 If a transaction is aborted for any reason (say, by the deadlock detection method), it is simply
resubmitted, since it has not changed the database on disk.
 A drawback of the method described here is that it limits the concurrent execution of
transactions because all write-locked items remain locked until the transaction reaches
its commit point.
 Additionally, it may require excessive buffer space to hold all updated items until the
transactions commit.
 The method’s main benefit is that transaction operations never need to be undone, for two
reasons:
1. A transaction does not record any changes in the database on disk until after it reaches
its commit point—that is, until it completes its execution successfully. Hence, a
transaction is never rolled back because of failure during transaction execution.
2. A transaction will never read the value of an item that is written by an uncommitted
transaction, because items remain locked until a transaction reaches its commit point.
Hence, no cascading rollback will occur.

Recovery Techniques Based on Immediate Update

 In these techniques, when a transaction issues an update command, the database on disk can be
updated immediately, without any need to wait for the transaction to reach its commit point.
 Notice that it is not a requirement that every update be applied immediately to disk; it is
just possible that some updates are applied to disk before the transaction commits.
 Theoretically, we can distinguish two main categories of immediate update algorithms.
1. If the recovery technique ensures that all updates of a transaction are recorded in the
database on disk before the transaction commits, there is never a need to REDO any
operations of committed transactions. This is called the UNDO/NO-REDO recovery
algorithm.
 In this method, all updates by a transaction must be recorded on disk before the
transaction commits, so that REDO is never needed. Hence, this method must utilize
the force strategy for deciding when updated main memory buffers are written back to
disk.

AUWC School of Technology and InformaticsPage 51

 Advanced Database Management
2. If the transaction is allowed to commit before all its changes are written to the database, we
have the most general case, known as the UNDO/REDO recovery algorithm. In this case, the
steal/no-force strategy is applied.

Shadow Paging
 This recovery scheme does not require the use of a log in a single-user environment.
 In a multiuser environment, a log may be needed for the concurrency control method.
 Shadow paging considers the database to be made up of a number of fixed size disk
pages (or disk blocks)—say, n—for recovery purposes.
 A directory with n entries is constructed, where the ith entry points to the ith database page on
disk.
 The directory is kept in main memory if it is not too large, and all references—read or writes—to
database pages on disk go through it.
 When a transaction begins executing, the current directory—whose entries point to the most
recent or current database pages on disk—is copied into a shadow directory.
 The shadow directory is then saved on disk while the current directory is used by the transaction.
 During transaction execution, the shadow directory is never modified. When a write item
operation is performed, a new copy of the modified database page is created, but the old copy of
that page is not overwritten.
 The database thus is returned to its state prior to the transaction that was executing when the crash
occurred, and any modified pages are discarded. Committing a transaction corresponds to
discarding the previous shadow directory. Since recovery involves neither undoing nor redoing
data items, this technique can be categorized as a NO-UNDO/ NO-REDO technique for
recovery.

The ARIES Recovery Algorithm

 It is used in many relational database-related products of IBM.
 ARIES uses a steal/no-force approach for writing, and it is based on three concepts:
1. Write-ahead logging, repeating history during redo, and logging changes during undo.
2. The second concept, repeating history, means that ARIES will retrace all actions of the
database system prior to the crash to reconstruct the database state when the crash occurred.
a. Transactions that were uncommitted at the time of the crash (active transactions)
are undone.
3. The third concept, logging during undo, will prevent ARIES from repeating the completed
undo operations if a failure occurs during recovery, which causes a restart of the recovery
process.
 The ARIES recovery procedure consists of three main steps: analysis, REDO, and UNDO.
1. Analysis: The analysis step identifies the dirty (updated) pages in the buffer and the set
of transactions active at the time of the crash.
2. Redo
 The appropriate point in the log where the REDO operation should start is also
determined.
 The REDO phase actually reapplies updates from the log to the database.
 Generally, the REDO operation is applied only to committed transactions.

AUWC School of Technology and InformaticsPage 52

 Advanced Database Management
 In ARIES, every log record has an associated log sequence number (LSN) that is
monotonically increasing and indicates the address of the log record on disk.
 Each LSN corresponds to a specific change (action) of some transaction.
 Also, each data page will store the LSN of the latest log record corresponding to
a change for that page.
 A log record is written for any of the following actions: updating a page (write), committing a
transaction (commit), aborting a transaction (abort), undoing an update (undo), and ending
a transaction (end).

Recovery in Multidata base Systems

 These databases may even be stored on different types of DBMSs; for example, some DBMSs
may be relational, whereas others are object oriented, hierarchical, or network DBMSs.
 In such a case, each DBMS involved in the multidata base transaction may have its own
recovery technique and transaction manager separate from those of the other DBMSs.
 This situation is somewhat similar to the case of a distributed database management
system, where parts of the database reside at different sites that are connected by a
communication network.
 To maintain the atomicity of a multidata base transaction, it is necessary to have a two-level
recovery mechanism.
 A global recovery manager, or coordinator, is needed to maintain information needed for
recovery, in addition to the local recovery managers and the information they maintain (log,
tables).
 The coordinator usually follows a protocol called the two-phase commit protocol; whose two
phases can be stated as follows:
 Phase 1. When all participating databases signal the coordinator that the part of the
multidata base transaction involving each has concluded, the coordinator sends a message
prepare for commit to each participant to get ready for committing the transaction.
 Each participating database receiving that message will force-write all log records
and needed information for local recovery to disk and then send a ready to commit or
OK signal to the coordinator.
 If the force-writing to disk fails or the local transaction cannot commit for some
reason, the participating database sends cannot commit or not OK signal to the
coordinator.
 If the coordinator does not receive a reply from the database within a certain time
out interval, it assumes a not OK response.
 Phase 2. If all participating databases reply OK, and the coordinator’s vote is also OK, the
transaction is successful, and the coordinator sends a commit signal for the transaction to the
participating databases.
 The net effect of the two-phase commit protocol is that either all participating databases
commit the effect of the transaction or none of them do.
 In case any of the participants—or the coordinator—fails, it is always possible to
recover to a state where either the transaction is committed or it is rolled back.
 A failure during or before Phase 1 usually requires the transaction to be rolled back, whereas
a failure during Phase 2 means that a successful transaction can recover and commit.

AUWC School of Technology and InformaticsPage 53

 Advanced Database Management
Database Backup and Recovery from Catastrophic Failures
 So far, all the techniques we have discussed apply to non-catastrophic failures.
 A key assumption has been that the system log is maintained on the disk and is not lost as a
result of the failure.
 Similarly, the shadow directory must be stored on disk to allow recovery when shadow paging is
used.
 The recovery techniques we have discussed use the entries in the system log or the
shadow directory to recover from failure by bringing the database back to a consistent
state.
 The recovery manager of a DBMS must also be equipped to handle more catastrophic failures
such as disk crashes.
 The main technique used to handle such crashes is a database backup, in which the
whole database and the log are periodically copied onto a cheap storage medium such
as magnetic tapes or other large capacity offline storage devices.
 In case of a catastrophic system failure, the latest backup copy can be reloaded from the tape to
the disk, and the system can be restarted.
 Data from critical applications such as banking, insurance, stock market, and other databases is
periodically backed up in its entirety and moved to physically separate safe locations.
 To avoid losing all the effects of transactions that have been executed since the last
backup, it is customary to back up the system log at more frequent intervals than full
database backup by periodically copying it to magnetic tape.
 The system log is usually substantially smaller than the database itself and hence can be backed
up more frequently. Therefore, users do not lose all transactions they have performed since the
last database backup.
 All committed transactions recorded in the portion of the system log that has been backed up to
tape can have their effect on the database redone.
 A new log is started after each database backup.
 Hence, to recover from disk failure, the database is first recreated on disk from its
latest backup copy on tape.
 Following that, the effects of all the committed transactions whose operations have been recorded
in the backed-up copies of the system log are reconstructed.

