CPSC 304

Database Systems

n Instructors: Hazra Imran and Laks

V.S. Lakshmanan
n Textbook: Database Management
Systems by Raghu Ramakrishnan &
Johannes Gehrke 3rd Edition.

Most (not all) of my slides are based on the above textbook.

Laks V.S. Lakshmanan; Based on Ramakrishnan & Gehrke, DB Management Systems
and on George Tsiknis and Raymond Ngs slides.

Administrivia

Laks V.S. Lakshmanan; Based on Ramakrishnan & Gehrke, DB Management Systems

and on George Tsiknis and Raymond Ngs slides.

Two Sections
n Lecture notes separate for the two sections (Hazra) and

(Laks).
n However, tutorials are common: you may go to any one
tutorial.
n Lecture notes for both sections will be posted on piazza.

Unit 1:Intro

Your TAs
n . See

https://piazza.com/ubc.ca/winterterm22016/cpsc304/staff
for info. on your TAs.
nFastest way to get info./help about CPSC 304 our online

discussion forum Piazza.

Signup: https://piazza.com/ubc.ca/winterterm22016/cpsc304
Piazza Class Home: piazza.com/ubc.ca/winterterm22016/
cpsc304/home (go-to place for most things CPSC 304).
Course Home Page (Laks section):
http://www.cs.ubc.ca/~laks/cpsc304/304home.html.

Unit 1:Intro

A few Administrative Details

n Online Discussion of Course Material:
n We will use the Piazza system (www.piazza.com) for all

online discussion of course material. Piazza is a next

generation Question & Answer system specifically
designed to help you get answers to your questions fast.
The best way to get an answer to a question about 304 is
to post it on Piazza. Piazza allows both instructor, TAs
and students to answer questions, and makes it easy to
edit both questions and answers to improve them. To join
the Piazza group for 304, please go to
piazza.com/ubc.ca/winterterm22016/cpsc304 and follow
the instructions. Be sure to indicate your full name,
student ID, and the email address that you want to be
registered on Piazza (for this course) with.
Unit 1:Intro

More Admin. Details

n Please note that Piazza is a service that is hosted in the

United States. (It is a startup that originated at Stanford

University.) If you wish to preserve your anonymity on
Piazza, so that none of your personal information is
stored in the United States, you can create a new
anonymous email account and request Dylan (aka
Wengqiang Dong) ([email protected]) to use that email
account for registering you. (If you already use a gmail,
hotmail or other US hosted email account, this is
essentially moot.)
n Well be using Piazza also for posting announcements
and additional course related material.
n Register your clickers ASAP! (See course home page for

tips.)

Unit 1:Intro

Overview

Unit 1:Intro

Unit 1
Introduction
Read

Text: Chapter 1

Our focus
n tell you what the course is about
n tell you what to expect out of the course
Learning Goals
n Explain what a Database is
n Explain what a database management system (DBMS) is
n Explain benefits of using a DBMS
n Describe the basic structure of a DBMS
Laks V.S. Lakshmanan; Based on Ramakrishnan & Gehrke, DB Management Systems
and on George Tsiknis and Raymond Ngs slides.

Why database systems? 1/2

n One of the most successful industries.
n What powers your ATMs, or shopping portals, or travel

planning/reservations sites, ?
n Social Networking & Recommender Systems: DBMS
Underlying core powering facebook, myspace, flickr,
del.icio.us, Yahoo!Answers, rottentomatoes.com, .
n Data management encompasses all major applications of
CS.
n data management will remain important for ever:
Continued improvement/extensions of relational technology.
Developing technologies for managing data not managed (well):
e.g., text, multimedia, web data, graphs, matrices,
Unit 1:Intro

Why database systems? 2/2

Suppose we are building a system to store the information
pertaining to the university using, say Java.
What do we need to do?
nStore data on files (how should we organize them? )
nWrite programs to access and update the data
nMake sure that updates dont mess things up.
nProvide different views and access on the data to different
A lot of
users (registrar versus students)
work!
nNeed to write programs to deal with crashes.
nAnd do all the above from scratch for each application!
Alternatively:
nUse a DBMS to store and maintain the data.
Unit 1:Intro

10

What is a database?
n A database is an organized collection of related

data, usually (but not necessarily) stored on disk. It is

typically:
important, online
shared
secured
well-designed
variable size

n A DB typically models some real-world enterprise

Entities (e.g., students, courses, movies, songs, cameras, )

Relationships (e.g., Ford stars in Star Wars: The Force
Awakens;
Mary takes CPSC 304; Jack rented a Civic from Hertz.).

