CHAPTER 5:

PHYSICAL DATABASE DESIGN AND

PERFORMANCE

Modern Database Management

1
 OBJECTIVES
 Define terms
 Describe the physical database design process
 Choose storage formats for attributes
 Select appropriate file organizations
 Describe three types of file organization
 Describe indexes and their appropriate use
 Translate a database model into efficient structures
 Know when and how to use denormalization

2
 PHYSICAL DATABASE DESIGN

 Purpose–translate the logical description of

data into the technical specifications for
storing and retrieving data
 Goal–create a design for storing data that
will provide adequate performance and
ensure database integrity, security, and
recoverability

3
 PHYSICAL DESIGN PROCESS
Inputs Decisions
Normalized relations Attribute data types
Volume estimates Physical record descriptions
Attribute definitions (doesn’t always match
logical design)
Response time
expectations File organizations
Leads to
Data security needs Indexes and database
architectures
Backup/recovery needs
Query optimization
Integrity expectations
DBMS technology used

4
 PHYSICAL DESIGN FOR
REGULATORY COMPLIANCE
 Sarbanes- Oxley Act (SOX) – protect investors
by improving accuracy and reliability
 Committee of Sponsoring Organizations (COSO)
of the Treadway Commission
 IT Infrastructure Library (ITIL)
 Control Objectives for Information and Related
Technology (COBIT)
Regulations and standards that impact physical design decisions

5
Figure 5-1 Composite usage map
(Pine Valley Furniture Company)

6
Figure 5-1 Composite usage map
(Pine Valley Furniture Company) (cont.)

Data volumes

7
Figure 5-1 Composite usage map
(Pine Valley Furniture Company) (cont.)

Access Frequencies
(per hour)

8
8
Figure 5-1 Composite usage map
(Pine Valley Furniture Company) (cont.)

Usage analysis:
14,000 purchased parts
accessed per hour 
8000 quotations accessed
from these 140 purchased part
accesses 
7000 suppliers accessed from
these 8000 quotation accesses

9
Figure 5-1 Composite usage map
(Pine Valley Furniture Company) (cont.)
Usage analysis:
7500 suppliers accessed per
hour 
4000 quotations accessed
from these 7500 supplier
accesses 
4000 purchased parts
accessed from these 4000
quotation accesses

10
 DESIGNING FIELDS

 Field: smallest unit of application data

recognized by system software
 Field design
 Choosing data type
 Coding, compression, encryption
 Controlling data integrity

11
CHOOSING DATA TYPES

12
Figure 5-2 Example of a code look-up table
(Pine Valley Furniture Company)

Code saves space, but

costs an additional lookup
to obtain actual value

13
 FIELD DATA INTEGRITY
 Default value–assumed value if no explicit
value
 Range control–allowable value limitations
(constraints or validation rules)
 Null value control–allowing or prohibiting
empty fields
 Referential integrity–range control (and null
value allowances) for foreign-key to primary-
key match-ups
Sarbanes-Oxley Act (SOX) legislates importance of financial data integrity

14
HANDLING MISSING DATA

 Substitute an estimate of the missing value

(e.g., using a formula)
 Construct a report listing missing values
 In programs, ignore missing data unless the
value is significant (sensitivity testing)

Triggers can be used to perform these operations.

15
 DENORMALIZATION
 Transforming normalized relations into non-normalized
physical record specifications
 Benefits:
 Can improve performance (speed) by reducing number of table lookups
(i.e. reduce number of necessary join queries)
 Costs (due to data duplication)
 Wasted storage space
 Data integrity/consistency threats
 Common denormalization opportunities
 One-to-one relationship (Fig. 5-3)
 Many-to-many relationship with non-key attributes (associative entity)
(Fig. 5-4)
 Reference data (1:N relationship where 1-side has data not used in any
other relationship) (Fig. 5-5)

16
Figure 5-3 A possible denormalization situation: two entities with one-
to-one relationship

17
Figure 5-4 A possible denormalization situation: a many-to-many
relationship with nonkey attributes

Extra table
access
required

Null description possible

18
Figure 5-5
A possible
denormalization
situation:
reference data

Extra table
access
required

Data duplication

19
 DENORMALIZE WITH CAUTION
 Denormalization can
 Increase chance of errors and inconsistencies
 Reintroduce anomalies
 Force reprogramming when business rules change
 Perhaps other methods could be used to
improve performance of joins
 Organization of tables in the database (file
organization and clustering)
 Proper query design and optimization