AUWC School of Technology and InformaticsPage 54

 Advanced Database Management
CHAPTER SIX
DISTRIBUTED DATABASE SYSTEMS
Distributed Database Concepts
 Distributed databases bring the advantages of distributed computing to the database management
domain.
 A distributed computing system consists of a number of processing elements, not necessarily
homogeneous, that are interconnected by a computer network, and that cooperate in performing
certain assigned tasks.
 As a general goal, distributed computing systems partition a big, unmanageable problem into
smaller pieces and solve it efficiently in a coordinated manner.
 The economic viability of this approach stems from two reasons:
1. More computer power is harnessed(connected) to solve a complex task, and
2. Each autonomous (independent) processing element can be managed
independently and develop its own applications.

Generally
 Distributed database (DDB) as a collection of multiple logically interrelated databases
distributed over a computer network.
 Distributed database management system (DDBMS) as a software system that manages a
distributed database while making the distribution transparent to the user.
 A collection of files stored at different nodes of a network and the maintaining of inter
relationships among them via hyperlinks has become a common organization on the Internet,
with files of Web pages.
 Examples DDBMS Operational database, Analytical database, Hypermedia database

Parallel Versus Distributed Technology

 Turning our attention to system architectures, there are two main types of multiprocessor system
architectures that are commonplace:
1. Shared memory (tightly coupled) architecture: Multiple processors share secondary (disk)
storage and also share primary memory.
2. Shared disk (loosely coupled) architecture: Multiple processors share secondary (disk)
storage but each has their own primary memory.
 These architectures enable processors to communicate without the overhead of exchanging
messages over a network.
 Database management systems developed using the above types of architectures are termed
parallel database management systems rather than DDBMS, since they utilize parallel processor
technology.
 Another type of multiprocessor architecture is called shared nothing architecture. In this
architecture, every processor has its own primary and secondary (disk) memory, no common
memory exists, and the processors communicate over a high-speed interconnection network (bus
or switch).

Advantages of Distributed Databases

AUWC School of Technology and InformaticsPage 55

 Advanced Database Management
 Distributed database management has been proposed for various reasons ranging from
organizational decentralization and economical processing to greater autonomy.
 We highlight some of these advantages here.
1. Management of distributed data with different levels of transparency:
 Ideally, a DBMS should be distribution transparent in the sense of hiding the details
of where each file (table, relation) is physically stored within the system.
 The EMPLOYEE, PROJECT, and WORKS_ON tables may be fragmented
horizontally (that is, into sets of rows and stored with possible replication).
 The following types of transparencies are possible:
 Distribution or network transparency: This refers to freedom for the user from the
operational details of the network. It may be divided into location transparency and
naming transparency.
A. Location transparency refers to the fact that the command used to perform a task is
independent of the location of data and the location of the system where the
command was issued.
B. Naming transparency implies that once a name is specified, the named objects can
be accessed unambiguously without additional specification.
 Replication transparency: copies of data may be stored at multiple sites for better
availability, performance, and reliability. Replication transparency makes the user
unaware (uninformed) of the existence of copies.
 Fragmentation transparency: Two types of fragmentation are possible.
A. Horizontal fragmentation distributes a relation into sets of tuples (rows).
B. Vertical fragmentation distributes a relation into sub relations where each sub
relation is defined by a subset of the columns of the original relation.
A global query by the user must be transformed into several fragment queries.
Fragmentation transparency makes the user unaware of the existence of
fragments.
2. Increased reliability and availability:
 These are two of the most common potential advantages cited for distributed databases.
A. Reliability is broadly defined as the probability that a system is running (not
down) at a certain time point.
B. Availability is the probability that the system is continuously available during a
time interval.
 When the data and DBMS software are distributed over several sites, one site may fail
while other sites continue to operate.
 Only the data and software that exist at the failed site cannot be accessed. This improves
both reliability and availability. Further improvement is achieved by judiciously
replicating data and software at more than one site.
 In a centralized system, failure at a single site makes the whole system unavailable to all
users.
 In a distributed database, some of the data may be unreachable, but users may still be
able to access other parts of the database.

3. Improved performance:

AUWC School of Technology and InformaticsPage 56

 Advanced Database Management
 A distributed DBMS fragments the database by keeping the data closer to where it is
needed most.
 Data localization reduces the contention for CPU and I/O services and
simultaneously reduces access delays involved in wide area networks.
 When a large database is distributed over multiple sites, smaller databases exist at
each site.
 As a result, local queries and transactions accessing data at a single site have
better performance because of the smaller local databases.
 In addition, each site has a smaller number of transactions executing than if all
transactions are submitted to a single centralized database.
 Moreover, inter query and intra query parallelism can be achieved by executing
multiple queries at different sites, or by breaking up a query into a number of sub
queries that execute in parallel. This contributes to improved performance.
4. Easier expansion (scalability):
 In a distributed environment, expansion of the system in terms of adding more data,
increasing database sizes, or adding more processors is much easier.
 The transparencies we discussed in (1) above lead to a compromise between ease of use
and the overhead cost of providing transparency.
 Total transparency provides the global user with a view of the entire DDBS as if it is a
single centralized system.
 Transparency is provided as a complement to autonomy, which gives the users tighter
control over their own local databases.
 Transparency features may be implemented as a part of the user language, which may
translate the required services into appropriate operations. In addition, transparency
impacts the features that must be provided by the operating system and the DBMS.

Disadvantages of Distributed Databases

 Complexity- The data replication, failure recovery, network management …makes the system
more complex than the central DBMSs.
 Cost- since DDBMS needs more people and more hardware, maintaining and running the system
can be more expensive than the centralized system.
 Problem of connecting Dissimilar Machine- Additional layers of operating system software are
needed to translate and coordinate the flow of data between machines.
 Data integrity and security problem- Because data maintained by distributed systems can be
accessed at any locations in the network, controlling the integrity of a database can be difficult .
Additional Functions of Distributed Databases
 Distribution leads to increased complexity in the system design and implementation.
 To achieve the potential advantages listed previously, the DDBMS software must be able to
provide the following functions in addition to those of a centralized DBMS:
 Keeping track of data: The ability to keep track of the data distribution, fragmentation, and
replication by expanding the DDBMS catalog.
 Distributed query processing: The ability to access remote sites and transmit queries and
data among the various sites via a communication network.

AUWC School of Technology and InformaticsPage 57

 Advanced Database Management
 Distributed transaction management: The ability to devise/plan execution strategies for
queries and transactions that access data from more than one site and to synchronize the
access to distributed data and maintain integrity of the overall database.
 Replicated data management: The ability to decide which copy of a replicated data item to
access and to maintain the consistency of copies of a replicated data item.
 Distributed database recovery: The ability to recover from individual site crashes and from
new types of failures such as the failure of a communication links.
 Security: Distributed transactions must be executed with the proper management of the
security of the data and the authorization/access privileges of users.
 Distributed directory (catalog) management: A directory contains information (metadata)
about data in the database. The directory may be global for the entire DDB, or local for each
site. The placement and distribution of the directory are design and policy issues.
 These functions themselves increase the complexity of a DDBMS over a centralized
DBMS.
 Before we can realize the full potential advantages of distribution, we must find satisfactory
solutions to these design issues and problems.
 Including all this additional functionality is hard to accomplish, and finding optimal solutions is a
step beyond that.
 At the physical hardware level, the following main factors distinguish a DDBMS from a
centralized system:
 There are multiple computers, called sites or nodes.
 These sites must be connected by some type of communication network to transmit data
and commands among sites.
 The sites may all be located in physical proximity—say, within the same building or group of
adjacent buildings—and connected via a local area network, or they may be geographically
distributed over large distances and connected via a long-haul or wide area network.
 Local area networks typically use cables, whereas long-haul networks use telephone lines or
satellites. It is also possible to use a combination of the two types of networks.

Data Fragmentation, Replication, and Allocation Techniques for Distributed Database

Design

 Techniques that are used to break up the database into logical units, called fragments, which
may be assigned for storage at the various sites.
 Data replication, which permits certain data to be stored in more than one site, and the process
of allocating fragments—or replicas of fragments—for storage at the various sites.
 These techniques are used during the process of distributed database design.
 The information concerning data fragmentation, allocation, and replication is stored in a global
directory that is accessed by the DDBS applications as needed.

