|
 |

 


  


 `
  
 
˜ What a distributed database management
system (DDBMS) is and what its components
are
˜ How database implementation is affected by
different levels of data and process distribution
˜ How transactions are managed in a distributed
database environment
˜ How database design is affected by the
distributed database environment
¦ 
|
 |  


  


˜ Distributed database management
system (DDBMS)

Ô Governs storage and processing of

logically related data over
interconnected computer systems
in which both data and processing
functions are distributed among
several sites
¦ |
 
|  

  


 
˜ entralized database required that
corporate data be stored in a
single central site

˜ Dynamic business environment

and centralized database¶s
shortcomings spawned a demand
for applications based on data
access from different sources at
multiple locations
  |  


  

 ||
  

˜ Data are located near ³greatest demand´ site
˜ Faster data access
˜ Faster data processing
˜ Growth facilitation
˜ Improved communications
˜ Reduced operating costs
˜ User-friendly interface
˜ Less danger of a single-point failure
˜ Processor independence
 ||
|
  

˜ omplexity of management and control
˜ Security
˜ Lack of standards
˜ Increased storage requirements
˜ Greater difficulty in managing the data
environment
˜ Increased training cost
|
  


  
|
 |  

  
  
 
|
 

  


˜ ^pplication interface
˜ Validation
˜ Transformation
˜ Query optimization
˜ Mapping
˜ I/O interface
˜ Formatting
˜ Security
˜ Backup and recovery
˜ DB administration
˜ oncurrency control
˜ Transaction management
  
 

|
 
  


 
˜ Must perform all the functions of a
centralized DBMS

˜ Must handle all necessary

functions imposed by the
distribution of data and processing

˜ Must perform these additional

functions ?

?to the end
user
!|
 
|  

  


 ||
 

˜ Must include (at least) the following components:
Ô Com ? 
o ?
?o
Ô ?
o 



of?


Ô Comm 
?om

Ô Transaction processor (or, application processor,
or transaction manager)
å Software component found in each computer that
requests data
Ô Data processor or data manager
å Software component residing on each computer
that stores and retrieves data located at the site
å May be a centralized DBMS
|
 |  


  

|  



"
|   


|
  
  #  


 # |   |
˜ ^ll processing is done on single PU or host
computer (mainframe, midrange, or P)
˜ ^ll data are stored on host computer¶s local disk
˜ Processing cannot be done on end user¶s side of
the system
˜ Typical of most mainframe and midrange
computer DBMSs
˜ DBMS is located on the host computer, which is
accessed by dumb terminals connected to it
˜ ^lso typical of the first generation of single-user
microcomputer databases
 #  


 # |  
  

 #  


 # |  
|
˜ Multiple processes run on different
computers sharing a single data
repository
˜ MPSD scenario requires a network
file server running conventional
applications that are accessed
through a L^
˜ Many multi-user accounting
applications, running under a
personal computer network, fit

 #  


 # | 

 #  



 # |  

|
˜ Fully distributed database management system
with support for multiple data processors and
transaction processors at multiple sites
˜ lassified as either homogeneous or
heterogeneous
˜ Homogeneous DDBMSs
Ô Integrateonly one type of centralized DBMS
over a network

 #  



 # |  

| 
˜ Heterogeneous DDBMSs
Ô Integratedifferent types of centralized DBMSs
over a network
˜ Fully heterogeneous DDBMS
Ô Support different DBMSs that may even support
different data models (relational, hierarchical, or
network) running under different computer
systems, such as mainframes and
microcomputers
a

|
 
|  
  
 |
 |  

¦ 
 ! 

˜ ^llow end user to feel like database¶s only
user
˜ Features include:
Ô Distribution transparency
Ô Transaction transparency
Ô Failure transparency
Ô Performance transparency
Ô Heterogeneity transparency
 |
  ¦ 
 
˜ ^llows management of a physically dispersed
database as though it were a centralized
database
˜ Three levels of distribution transparency are
recognized:
Ô Fragmentation transparency
Ô Location transparency
Ô Local mapping transparency
 
¦ 
 ! 

!  "  

¦ 
 ¦ 
 
˜ ½nsures database transactions will
maintain distributed database¶s
integrity and consistency
 |
 $%

 
|
 ¦ 
 

˜ Distributed transaction
Ô an update or request data from several
different remote sites on a network
˜ Remote request
Ô Letsa single SQL statement access data to be
processed by a single remote database
processor
˜ Remote transaction
Ô ^ccesses data at a single remote site
|
 $%

 
|
 ¦ 
 

 
˜ Distributed transaction

Ô ^llows a transaction to reference

several different (local or remote)
DP sites

˜ Distributed request

Ô Lets a single SQL statement

reference data located at several
different local or remote DP sites
$ $%

$ ¦ 
 
|
 ¦ 
 
|
 $%

|
 
$%

|
 

˜ Multisite, multiple-process
operations are much more likely to
create data inconsistencies and
deadlocked transactions than are
single-site systems
¦   


