Transcript Database Management System

Database Technology:
Achievements and Opportunities
Dr M Saraee
Dept. of Electrical and Computer Engineering
Isfahan University of Technology
1
Outline of Discussion
Why do we need databases
Database Systems: Past, Present and Future
Overview of New Approaches
2
File-based Systems
A collection of application programs that perform services for the end users
(e.g. reports). Each program defines and manages its own data.
3
Database / DBMS
Database: A shared collection of logically related data (and a
description of this data), designed to meet the information needs
of an organisation.
DBMS: A software system that enables users to define, create,
and maintain the database and which provides controlled access
to this database.
4
Database Management System (DBMS)
5
Why do we need database management systems?
• A Database Management System (DBMS) is a tool that allows to
store, modify and query data.
However, we can store,
modify and query data in a
text file!
What can a DBMS do that
we can’t do with the text file
solution.
File-based solution to manage data, stick it all in a
text file!
6
Enforcing Constraints
• With the File-based Systems there is no way to enforce integrity
constraints on the data. In other words people can put bad data into the
text file.
• In contrast, a DBMS allows us to enforce all kinds of constraints. This
really helps (but does not guarantee) that our data is correct.
A typo gives Roberta Wickham a GPA of 44.00
7
Scalability
• The File-based Systems old method, might work for small datasets. What
happens when we have big datasets…
• Most real world datasets are so large that we can only have a small
fraction of them in main memory at any time, the rest has to stay on disk.
• Even if we had lots of main memory, with 32 bit addressing we can only
refer to 4GB of data!
8
Query Expressiveness
• The File-based Systems would allow us to search for keywords or certain
numbers (slowly).
• With a DBMS we can search with much more expressive queries. For
example I can ask.. “Find all students whose GPA is greater than 2.5, and
who don’t own a phone” or “what is the average GPA of the students”
9
Query Expressiveness II
• We could write some program that might allow more expressive queries
on my text file, but it would tied into the structure of our data and the
operating system etc..
• With a DBMS we are completely isolated from the physical structure of
our data. If we change the structure of our data (by adding a field, for
example) or moving from a PC to a Mac, nothing changes at the front
end!
10
Different Views
• The File-based System only allows one view of the data.
•With a DBMS we can arrange for different people to have different views
of the data. For example, I can see everything, a student can see only
his/her data, the TA can see…
11
Concurrency
• Suppose we leave my text file on UNIX account, and weI log in and
begin to modify it at the same time our TA is modifying it!
• A DBMS will automatically make sure that this kind of thing cannot
happen.
12
Security
• Suppose I leave my text file on UNIX account, and a student hacks in
and changes their grades…
• A DBMS will allow multiple levels of security.
13
Crash Recovery
• Suppose I am editing my text file and the system crashes!
• A DBMS is able to guarantee 100% recovery from system crashes.
14
Roles in the Database Environment
Data Administrator (DA)
Database Administrator (DBA)
Database Designers (Logical and Physical)
Application Programmers
End Users (native and sophisticated)
15
Database Systems: Achievements
Relational Database Systems
Transaction Management
Distributed Relational Database Systems
16
Database Systems: Current Research
Object-Oriented DBMS
– Object-Relational DBMS
Support for New Data Types
– temporal data
– spatial data
Transaction Processing
17
Technology vs Functionality Matrix
Functionality
Active
Temporal Multimedia
Geographical
SpatioTemporal
Technology
Relational
Object-Oriented
Interoperable
Add functionality to
existing technology
Combine functionality
from existing technology
18
Database Systems: New Applications
Earth Observation Data
Electronic Commerce
Health-Care Information Systems
Digital Publishing
Collaborative Design
19
Earth Observation Data
 The Earth Observing System (EOS) will gather data about
the atmosphere, oceans and land.
NASA satellites have been launched from 1998 onwards
Data transmission is estimated to be 1/3 petabyte per year
(1PB = 109 MB=1015 bytes)
EOS Data and Information System (EOSDIS) will support
on-line access and maintenance of EOS data.
20
Electronic Commerce
Needs support for browsing of catalogs and electronic
purchasing of goods
Large number of consumers and suppliers
Database challenges
– heterogeneous distributed information sources
– distributed authentication and fund transfers
21
Health-Care Information Systems
Improve quality and quantity of health care
Needs support for medical records across hospitals,
medical offices and insurance offices and across
countries
Collection of historical information about a patient
Database challenges
– integration of heterogeneous legacy information
– access control to preserve confidentiality of medical records
– intelligent interfaces to be used by health-care professionals
22
Distributed Databases
A distributed database system consists of a collection of
sites, connected together via some kind of communications
network, in which
– each site is a database system in its own right, but
– the sites have agreed to work together (if necessary), so that a user at any
site can access data anywhere in the network exactly as if the data was all
stored at the user's own site
Fundamental principle of distributed databases
a distributed system should look exactly like a nondistributed system
23
Active Databases
Relational Databases are passive
– They execute queries or database operations only when explicitly
requested to do so by a user or an application program.
Active Databases are reactive
– They monitor conditions defined on states of the database, and then,
once these conditions occur, they invoke specified actions.
24
Active Database Components
The rule base
WHEN event IF condition THEN action
The database
The inference mechanism (rule manager)
The user interface
25
Information Systems and Time
event
UoD
IS i-1
T-1
i-1
event
UoD
IS i
T
i
UoD
IS
i+1
i+1
T+1
time axis
26
Why Time Modelling?
Historical queries about past status
Trend analysis
Representation of retroactive or proactive changes
Version control and design management
Scheduling and planning requirements
27
Time and IS Modelling
What was the salary of John Smith last year
What was the business policy for the Product Ordering
process under the previous managing director
What is the history of the Product Promotion strategy?
28
Conclusions
Knowledge Representation and Manipulation within a
computer is the single most important challenge for IS
Databases are the main focus of this work and they are the
most widely available technology outside operating systems
The next few years will see another rapid expansion of this
area
29