Data Fragmentation
 There are two approaches to store the relation in the distributed database:

Replication and Fragmentation.

AUWC School of Technology and InformaticsPage 58

 Advanced Database Management
 Data Fragmentation: is a technique used to break up the database into logically related units
called fragments. A database can be fragmented as:
1. Horizontal Fragmentation
2. Vertical Fragmentation
3. Mixed (Hybrid) Fragmentation
 The main reasons for fragmenting a relation are
1. Efficiency (good organization)- data that is not needed by the local applications is not
stored
2. Parallelism- a transaction can be divided into several sub queries that operate on fragments
which will increase the degree of concurrency.
 In a DDB, decisions must be made regarding which site should be used to store which portions
of the database.
 Before we decide on how to distribute the data, we must determine the logical units of the
database that are to be distributed.

Horizontal Fragmentation
 A horizontal fragment of a relation is a subset of the tuples in that relation.
 The tuples that belong to the horizontal fragment are specified by a condition on one or more
attributes of the relation. Often, only a single attribute is involved.
 For example, we may define three horizontal fragments on the EMPLOYEE relation: (DNO= 5),
(DNO= 4), and (DNO= 1)—each fragment contains the EMPLOYEE tuples working for a
particular department.
 Similarly, we may define three horizontal fragments for the PROJECT relation, with the
conditions (DNUM= 5), (DNUM= 4), and (DNUM= 1)—each fragment contains the PROJECT
tuples controlled by a particular department.
 Horizontal fragmentation divides a relation "horizontally" by grouping rows to create subsets
of tuples, where each subset has a certain logical meaning. These fragments can then be
assigned to different sites in the distributed system.
 Derived horizontal fragmentation applies the partitioning of a primary relation (DEPARTMENT
in our example) to other secondary relations (EMPLOYEE and PROJECT in our example), which
are related to the primary via a foreign key. This way, related data between the primary and the
secondary relations gets fragmented in the same way.

Vertical Fragmentation
 Each site may not need all the attributes of a relation, which would indicate the need for a
different type of fragmentation.
 Vertical fragmentation divides a relation "vertically" by columns.
 A vertical fragment of a relation keeps only certain attributes of the relation.
 For example, we may want to fragment the EMPLOYEE relation into two vertical fragments.
 The first fragment includes personal information—NAME, BDATE, ADDRESS, and
SEX—and
 The second includes work-related information—SSN, SALARY, SUPERSSN, DNO.
 This vertical fragmentation is not quite proper because, if the two fragments are stored
separately, we cannot put the original employee tuples back together, since there is no common
attribute between the two fragments.

AUWC School of Technology and InformaticsPage 59

 Advanced Database Management
 It is necessary to include the primary key or some candidate key attribute in every vertical
fragment so that the full relation can be reconstructed from the fragments. Hence, we must add
the SSN attribute to the personal information fragment.

Data Replication and Allocation

 Replication is useful in improving the availability of data.
 The most extreme case is replication of the whole database at every site in the distributed system,
thus creating a fully replicated distributed database.
 This can improve availability remarkably because the system can continue to operate
as long as at least one site is up.
 It also improves performance of retrieval for global queries, because the result of such
a query can be obtained locally from any one site; hence, a retrieval query can be
processed at the local site where it is submitted, if that site includes a server module.
 The disadvantage of full replication is that it can slow down update operations drastically,
since a single logical update must be performed on every copy of the database to keep the copies
consistent. This is especially true if many copies of the database exist.
 Full replication makes the concurrency control and recovery techniques more expensive than
they would be if there were no replication,
 The other extreme from full replication involves having no replication—that is, each fragment is
stored at exactly one site.
 In this case all fragments must be disjoint, except for the repetition of primary keys among
vertical (or mixed) fragments. This is also called non redundant allocation.
 Between these two extremes, we have a wide spectrum of partial replication of the data—that
is, some fragments of the database may be replicated whereas others may not.
 The number of copies of each fragment can range from one up to the total number of sites in
the distributed system.
 A special case of partial replication is occurring heavily in applications where mobile workers—
such as sales forces, financial planners, and claims adjustors—carry partially replicated databases
with them on laptops and personal digital assistants and synchronize them periodically with the
server database.
 A description of the replication of fragments is sometimes called a replication schema.
 Each fragment—or each copy of a fragment—must be assigned to a particular site in the
distributed system. This process is called data distribution (or data allocation).
 The choice of sites and the degree of replication depend on the performance and availability
goals of the system and on the types and frequencies of transactions submitted at each site .
 For example, if high availability is required and transactions can be submitted at any site and if
most transactions are retrieval only, a fully replicated database is a good choice.
 However, if certain transactions that access particular parts of the database are mostly submitted
at a particular site, the corresponding set of fragments can be allocated at that site only.
 Data that is accessed at multiple sites can be replicated at those sites. If many updates are
performed, it may be useful to limit replication. Finding an optimal or even a good solution to
distributed data allocation is a complex optimization problem.

AUWC School of Technology and InformaticsPage 60

 Advanced Database Management
Types of Distributed Database Systems
 The term distributed database management system can describe various systems that differ from
one another in many respects.
 The main thing that all such systems have in common is the fact that data and software are
distributed over multiple sites connected by some form of communication network.
 In this section we discuss a number of types of DDBMSs and the criteria and factors that make
some of these systems different. The first factor we consider is the degree of homogeneity of the
DDBMS software. If all servers (or individual local DBMSs) use identical software and all
users (clients) use identical software, the DDBMS is called homogeneous; otherwise, it is
called heterogeneous.
 Another factor related to the degree of homogeneity is the degree of local autonomy.
 If there is no provision for the local site to function as a stand-alone DBMS, then the system
has no local autonomy.
 On the other hand, if direct access by local transactions to a server is permitted, the system has
some degree of local autonomy. At one extreme of the autonomy spectrum, we have a DDBMS
that "looks like" a centralized DBMS to the user.
 A single conceptual schema exists, and all access to the system is obtained through a site that is
part of the DDBMS—which means that no local autonomy exists.
 At the other extreme we encounter a type of DDBMS called a federated DDBMS (or a multi
database system).
 In such a system, each server is an independent and autonomous centralized DBMS that has its
own local users, local transactions, and DBA and hence has a very high degree of local
autonomy.
 The term federated database system (FDBS) is used when there is some global view or schema
of the federation of databases that is shared by the applications.
 On the other hand, a multi database system does not have a global schema and interactively
constructs one as needed by the application. Both systems are hybrids between distributed and
centralized systems and the distinction we made between them is not strictly followed.
 In a heterogeneous FDBS, one server may be a relational DBMS, another network DBMS, and a
third an object or hierarchical DBMS; in such a case it is necessary to have a canonical system
language and to include language translators to translate sub queries from the canonical language
to the language of each server.

Federated Database Management Systems Issues

 The type of heterogeneity present in FDBSs may arise from several sources.
 We discuss these sources first and then point out how the different types of autonomies contribute
to a semantic heterogeneity that must be resolved in a heterogeneous FDBS.
A. Differences in data models: Databases in an organization come from a variety of data
models including the so-called legacy models, the relational data model, the object data
model, and even files.
The modeling capabilities of the models vary. Hence, to deal with them uniformly via a single
global schema or to process them in a single language is challenging.
Even if two databases are both from the RDBMS environment, the same information may be
represented as an attribute name, as a relation name, or as a value in different databases.

AUWC School of Technology and InformaticsPage 61

 Advanced Database Management
This calls for an intelligent query processing mechanism that can relate information based on
metadata.
B. Differences in constraints: Constraint facilities for specification and implementation vary
from system to system. There are comparable features that must be reconciled in the
construction of a global schema. For example, the relationships from ER models are
represented as referential integrity constraints in the relational model. Triggers may have to
be used to implement certain constraints in the relational model. The global schema must
also deal with potential conflicts among constraints.
C. Differences in query languages: Even with the same data model, the languages and their
versions vary. For example, SQL has multiple versions like SQL-89, SQL-92 (SQL2), and
SQL3, and each system has its own set of data types, comparison operators, string
manipulation features, and so on.