`¦
 ¦# 
 

˜ Distributed databases make it possible for a
transaction to access data at several sites
˜ Final OMMIT must not be issued until all
sites have committed their parts of the
transaction
˜ Two-phase commit protocol requires each
individual DP¶s transaction log entry be
written before the database fragment is
actually updated
  
¦ 
 
&    
˜ Objective of query optimization
routine is to minimize total cost
associated with the execution of a
request
˜ osts associated with a request
are a function of the:
Ô ^ccess time (I/O) cost
Ô ommunication cost
Ô PU time cost
  ¦ 
 
&     
˜ Must provide distribution transparency as well as

transparency
˜ Replica transparency:
Ô DDBMS¶s ability to hide the existence of multiple
copies of data from the user
˜ Query optimization techniques:
Ô Manual or automatic
Ô Static or dynamic
Ô Statistically based or rule-based algorithms
 |
 |  

|

˜ Data fragmentation:

Ô How to partition the database into fragments

˜ Data replication:

Ô Which fragments to replicate

˜ Data allocation:

Ô Where to locate those fragments and replicas

 |  !    
˜ Breaks single object into two or more
segments or fragments

˜ ½ach fragment can be stored at any site over

a computer network

˜ Information about data fragmentation is

stored in the distributed data catalog (DD),
from which it is accessed by the TP to
process user requests
 |  !    
  

˜ Horizontal fragmentation:
Ô Division of a relation into subsets (fragments)
of tuples (rows)
˜ Vertical fragmentation:
Ô Division of a relation into attribute (column)
subsets
˜ Mixed fragmentation:
Ô ombination of horizontal and vertical
strategies
 
¦
$
¦ 
a  !    

¦
$¦ 
 
¦ !  
 ¦
"  

[  !  
¦ 


'!    


¦
$¦ 
 |  $   
˜ Storage of data copies at multiple sites served
by a computer network

˜ Fragment copies can be stored at several sites

to serve specific information requirements

Ô an enhance data availability and response time

Ô an help to reduce communication and total

query costs
¦ 


'!    



|  $   
$     

˜ Fully replicated database:
Ô Storesmultiple copies of 
database
fragment at multiple sites
Ô an be impractical due to amount of overhead
˜ Partially replicated database:
Ô Stores multiple copies of omdatabase
fragments at multiple sites
Ô Most DDBMSs are able to handle the partially
replicated database well
˜ Unreplicated database:
Ô Stores each database fragment at a single
site
Ô o duplicate database fragments
|    
˜ Deciding where to locate data
˜ ^llocation strategies:
Ô entralized data allocation
å ½ntire database is stored at one site
Ô Partitioned data allocation
å Database is divided into several disjointed parts
(fragments) and stored at several sites
Ô Replicated data allocation
å opies of one or more database fragments are
stored at several sites
˜ Data distribution over a computer network is
achieved through data partition, data
replication, or a combination of both
  (
)||

˜ Way in which computers interact to form a
system

˜ Features a  of resources, or a client, and

a  ov of resources, or a server

˜ an be used to implement a DBMS in which

the client is the TP and the server is the DP
 (  

˜ Less expensive than alternate minicomputer or
mainframe solutions
˜ ^llow end user to use microcomputer¶s GUI,
thereby improving functionality and simplicity
˜ More people with P skills than with mainframe
skills in the job market
˜ P is well established in the workplace
˜ umerous data analysis and query tools exist to
facilitate interaction with DBMSs available in the
P market
˜ onsiderable cost advantage to offloading
applications development from the mainframe to
powerful Ps
 (|
  

˜ reates a more complex environment, in which
different platforms (L^ s, operating systems,
and so on) are often difficult to manage
˜ ^n increase in the number of users and
processing sites often paves the way for security
problems
˜ Possible to spread data access to a much wider
circle of userså increases demand for people with
broad knowledge of computers and softwareå
increases burden of training and cost of
maintaining the environment
 )*)| +
¦
  


|
 |  


Local site independence
2 entral site independence
3 Failure independence
4 Location transparency
5 Fragmentation transparency
6 Replication transparency
7 Distributed query processing
8 Distributed transaction processing
9 Hardware independence

Operating system independence
 etwork independence
2 Database independence
  
˜ Distributed database stores logically related
data in two or more physically independent
sites connected via a computer network
˜ Database is divided into fragments
˜ Distributed databases require distributed
processing
˜ Main components of a DDBMS are the
transaction processor and the data processor
  
˜ urrent database systems can be classified by
extent to which they support processing and
data distribution
˜ DDBMS characteristics are best described as a
set of transparencies
˜ ^ transaction is formed by one or more
database requests
˜ ^ database can be replicated over several
different sites on a computer network
˜ lient/server architecture refers to the way in
which two computers interact over a computer
network to form a system