Unit 1:Intro

11

What is a DBMS?
n A Database Management System (DBMS) is a

bunch of software designed to store and manage

databases. It is used to:
define, modify, and query a database
provide access control
allow concurrent access
maintain integrity
provide backup and recovery from crashes
Provides a uniform data administration/maintenance
o provides centralized control and easy data management
o reduces application development time/effort.

etc.

n OK, We have to do much less now, but

Unit 1:Intro

12

How do we design such large systems?

We do it in stages and we use different models to represent
the info at each stage:
1.First we model the real world concepts using some high
level models, called conceptual models
2.Then we translate the world model into a database model
(called logical data model) that the database system
understands
3.We make this data model as efficient as possible
(optimization stage)
4.Design the code to query and maintain the data
5.Write the code for the application we want to develop.
In this course well look at all of these .
Unit 1:Intro

13

Conceptual Data Models

n Also called semantic data models
n High level (i.e.,abstract) models used for capturing the real world at

the initial stage.

n Well use the most popular: Entity-Relationship model (ER)

Unit 1:Intro

14

Logical Data Models

n A major purpose of a DB system -- provide an abstract

view of the data using a data model

n Data model : a collection of tools for describing
data , data relationships, semantics, constraints

n Well use the relational model : most widely used model

today.

Main concept: relation, basically a table with rows and columns.

Relations represent all the info. needed for the application.

n Other models:
object-oriented model
semi-structured data models
older models: network model and hierarchical model

Unit 1:Intro

15

An Example of a Relational Database

student
sid
92001200

name

address

Smith G.

1234 W. 12th Ave, Van.

major

gpa

CPSC

85

93001250

Chan J.

2556 Fraser St., Van.

MATH

80

94001150

Campeau J.

null

null

null

course
did

num

title

credits

CPSC

124

Principles of CS I

CPSC

126

Principles of CS II

MATH

100

Calculus I

etc.
Unit 1:Intro

16

Optimizing the Database

n Well learn how to create a design that

captures all the information we need to store

takes up less space
is easy to maintain
n Well look at different Normal Forms
(dont worry about it for now!)

Unit 1:Intro

17

Designing queries
n Well mainly use Structured Query Language (SQL):

Example: Find all the CPSC students who have at least

75% gpa:
Student(sid, name, address, major, gpa)
select name
from Student
where major=CPSC and
gpa >=75
n A declarative language the query processor figures out

how to answer the query efficiently

n Well also look at Relational Algebra and a high level and
intuitive logical language called Datalog:
Unit 1:Intro

18

Datalog Sampler
n Redo the previous query using Datalog:
n Find all the CPSC students who have at least 75% gpa:
n answer(Name) student(Sid, Name, Addr, Major, Gpa),

Major = CPSC, Gpa >= 75.

Constants

Variables

Student(sid, name, address, major, gpa)

Unit 1:Intro

19

Datalog Sampler
n Redo the previous query using Datalog:
n Find all the CPSC students who have at least 75% gpa:
n answer(Name) student(Sid, Name, Addr, Major, Gpa),

Major = CPSC, Gpa >= 75.

Dont Care Variables

Student(sid, name, address, major, gpa)

Unit 1:Intro

20

Datalog Sampler
n Redo the previous query using Datalog:
n Find all the CPSC students who have at least 75% gpa:
n answer(Name) student(_, Name, _, CPSC, Gpa),

Gpa >= 75.

Note: Datalog is essentially a pattern-driven query

language.
Student(sid, name, address, major, gpa)
Unit 1:Intro

21

An Intuitive Way to Think about Datalog

student
sid

name

address

major

gpa

CPSC

conditions
G >= 75

Unit 1:Intro

Values of (variable) N are returned as answer.

Var G is constrained.
CPSC and 75 are constants.
We dont care about values of sid and address.

22

General Remarks about

the Course

Unit 1:Intro

23

Create an Application
n In the term project you will use

a programming language to write code for an

application
and a standard interface to the database to access the
data for the application

n Well focus on Java & JDBC

you can also use C++ and ODBC
or HTML and PHP

Unit 1:Intro

24

Special Topics
In addition to core topics on relational databases,
n well discuss two special topics on Data Analytics
Data Warehousing & OLAP
Data Mining.

Unit 1:Intro

25

Course Learning Outcomes

At the end of the course you will be able to :
n describe how relational databases store and retrieve information
n develop a database that satisfies the needs of a small enterprise using
n
n
n
n

the principles of relational database design.

Express queries using formal database languages like relational
algebra and datalog.
Express queries using SQL
develop a complete data-centric application with transactions and user
interface using Java, JDBC (or equivalent) and a popular DBMS
explain key concepts and techniques used in:
Data Warehousing & OLAP.
Data Mining.

Unit 1:Intro

26

Course Activities
n Youll learn the most by doing:

assigned reading before each lecture

in-class clicker exercises (lectures)
homework assignments started in tutorials
project in groups of size 4 will go through the design
and implementation of a realistic database application
additional exercises posted on the Piazza from time to
time (for your own practice) not for grade.