Semantic Heterogeneity
 Semantic heterogeneity occurs when there are differences in the meaning, interpretation, and
intended use of the same or related data.
 Semantic heterogeneity among component database systems (DBSs) creates the biggest hurdle in
designing global schemas of heterogeneous databases.
 The design autonomy of component DBSs refers to their freedom of choosing the following
design parameters, which in turn affect the eventual complexity of the FDBS:
A. The universe of discourse from which the data is drawn: For example, two customer
accounts databases in the federation may be from United States and Japan with entirely
different sets of attributes about customer accounts required by the accounting practices.
Currency rate fluctuations would also present a problem. Hence, relations in these two
databases which have identical names—CUSTOMER or ACCOUNT—may have some
common and some entirely distinct information.
B. Representation and naming: The representation and naming of data elements and the
structure of the data model may be pre specified for each local database.
C. The understanding, meaning, and subjective interpretation of data. This is a chief
contributor to semantic heterogeneity.
D. Transaction and policy constraints: this deal with Serializability criteria, compensating
transactions, and other transaction policies.
E. Derivation of summaries: Aggregation, summarization, and other data-processing features
and operations supported by the system.
 Communication autonomy of a component DBS refers to it stability to decide whether to
communicate with another component DBSs.
 Execution autonomy refers to the ability of a component DBS to execute local operations
without interference from external operations by other component DBSs and its ability to decide
the order in which to execute them.
 The association autonomy of a component DBS implies that it has the ability to decide whether
and how much to share its functionality (operations it supports) and resources (data it manages)
with other component DBSs.
 The major challenge of designing FDBSs is to let component DBSs interoperate while still
providing the above types of autonomies to them.

AUWC School of Technology and InformaticsPage 62

 Advanced Database Management
 Typical five-level schema architecture to support global applications in the FDBS environment.
In this architecture, the local schema is the conceptual schema (full database definition) of a
component database, and the component schema is derived by translating the local schema into a
canonical data model or common data model (CDM) for the FDBS.
 Schema translation from the local schema to the component schema is accompanied by
generation of mappings to transform commands on a component schema into commands on the
corresponding local schema.
 The export schema represents the subset of a component schema that is available to the FDBS.
 The federated schema is the global schema or view, which is the result of integrating all the
shareable export schemas.
 The external schemas define the schema for a user group or an application, as in the three-level
schema architecture.

Query Processing in Distributed Databases

Data Transfer Costs of Distributed Query Processing
 In a distributed system, several additional factors further complicate query processing.
 The first is the cost of transferring data over the network.
 This data includes intermediate files that are transferred to other sites for further processing,
as well as the final result files that may have to be transferred to the site where the query
result is needed.
 Although these costs may not be very high if the sites are connected via a high-performance
local area network, they become quite significant in other types of networks.
 Hence, DDBMS query optimization algorithms consider the goal of reducing the amount of
data transfer as an optimization criterion in choosing a distributed query execution strategy.
 We illustrate this with two simple example queries.
 Suppose that the EMPLOYEE and DEPARTMENT relations of.
 The size of the EMPLOYEE relation is 100 * 10,000 = 106 bytes, and
 The size of the DEPARTMENT relation is 35 * 100 = 3500 bytes.
 Consider the query Q: "For each employee, retrieve the employee name and the name of the
department for which the employee works." This can be stated as follows in the relational
algebra:

Q: pFNAME, LNAME, DNAME (EMPLOYEEDNO=DNUMBER DEPARTMENT)

 The result of this query will include 10,000 records, assuming that every employee is related to a
department.
 Suppose that each record in the query result is 40 bytes long.
 The query is submitted at a distinct site 3, which is called the result site because the query result
is needed there.
 Neither the EMPLOYEE nor the DEPARTMENT relations reside at site 3.

There are three simple strategies for executing this distributed query:
1. Transfer both the EMPLOYEE and the DEPARTMENT relations to the result site, and
perform the join at site 3. In this case a total of 1,000,000+ 3500 = 1,003,500 bytes must be
transferred.

AUWC School of Technology and InformaticsPage 63

 Advanced Database Management
2. Transfer the EMPLOYEE relation to site 2, execute the join at site 2, and send the result to site
the size of the query result is 40 * 10,000 = 400,000 bytes, so 400,000 + 1,000,000 = 1,400,000
bytes must be transferred.
If minimizing the amount of data transfer is our optimization criterion, we should choose strategy 3. Now
consider another query Q: "For each department, retrieve the department name and the name of the
department manager." This can be stated as follows in the relational algebra:

An Overview of Client-Server Architecture and Its Relationship to Distributed

Databases
 Distributed database applications are being developed in the context of the client-server
architecture.
 Exactly how to divide the DBMS functionality between client and server has not yet been
established.
 Different approaches have been proposed. One possibility is to include the
functionality of a centralized DBMS at the server level.
 A number of relational DBMS products have taken this approach, where an SQL server is
provided to the clients.
 Each client must then formulate the appropriate SQL queries and provide the user interface and
programming language interface functions.
 Since SQL is a relational standard, various SQL servers, possibly provided by different vendors,
can accept SQL commands.
 The client may also refer to a data dictionary that includes information on the distribution of data
among the various SQL servers, as well as modules for decomposing a global query into a
number of local queries that can be executed at the various sites.
 Interaction between client and server might proceed as follows during the processing of an SQL
query:
1. The client parses a user query and decomposes it into a number of independent site
queries. Each site query is sent to the appropriate server site.
2. Each server processes the local query and sends the resulting relation to the client site.
3. The client site combines the results of the sub queries to produce the result of the originally
submitted query.
 In this approach, the SQL server has also been called a transaction server (or a database
processor (DP) or a back-end machine), whereas the client has been called an application
processor (AP)(or a front-end machine).
 The interaction between client and server can be specified by the user at the client level or via a
specialized DBMS client module that is part of the DBMS package.
 For example, the user may know what data is stored in each server, break down a query
request into site sub queries manually, and submit individual sub queries to the various
sites.
 The resulting tables may be combined explicitly by a further user query at the client level.
 The alternative is to have the client module undertake these actions automatically.
 In a typical DDBMS, it is customary to divide the software modules into three levels:
1. The server software is responsible for local data management at a site, much like
centralized DBMS software.

AUWC School of Technology and InformaticsPage 64

 Advanced Database Management
2. The client software is responsible for most of the distribution functions; it accesses data
distribution information from the DDBMS catalog and processes all requests that
require access to more than one site. It also handles all user interfaces.
3. The communications software (sometimes in conjunction with a distributed operating
system) provides the communication primitives that are used by the client to transmit
commands and data among the various sites as needed. This is not strictly part of the
DDBMS, but it provides essential communication primitives and services.
 The client is responsible for generating a distributed execution plan for a multisite query or
transaction and for supervising distributed execution by sending commands to servers.
 These commands include local queries and transactions to be executed, as well as commands to
transmit data to other clients or servers.
 Hence, client software should be included at any site where multisite queries are submitted.
Another function controlled by the client (or coordinator) is that of ensuring consistency of
replicated copies of a data item by employing distributed (or global) concurrency control
techniques.
 The client must also ensure the atomicity of global transactions by performing global recovery
when certain sites fail.

One possible function of the client is to hide the details of data distribution from the user; that is, it
enables the user to write global queries and transactions as though the database were centralized,
without having to specify the sites at which the data referenced in the query or transaction resides. This
property is called distribution transparency.