20
 PARTITIONING
 Horizontal Partitioning: Distributing the rows of a
logical relation into several separate tables
 Useful for situations where different users need access to
different rows
 Three types: Key Range Partitioning, Hash Partitioning, or
Composite Partitioning
 Vertical Partitioning: Distributing the columns of a
logical relation into several separate physical tables
 Useful for situations where different users need access to
different columns
 The primary key must be repeated in each file
 Combinations of Horizontal and Vertical
21
 PARTITIONING PROS AND CONS
 Advantages of Partitioning:
 Efficiency: Records used together are grouped together
 Local optimization: Each partition can be optimized for performance
 Security: data not relevant to users are segregated
 Recovery and uptime: smaller files take less time to back up
 Load balancing: Partitions stored on different disks, reduces
contention
 Disadvantages of Partitioning:
 Inconsistent access speed: Slow retrievals across partitions
 Complexity: Non-transparent partitioning
 Extra space or update time: Duplicate data; access from multiple
partitions

22
ORACLE’S HORIZONTAL PARTITIONING
 Range partitioning
 Partitions defined by range of field values
 Could result in unbalanced distribution of rows
 Like-valued fields share partitions
 Hash partitioning
 Partitions defined via hash functions
 Will guarantee balanced distribution of rows
 Partition could contain widely varying valued fields
 List partitioning
 Based on predefined lists of values for the partitioning key
 Composite partitioning
 Combination of the other approaches

23
DESIGNING PHYSICAL DATABASE FILES
 Physical File:
 A named portion of secondary memory allocated for
the purpose of storing physical records
 Tablespace–named logical storage unit in which data
from multiple tables/views/objects can be stored
 Tablespace components
 Segment – a table, index, or partition
 Extent–contiguous section of disk space
 Data block – smallest unit of storage

24
Figure 5-6 DBMS terminology in an Oracle 11g environment

25
FILE ORGANIZATIONS
 Technique for physically arranging records of a file on
secondary storage
 Factors for selecting file organization:
 Fast data retrieval and throughput
 Efficient storage space utilization
 Protection from failure and data loss
 Minimizing need for reorganization
 Accommodating growth
 Security from unauthorized use
 Types of file organizations
 Sequential
 Indexed
 Hashed

26
Figure 5-7a
Sequential file
organization

Records of the If sorted –

every insert or
file are stored delete requires
in sequence by resort
the primary key
field values.
If not sorted
Average time to
find desired record
= n/2

27
 INDEXED FILE ORGANIZATIONS
 Storage of records sequentially or nonsequentially with
an index that allows software to locate individual
records
 Index: a table or other data structure used to determine
in a file the location of records that satisfy some
condition
 Primary keys are automatically indexed
 Other fields or combinations of fields can also be
indexed; these are called secondary keys (or nonunique
keys)

28
Figure 5-7b Indexed file organization

uses a tree search

Average time to find desired
record = depth of the tree

29
Figure 5-7c
Hashed file
organization

Hash algorithm
Usually uses division-
remainder to determine
record position. Records
with same position are
grouped in lists.

30
Figure 6-8 Join Indexes–speeds up join operations

b) Join index for matching foreign

key (FK) and primary key (PK)

a) Join index
for common
non-key
columns

31
32
 CLUSTERING FILES
 In some relational DBMSs, related records from
different tables can be stored together in the same
disk area
 Useful for improving performance of join
operations
 Primary key records of the main table are stored
adjacent to associated foreign key records of the
dependent table
 e.g. Oracle has a CREATE CLUSTER command

33
RULES FOR USING INDEXES

1. Use on larger tables

2. Index the primary key of each table
3. Index search fields (fields frequently in
WHERE clause)
4. Fields in SQL ORDER BY and GROUP BY
commands
5. When there are >100 values but not when there
are <30 values

34
 RULES FOR USING INDEXES (CONT.)
6. Avoid use of indexes for fields with long values;
perhaps compress values first
7. If key to index is used to determine location of
record, use surrogate (like sequence number) to
allow even spread in storage area
8. DBMS may have limit on number of indexes per
table and number of bytes per indexed field(s)
9. Be careful of indexing attributes with null
values; many DBMSs will not recognize null
values in an index search

35
 QUERY OPTIMIZATION
 Parallel query processing–possible when working in multiprocessor
systems
 Overriding automatic query optimization–allows for query writers
to preempt the automated optimization
 Oracle example:

/* */ clause is a hint to override Oracle’s default query plan

36