Unit 1:Intro

27

Examinations and Grading

n Midterms (70 minutes each) :

Tuesday, January 31, in class.

Tuesday, March 7, in class.
n Final examination:
April, date to be scheduled by the university.
n Course grades will be calculated as follows:
12% for the tutorial work (homework assignments)
3% for the clicker questions
20% for the project
20% for midterm exams, and
45% for the final exam.
Plus 0.5 point for just completing a midterm survey. If there are
other surveys, you will hear from us.
Unit 1:Intro

28

Passing the course

n To pass the course, you must obtain an

overall passing mark and pass the project

and the final.

Unit 1:Intro

29

Course Resources
n Piazza home for course: piazza.com/ubc.ca/

winterterm22016/cpsc304/home

all course materials: notes, sample exams, solutions,

Online discussion forum: Piazza.
o Announcements, additional material.
o tutorials, project, homework, etc.

n CPSC 304 home page (Laks section):

http://www.cs.ubc.ca/~laks/cpsc304/304home.html
n Office Hours:
Laks: Face2Face hour: Tues 5:00-6:00 pm; Piazza hour: Thu
5:00-6:00 pm.
See course home page for details on TAs and tutorials/labs.

n Keep visiting course pages and piazza very often to catch

latest updates.

Unit 1:Intro

30

Back to Databases - DBMSs

n What else DBMSs are good for?
n Who are the main users of databases?
n What is the basic structure of a DBMS?

Unit 1:Intro

31

Levels of Abstraction
n Usually database systems allow

users to access the data at three

abstraction levels:

Physical level: shows how data are

actually stored
Logical (or conceptual) level: shows
data using the systems data model
( i.e., tables, etc.)
External (or View) level: describes
different part of the database to different
users
o
o

Unit 1:Intro

View 1 View 2 View 3

Logical Level
Physical Level

for convenience, security, or other

reasons
i.e. compare views of a bank database
for customer, teller, bank manager, &
database administrator
32

Schema and Instances

n Similar to types and variables in programming languages
n Schema the structure of the database
Physical schema: database structure at the physical level
Logical schema (or Conceptual schema or just Schema):
database structure at the logical level

n Instance the actual content of the database at a

particular point in time

Analogous to the value of a variable

n Physical Data Independence the ability to modify the

physical schema without changing the logical schema

Applications depend on the logical schema

n Logical Data Independence Provided by the views

Ability to change the logical schema without changing the
applications that depend on the view
Unit 1:Intro

33

Relational Example: University Database

n Logical schema:

student(sid: integer, name: string, address: string,

major: string, gpa:float)
course(dept: string, num:string, credits:integer)
enrolled(sid:integer, dept: string, num:string,
grade:integer)

n Physical schema will define:

The type of file that stores each relation

The type of records for each file
Any Indexes on the files, etc.

n External Schema (View):

Unit 1:Intro

course_info(dept: string, num:string, enrollment:integer)

34

An Instance of the University Database

student
sid
92001200

name

address

Smith G.

1234 W. 12th Ave, Van.

major

gpa

CPSC

85

93001250

Chan J.

2556 Fraser St., Van.

MATH

80

94001150

Campeau J.

null

null

null

course
did

num

title

credits

CPSC

124

Principles of CS I

CPSC

126

Principles of CS II

MATH

100

Calculus I

etc.
Unit 1:Intro

35

Database Primary Users

n End users (including senior mgmt)
n DB application programmers
n Webmasters
n DB Administrator
maintains conceptual, physical and external schemas
o

modifies the design, as reqts change; determines impact

handles security and authorization

handles backups and recoveries
checks database performance and performs necessary tuning
o indexes, reorganization, query analysis

Unit 1:Intro

36

Structure of a DBMS
Query Parsing,
and Optimization

n A typical DBMS has

a layered
architecture.
n This is one of
several possible
Transacarchitectures; each tion
system has its own
&
variations.
Lock

Query (Operator)
Evaluator
Files and Access Methods

Recov
ery

Buffer Management

Mana
ger

Disk Space Management

Manager

DB
Unit 1:Intro

37

Key Takeaways from this lecture

n Why DBMS?
n Physical, logical, and conceptual models
n Schema and instance
n (super) simple queries using SQL and Datalog

Unit 1:Intro

38

Summary
n DBMS provide many benefits in maintaining & querying

large datasets including

abstract representations of the data at different levels,
recovery from system crashes,
concurrent access, data integrity, and security
quick application development.

n Main goal of 304 is to:

give you the knowledge of how a DB application works
train you as a database programmer
provide you with the early steps to become a DBA

n 404, on the other hand:

looks deeper into a DBMS
trains you more as a DBA
Unit 1:Intro

39