AUWC School of Technology and InformaticsPage 65

 Advanced Database Management
CHAPTER SEVEN
Spatial /multimedia/mobile databases
What is a Spatial Database?
 A spatial database is a database that is enhanced to store and access spatial data or data that
defines a geometric space.
 Data on spatial databases are stored as coordinates, points, lines, polygons and topology.
 Spatial RDBMS allows to use SQL data types, such as int and varchar, as well as spatial data
types, such as Point, Line string and Polygon for geometric calculations like distance or
relationships between shapes.
 RDBMS uses the B-Tree series or Hash Function to process indexes; only one-dimensional data
can be processed. Since spatial data types are two or three dimensional:
 R-Tree series or Quad Trees can be used in Spatial RDBMS to process such data;
 Or it is necessary to transform two- or three-dimensional data to one dimensional data,
then B-Tree can be used.
Spatial Database Applications
GIS applications (maps):
 Urban planning, route optimization, fire or pollution monitoring, utility networks, etc.
Other applications:
 VLSI design, CAD/CAM, model of human brain, etc.
Traditional applications:
 Multidimensional records
Spatial Data Types
line region
point

Point: 2 real numbers

Line: sequence of points
Region: area included inside n-points
An Example of Spatial Representation

AUWC School of Technology and InformaticsPage 66

 Advanced Database Management

Spatial Relationship
 Topological relationships:
 adjacent, inside, disjoint, etc.
 Direction relationships:
 Above, below, north of, etc.
 Metric relationships: “distance < 100”
EXAMPLE
A database:
 Relation states (sname: string, area: region, spop: int)
 Relation cities (cname: string, center: point; ext: region)
 Relation rivers (rname: string, route:line)
SELECT * FROM rivers WHERE route intersects Ethiopia
SELECT cname, sname FROM cities, states WHERE center inside area
SELECT rname, length (intersection (route, California)) FROM rivers WHERE route intersects
Oromia
Spatial Queries
 Selection queries:
“Find all objects inside query q”, inside-> intersects, north
 Nearest Neighbor-queries:
“Find the closets object to a query point q”, k-closest objects
 Spatial join queries: Two spatial relations S1 and S2, find all pairs: {x in S1, y in S2, and x
rel y= true}, rel= intersect, inside, etc

AUWC School of Technology and InformaticsPage 67

 Advanced Database Management
Example of Spatial SQL Nearest Neighbor search

Access Methods
Point Access Methods (PAMs):
 Index methods for 2 or 3-dimensional points (k-d trees, Z-ordering, grid-file)
Spatial Access Methods (SAMs):
 Index methods for 2 or 3-dimensional regions and points (R-trees)
Mobile Database
A mobile database is either a stationary database that can be connected to by a mobile computing device
such as smart phones or PDAs over a mobile network, or a database which is actually carried by the
mobile device. This could be a list of contacts, price information, distance travelled, or any other
information.

 Many applications require the ability to download information from an information

repository and operate on this information even when out of range or disconnected.

Mobile databases are highly concentrated in the retail and logistics industries. They are
increasingly being used in aviation and transportation industry.

Home Directory
An example of this is a mobile workforce. In this scenario, a user would require access to update
information from files in the home directories on a server or customer records from a database.

A home directory is a file system directory on a multi-user operating system containing files for a
given user of the system.

AUWC School of Technology and InformaticsPage 68

 Advanced Database Management
The specifics of the home directory (such as its name and location) is defined by the operating
system involved; for example, Windows systems between 2000 and 2003 keep home directories
in a folder called Documents and Settings.

Mobile Database Considerations

Mobile users must be able to work without a network connection due to poor or even non-existent
connections.

A cache could maintain to hold recently accessed data and transactions so that they are not lost
due to connection failure.

Users might not require access to truly live data, only recently modified data, and uploading of
changing might be deferred until reconnected.

Bandwidth must be conserved (a common requirement on wireless networks that charge per
megabyte.

Mobile computing devices tend to have slower CPUs and limited battery life.

Users with multiple devices (i.e.: smartphone and tablet) may need to synchronize their devices to
a centralized data store. This may require application-specific automation features.

Users may change location geographically and on the network. Usually dealing with this, is left to
the operating system, which is responsible for maintaining the wireless network connection.

Mobile Database Capabilities

 Can physically move around without affecting data
availability
 Can reach to the place data is stored
 Can process special types of data efficiently
 Not subjected to connection restrictions
 Very high reachability
 Highly portable
Mobile Database Limitations
 Limited wireless bandwidth
 Wireless communication speed
 Limited energy source (battery power)
 Less secured
 Vulnerable to physical activities
 Hard to make theft proof.
Mobile Database Applications
 Insurance companies
 Emergencies services (Police, medical,
etc.)

AUWC School of Technology and InformaticsPage 69

 Advanced Database Management
 Traffic control
 Taxi dispatch
 E-commerce
Multimedia Database Management System (MMDBMS)
A true MMDBMS should be able to:
Operate (create, update, delete, retrieve) with at least all the audio-visual multimedia data types
(text, images, video, audio) and perhaps some non-audio-visual types (e.g. olfactory, taste,
haptic)
Fullfil all the standard requirements of a DBMS i.e. data integrity, ppersistence, transaction
management, concurrency control, system recovery, queries, versioning, data integrity, data
security, etc.
Manipulate huge data volumes (virtually no restriction concerning the number of multimedia
structures and their size)
Allow interaction with the user e.g. object searches, generated media
Retrieve multimedia data based on their content (attributes, features and concepts)
Efficiently manage data distribution over the nodes of a computer network (distributed
architecture)
By way of combination of basic multimedia objects, it should be possible to create new complex
objects (i.e. “specialisation” and “inheritance”)
Because many different data types are involved, special methods may be needed for optimal
storage, access, indexing, and retrieval - in particular, for time-based media objects
May include advanced features such as character recognition, voice recognition, and image
analysis
With hypermedia databases, “Hyper bases”, links between generated pages/reports are derived
from associations between objects
Many hypermedia systems are based on relational or object-relational database technologies e.g.
Active Server Pages, ColdFusion, Java Server Pages

Multimedia Data Types

 DBMS provide different kinds of domains for multimedia data:
Large object domains – these are long unstructured sequences of data e.g. Bbinary
Large Objects (BLOBs) or Character Large Objects (CLOBs)
File references – instead of holding the data, a file reference contains a link to the data
MS Windows Object Linking and Embedding (OLE) as in Microsoft Access
Actual multimedia domains
 object-relational databases support specialised multimedia data types
 object-oriented databases support specialised multimedia classes
Multimedia Database Models
 Multimedia File Managers
Programs which organise, index, catalogue, search and retrieve multimedia objects stored
as conventional standalone files in the operating system environment
May include thumbnail panel describing each object and its file characteristics
May also include a “preview” facility, and / or facility to launch associated media-editing
tools

AUWC School of Technology and InformaticsPage 70

 Advanced Database Management
Vary from simple “scrapbook” catalogue to complex programs offering off-line storage
or mounting of files via network
May support data compression, such as JPEG
 Relation DBMS with Extended / Add-on Capabilities
extension of ANSI SQL data types to facilitate storage and manipulation of binary data
e.g. BLOBs
use of filters / windows / external viewers / OLE to handle rich multimedia data
 Object-Oriented Databases
handle all kinds of multimedia data is a specialised classis
quite low market share in database technology
 Object-Relational Databases
hybrid combination of features of RDBMS and OODB
fully SQL-compliant, with added support for complex data types via object definition
language (ODL)
 Document Management Systems / Content Management Systems
Specialised systems for managing documents/content of many different types
Automate workflow, pages/reports generated automatically
Very popular for Web sites where content changes rapidly e.g. Kenny’s Bookshop
(www.kennys.ie), Irish Times (ireland.com)
Can buy off-the-shelf (e.g. Terminal Four, Broad Vision), acquire via open source, or
build it yourself! (e.g. Lotus Notes)
Multimedia database management system: system requirements
 High network and communications bandwidth
network standards such as asynchronous transfer mode (ATM), fibre distributed data
interface (FDDI), and Fast Ethernet
improved communications through ISDN, fibre optic technologies, wireless networks,
and cable TV
 High capacity storage, and fast retrieval mechanisms
enhancements in compression algorithms
 Uncompressed CD = 10 MB/min (30 songs approx. 1GB)
 MP3 high-quality compression = 1 MB/min (300 songs approx. 1GB)
 MPEG-AAC (advanced audio) = 500 KB/min (600 songs approx. 1 GB)
optical disks and magneto-optic technologies can store many terabytes (TB) of data
which is accessible near-line
 Highly-specified desktop PCs / workstations
desktop machines must be capable of processing multimedia data, and must be fitted with
requisite peripheral devices
 User interfaces
 interfaces need to be redesigned to facilitate real-time multi-user interactive multimedia
applications
Multimedia data storage
 High capacity storage, and fast retrieval mechanisms
 preferred approach storage is HSM (hierarchical storage management)
AUWC School of Technology and InformaticsPage 71
 Advanced Database Management

Main
Memory
increasing probability (RAM) increasing storage capacity
of access increasing permanence
increasing cost increasing access
improving time
On-line devices:
performance magnetic disks,
optical disks

Near-line devices:
optical storage

Off-line devices:
magnetic tapes, optical storage

Applications of Multimedia Databases

Generic business & Office Information Systems: document imaging, document editing tools,
mmultimedia email, mmultimedia conferencing, mmultimedia workflow management, teleworking
Software development: multimedia object libraries; computer-aided software engineering
(CASE); multimedia communication histories; multimedia system artefacts
Education: multimedia encyclopaedia; multimedia courseware & training materials; education-
on-demand (distance education, JIT learning)
Banking: tele-banking
Retail: home shopping; customer guidance
Tourism & hospitality: trip visualisation & planning; sales & marketing
Publishing: electronic publishing; document editing tools; multimedia archives
Medicine: digitised x-rays; patient histories; image analysis; tele-consultation
Criminal investigation: biometrics; fingerprint matching; “photo-fit” analysis
entertainment: video-on-demand; interactive TV; Virtual Reality gaming
Museums & Libraries
science: spatial data analysis; cartographic databases; geographic information systems (GIS’s)
engineering: computer-aided design / manufacture (CAD / CAM); collaborative design;
concurrent engineering
pharmaceuticals: dossier management for new drug applications (NDA’s)

AUWC School of Technology and InformaticsPage 72

 Advanced Database Management
CHAPTER EIGHT
WEB- BASED DATABASES
Databases on the World Wide Web
The World Wide Web (WWW)—popularly known as "the Web"—originally developed in Switzerland at
CERN in early 1990 as a large-scale hypermedia information service system for biological scientists to
share information. Today this technology allows universal access to this shared information to anyone
having access to the Internet and the Web contains hundreds of millions of Web pages within the reach of
millions of users.

In Web technology, basic client-server architecture underlies all activities. Information is stored on
computers designated as Web servers in publicly accessible shared files encoded using Hyper Text
Markup Language (HTML).

A number of tools enable users to create Web pages formatted with HTML tags, freely mixed with
multimedia content from graphics to audio and even to video. A page has many interspersed hyperlinks
literally a link that enables a user to "browse" or move from one page to another across the Internet. This
ability has given a tremendous power to end users in searching and navigating related information often
across different continents.

Information on the Web is organized according to a Uniform Resource Locator (URL) something
similar to an address that provides the complete pathname of a file. The pathname consists of a string of
machine and directory names separated by slashes and ends in a filename. For example, the table of
contents of this book is currently at the following URL:

A URL always begins with a hypertext transport protocol (http), which is the protocol used by the Web
browsers, a program that communicates with the Web server, and vice versa. Web browsers interpret and
present HTML documents to users. Popular Web browsers include the Internet Explorer of Microsoft and
the Netscape Navigator. A collection of HTML documents and other files accessible via the URL on a
Web server is called a Web site. In the above URL, "www.awl.com" may be called the Web site of
Addison Wesley Publishing.

Providing Access to Databases on the World Wide Web

Today’s technology has been moving rapidly from static to dynamic Web pages, where content may be in
a constant state of flux. The Web server uses a standard interface called the Common Gateway Interface
(CGI) to act as the middleware the additional software layer between the user interface front-end and the
DBMS back-end that facilitates access to heterogeneous databases.

AUWC School of Technology and InformaticsPage 73

 Advanced Database Management
The CGI middleware executes external programs or scripts to obtain the dynamic information, and it
returns the information to the server in HTML, which is given back to the browser. As the Web
undergoes its latest transformations, it has become necessary to allow users access not only to file systems
but to databases and DBMSs to support query processing, report generation, and so forth. The existing
approaches may be divided into two categories:

1. Access using CGI scripts: The database server can be made to interact with the Web server via
CGI. The main disadvantage of this approach is that for each user request, the Web server must
start a new CGI process: each process makes a new connection with the DBMS and the Web
server must wait until the results are delivered to it. No efficiency is achieved by any grouping of
multiple users’ requests; moreover, the developer must keep the scripts in the CGI-bin
subdirectories only, which opens it to a possible breach of security. The fact that CGI has no
language associated with it but requires database developers to learn PERL or Tcl is also a
drawback. Manageability of scripts is another problem if the scripts are scattered everywhere.
2. Access using JDBC: JDBC is a set of Java classes developed by Sun Microsystems to allow
access to relational databases through the execution of SQL statements. It is a way of connecting
with databases, without any additional processes for each client request. Note that JDBC is a
name trademarked by Sun; it does not stand for Java Data Base connectivity as many believe.
JDBC has the capabilities to connect to a database, send SQL statements to a database and to
retrieve the results of a query using the Java classes Connection, Statement, and Result Set
respectively. With Java’s claimed platform independence, an application may run on any Java-
capable browser, which loads the Java code from the server and runs it on the client’s browser.
The Java code is DBMS transparent; the JDBC drivers for individual DBMSs on the server end
carry the task of interacting with that DBMS. If the JDBC driver is on the client, the application
runs on the client and its requests are communicated to the DBMS directly by the driver. For
standard SQL requests, many RDBMSs can be accessed this way. The drawback of using JDBC
is the prospect of executing Java through virtual machines with inherent efficiency. The JDBC
bridge to Object Database Connectivity (ODBC) remains another way of getting to the
RDBMSs.

Besides CGI, other Web server vendors are launching their own middleware products for providing
multiple database connectivity. These include Internet Server API (ISAPI) from Microsoft and
Netscape API (NSAPI) from Netscape

AUWC School of Technology and InformaticsPage 74

 Advanced Database Management
The Web Integration Option of INFORMIX
Informix has addressed the limitations of CGI and the incompatibilities of CGI, NSAPI, and ISAPI by
creating the Web Integration Option (WIO).

WIO eliminates the need for scripts. Developers use tools to create intelligent HTML pages called
Application Pages (or App Pages) directly within the database. They execute SQL statements
dynamically, format the results inside HTML, and return the resulting Web page to the end users.

Driver, a lightweight CGI process that is invoked when a URL request is received by the Web server. A
unique session identifier is generated for each request but the WIO application is persistent and does not
terminate after each request. When the WIO application receives a request from the Web driver, it
connects to the database and executes Web Explode, a function that executes queries within Web pages
and formats results as a Web page that goes back to the browser via the Web driver.

Informix HTML tag extensions allow Web authors to create applications that can dynamically construct
Web page templates from the Informix Dynamic Server and present them to the end users .

WIO also lets users create their own customized tags to perform specialized tasks. Thus, without resorting
to any programming or script development, powerful applications can be designed.

Another feature of WIO helps transaction-oriented applications by providing an application programming

interface (API) that offers a collection of basic services such as connection and session management that
can be incorporated into Web application.

WIO supports applications developed in C, C++, and Java. This flexibility lets developer’s port existing
applications to the Web or develops new applications in these languages. The WIO is integrated with
Web server software and utilizes the native security mechanism of the Informix Dynamic Server. The
open architecture of WIO allows the use of various Web browsers and servers.

The ORACLE Web Server

The client requests files that are called "static" or "dynamic" files from the Web server. Static files have a
fixed content whereas dynamic files may have content that includes results of queries to the database.

There is an HTTP demon (a process that runs continuously) called Web Listener running on the server
that listens for the requests originating in the clients. A static file (document) is retrieved from the file
system of the server and displayed on the Web browser at the client. Request for a dynamic page is
passed by the listener to a Web request broker (WRB), which is a multi-threaded dispatcher that adheres

AUWC School of Technology and InformaticsPage 75

 Advanced Database Management
to cartridges. Cartridges are software modules that perform specific functions on specific types of data;
they can communicate among themselves.

Currently cartridges are provided for PL/SQL, Java, and Live HTML; customized cartridges may be
provided as well. Web Server has been fully integrated with PL/SQL, making it efficient and scalable.

The cartridges give it additional flexibility, making it possible to work with other languages and software
packages. An advanced secure sockets layer may be used for secure communication over the Internet. The
Designer 2000 development has a Web generator that enables previous applications developed for LANs
to be ported to the Internet and Intranet environments.

Open Problems with Web Databases

The Web is an important factor in planning for enterprise-wide computing environments, both for
providing external access to the enterprise’s systems and information for customers and suppliers and for
marketing and advertising purposes. At the same time, due to security requirements, employees of some
organizations are restricted to operate within intranets—sub networks that cannot be accessed freely from
the outside world.

Among the prominent applications of the intranet and the WWW are databases to support electronic
storefronts, parts and product catalogs, directories and schedules, newsstands, and bookstores. Electronic
commerce the purchasing of products and services electronically on the Internet is likely to become a
major application supported by such databases.

The future challenges of managing databases on the Web will be many, among them the following:

 Web technology needs to be integrated with the object technology. Currently, the web can be
viewed as a distributed object system, with HTML pages functioning as objects identified by the
URL.
 HTML functionality is too simple to support complex application requirements. As we saw, the
Web Integration Option of Informix adds further tags to HTML. In general, additional facilities
will be needed to

1. To make Web clients function as application front ends, integrating data from multiple
heterogeneous databases;
2. To make Web clients present different views of the same data to different users; and
3. To make Web clients "intelligent" by providing additional data mining functionality

 Web page content can be made more dynamic by adding more "behavior" to it as an object (In
this respect

AUWC School of Technology and InformaticsPage 76

 Advanced Database Management
1. client and server objects (HTML pages) can be made to interact;
2. Web pages can be treated as collections of programmable objects; and
3. Client-side code can access these objects and manipulate them dynamically.
 The support for a large number of clients coupled with reasonable response times for queries
against very large (several tens of gigabytes in size) databases will be major challenges for Web
databases. They will have to be addressed both by Web servers and by the underlying DBMSs.
Efforts are underway to address the limitations of the current data structuring technology,
particularly by the World Wide Web Consortium (W3C). The W3C is designing a Web Object
Model. W3C is also proposing an Extensible Markup Language (XML) for structured document
interchange on the Web.

XML defines a subset of SGML (the Standard Generalized Markup Language), allowing customization of
markup languages with application-specific tags. XML is rapidly gaining ground due to its extensibility in
defining new tags.

W3C’s Document Object Model (DOM) defines an object-oriented API for HTML or XML documents
presented by a Web client. W3C is also defining metadata modeling standards for describing Internet
resources.

The technology to model information using the standards discussed above and to find information on the
Web is undergoing a major evolution. Overall, the Web servers have to gain robustness as a reliable
technology to handle production-level databases for supporting 24x7 applications—24 hours a day, 7
days a week.

Security remains a critical problem for supporting applications involving financial and medical databases.
Moreover, transferring from existing database application environments to those on the Web will need
adequate support that will enable users to continue their current mode of operation and an expensive
infrastructure for handling migration of data among systems without introducing inconsistencies. The
traditional database functionality of querying and transaction processing must undergo appropriate
modifications to support Web-based applications. One such area is mobile databases.

AUWC School of Technology and InformaticsPage 77

 Advanced Database Management
CHAPTER NINE
DATA WAREHOUSING
What is Data Warehouse?
 A data warehouse is a subject oriented, integrated, time-variant, and non-volatile
collection of data.
 This data helps analysts to take informed decisions in an organization.
 A Data Warehouse consists of data from multiple heterogeneous data sources and is
used for analytical reporting and decision making.
 Data Warehouse is a central place where data is stored from different data sources and
applications.
 An operational database undergoes frequent changes on a daily basis on account of the
transactions that take place.
 Suppose a business executive wants to analyze previous feedback on any data such as
a product, a supplier, or any consumer data, then the executive will have no data
available to analyze because the previous data has been updated due to transactions.
 A data warehouses provides us generalized and consolidated data in multidimensional
view.
 Along with generalized and consolidated view of data, a data warehouses also
provides us Online Analytical Processing (OLAP) tools.
These tools help us in interactive and effective analysis of data in a multidimensional space.
This analysis results in data generalization and data mining.
 Data mining functions such as association, clustering, classification, prediction can be
integrated with OLAP operations to enhance the interactive mining of knowledge at
multiple level of abstraction. That's why data warehouse has now become an important
platform for data analysis and online analytical processing.

 In the above image, you can see that the data is coming from multiple heterogeneous
data sources to a Data Warehouse. Common data sources for a data warehouse includes.
 Operational databases  Flat Files
 SAP and non-SAP Applications

AUWC School of Technology and InformaticsPage 78

 Advanced Database Management
 Data in data warehouse is accessed by BI Business Intelligence users for Analytical
Reporting, Data Mining and Analysis. This is used for decision making by Business
Users, Sales Manager, and Analysts to define future strategy.
Understanding a Data Warehouse
 A data warehouse is a database, which is kept separate from the organization's
operational database.
 There is no frequent updating done in a data warehouse.
 It possesses consolidated historical data, which helps the organization to analyze its
business.
 A data warehouse helps executives to organize, understand, and use their data to take
strategic decisions.
 Data warehouse systems help in the integration of diversity of application systems.
Why a Data Warehouse is separated from Operational Databases?
A data warehouses is kept separate from operational databases due to the following reasons −
 An operational database is constructed for well-known tasks and workloads such as
searching particular records, indexing, etc. In contract, data warehouse queries are
often complex and they present a general form of data.
 Operational databases support concurrent (parallel) processing of multiple
transactions. Concurrency control and recovery mechanisms are required for
operational databases to ensure robustness and consistency of the database.
 An Operational Database query allows reading and modifying operations insert, delete
and Update while an OLAP query needs only read-only access of stored data Select
statement.
 An operational database maintains current data of an organization. On the other hand,
a data warehouse maintains historical data.
Data Warehouse Applications
 A data warehouse helps business executives to organize, analyze, and use their data for
decision making.
 A data warehouse serves as a sole part of a plan-execute-assess "closed-loop" feedback
system for the enterprise management.
 Data warehouses are widely used in the following fields
 Financial services  Retail sectors
 Banking services  Controlled manufacturing
 Consumer goods
Types of Data Warehouse
Information processing, analytical processing and data mining are the three types of data
warehouse applications that are discussed below
A. Information Processing − A data warehouse allows to process the data stored in it. The
data can be processed by means of querying, basic statistical analysis, reporting using
crosstabs, tables, charts, or graphs.
B. Analytical Processing − A data warehouse supports analytical processing of the
information stored in it. The data can be analyzed by means of basic OLAP operations,
including slice-and-dice, drill down, drill up, and pivoting.
C. Data Mining − Data mining supports knowledge discovery by finding hidden patterns
and associations, constructing analytical models, performing classification and
prediction. These mining results can be presented using the visualization tools.

AUWC School of Technology and InformaticsPage 79

 Advanced Database Management
Characteristics of a Data Warehouse
The following are the key characteristics of a Data Warehouse −
A. Subject Oriented − In a DW system, the data is categorized and stored by a business
subject rather than by application like equity plans, shares, loans, etc.
B. Integrated − Data from multiple data sources are integrated in a Data Warehouse.
C. Non-Volatile − Data in data warehouse is non-volatile. It means when data is loaded in
DW system, it is not altered.
D. Time Variant − A DW system contains historical data as compared to Transactional
system which contains only current data. In a Data warehouse you can see data for 3
months, 6 months, 1 year, 5 years, etc.
Note − A data warehouse does not require transaction processing, recovery, and concurrency
controls, because it is physically stored and separate from the operational database.
OLTP vs. OLAP
 Firstly, OLTP stands for Online Transaction Processing, while OLAP stands for Online
Analytical Processing
 In an OLTP system, there are a large number of short online transactions such as
INSERT, UPDATE, and DELETE. Whereas, in an OLTP system, an effective measure
is the processing time of short transactions and is very less. It controls data integrity in
multi-access environments. For an OLTP system, the number of transactions per second
measures the effectiveness.
 An OLTP Data Warehouse System contains current and detailed data and is maintained
in the schemas in the entity model 3 NF.
 For Example − A Day-to-Day transaction system in a retail store, where the
customer records are inserted, updated and deleted on a daily basis. It provides
faster query processing. OLTP databases contain detailed and current data. The
schema used to store OLTP database is the Entity model.
 In an OLAP system, there is lesser number of transactions as compared to a
transactional system. The queries executed are complex in nature and involves data
aggregations.

AUWC School of Technology and InformaticsPage 80

 Advanced Database Management

Data Warehouse Architecture

 Data Warehousing involves data cleaning, data integration, and data consolidations.
 A Data Warehouse has a 3-layer architecture.
1. Data Source Layer: -It defines how the data comes to a Data Warehouse. It involves
various data sources and operational transaction systems, flat files, applications, etc.
2. Integration Layer: -It consists of Operational Data Store and Staging area. Staging
area is used to perform data cleansing, data transformation and loading data from
different sources to a data warehouse. As multiple data sources are available for
extraction at different time zones, staging area is used to store the data and later to
apply transformations on data.
3. Presentation Layer: -This is used to perform BI reporting by end users. The data in a
DW system is accessed by BI users and used for reporting and analysis. The
following illustration shows the common architecture of a Data Warehouse System.

AUWC School of Technology and InformaticsPage 81

 Advanced Database Management

What is an Aggregation?
 We save tables with aggregated data like yearly 1row, quarterly 4rows, and monthly
12rows or so, if someone has to do a year to year comparison, only one row will be
processed. However, in an unaggregated table it will compare all the rows. This is called
Aggregation.
 There are various Aggregation functions that can be used in an OLAP system like Sum,
Avg, Max, Min, etc.
 For Example −SELECT Avg(salary) FROM employee WHERE title = 'Programmer';
Key Differences
These are the major differences between an OLAP and an OLTP system.
 Indexes − An OLTP system has only few indexes while in an OLAP system there are
many indexes for performance optimization.
 Joins − In an OLTP system, large number of joins and data are normalized.
However, in an OLAP systems there are less joins and are de-normalized.
 Aggregation − In an OLTP system, data is not aggregated while in an OLAP
database more aggregations are used.
 Normalization − An OLTP system contains normalized data however data is not
normalized in an OLAP system.

What is Data Mining?

 Data Mining is the process of extracting useful information and patterns from
enormous data.
 Data Mining includes collection, extraction, analysis and statistics of data.
 It is also known as Knowledge discovery process, Knowledge Mining from Data or data/
pattern analysis.

AUWC School of Technology and InformaticsPage 82

 Advanced Database Management
 Data Mining is a logical process of finding useful information to find out useful data.
 Once the information and patterns are found it can be used to make decisions for
developing the business.
 Data mining tools can give answers to your various questions related to your business
which was too difficult to resolve. They also forecast the future trends which lets the
business people to make proactive decisions.
 The information or knowledge extracted so can be used for any of the following
applications area:
 Market Analysis  Production Control
 Fraud Detection  Science Exploration
 Customer Retention
(maintenance)
Data mining involves three steps. They are
 Exploration – In this step the data is cleared and converted into another form. The nature
of data is also determined
 Pattern Identification – The next step is to choose the pattern which will make the best
prediction
 Deployment – The identified patterns are used to get the desired outcome.
Benefits of Data Mining
 Automated prediction of trends and behaviors
 It can be implemented on new systems as well as existing platforms
 It can analyze huge database in minutes
 Automated discovery of hidden patterns
 There are a lot of models available to understand complex data easily
 It is of high speed which makes it easy for the users to analyze huge amount of data in
less time
 It yields improved predictions
Data Mining Techniques
 There are several major data mining techniques have been developing and using in data
mining projects recently including association, classification, clustering, prediction,
sequential patterns and decision tree. We will briefly examine those data mining
techniques in the following sections.
1. Association
 In association, a pattern is discovered based on a relationship between items in
the same transaction.
 That’s the reason why association technique is also known as relation
technique.
 The association technique is used in market basket analysis to identify a set of
products that customers frequently purchase together.
2. Classification
 Classification is a classic data mining technique based on machine learning.
 Basically, classification is used to classify each item in a set of data into one of
a predefined set of classes or groups.
 Classification method makes use of mathematical techniques such as decision
trees, linear programming, neural network, and statistics.
 In classification, we develop the software that can learn how to classify the data
items into groups.

AUWC School of Technology and InformaticsPage 83

 Advanced Database Management
 For example, we can apply classification in the application that “given all
records of employees who left the company; predict who will probably leave
the company in a future period.”
 In this case, we divide the records of employees into two groups that named
“leave” and “stay”. And then we can ask our data mining software to classify the
employees into separate groups.
3. Clustering
 Clustering is a data mining technique that makes a meaningful or useful cluster
of objects which have similar characteristics using the automatic technique.
 The clustering technique defines the classes and puts objects in each class, while
in the classification techniques, objects are assigned into predefined classes.
 To make the concept clearer, we can take book management in the library as an
example.
 In a library, there is a wide range of books on various topics available.
 The challenge is how to keep those books in a way that readers can take
several books on a particular topic without hassle.
 By using the clustering technique, we can keep books that have some kinds
of similarities in one cluster or one shelf and label it with a meaningful
name.
 If readers want to grab books in that topic, they would only have to go to
that shelf instead of looking for the entire library.
4. Prediction
 The prediction, as its name implied, is one of a data mining technique that
discovers the relationship between independent variables and relationship
between dependent and independent variables.
 For instance, the prediction analysis technique can be used in the sale to
predict profit for the future if we consider the sale is an independent variable,
profit could be a dependent variable.
 Then based on the historical sale and profit data, we can draw a fitted regression
curve that is used for profit prediction.
5. Sequential Patterns
 Sequential patterns analysis is one of data mining technique that seeks to discover
or identify similar patterns, regular events or trends in transaction data over a
business period.
 In sales, with historical transaction data, businesses can identify a set of items that
customers buy together different times in a year.
 Then businesses can use this information to recommend customers buy it with
better deals based on their purchasing frequency in the past.
6. Decision trees
 The A decision tree is one of the most commonly used data mining techniques
because its model is easy to understand for users.
 In decision tree technique, the root of the decision tree is a simple question or
condition that has multiple answers.
 Each answer then leads to a set of questions or conditions that help us determine
the data so that we can make the final decision based on it.
7. Data Mining Decision Tree

AUWC School of Technology and InformaticsPage 84

 Advanced Database Management
 Starting at the root node, if the outlook is overcast then we should definitely play
tennis. If it is rainy, we should only play tennis if the wind is the week. And if it is
sunny then we should play tennis in case the humidity is normal.
 We often combine two or more of those data mining techniques together to form
an appropriate process that meets the business needs.

AUWC School of Technology and InformaticsPage 85

 Advanced Database Management
Reference
1. David M. Kroenke. (1998). Database processing, 6th ed. Prentice Hall
2. Thomas M. Connolly and Carolyn E.Begg. (2004). A step by step approach to
building databases, 2nded.Pearson Education Limited.
3. Jeffrey A. Hoffer. V.Ramesh. Heikki Topi. Modern Database Management, Tenth
edition.
4. Phill Pratt. Mary Last. Concept of Database Management, 8th edition, Cengage
Learning.
5. Carlos Coronel and Steven Morris. Database System Design, Implementation and
Management, 13th edition.

AUWC School of Technology and InformaticsPage 86