Research Computing

Introduction to Databases
Presented by
Yun Shen (yshen16@bu.edu)
 Research Computing

Introduction
• What is Database
• Key Concepts
• Typical Applications and Demo
• Lastest Trends
 Research Computing

What is Database
• Three levels to view:
▫ Level 1: literal meaning – the place where data is stored
 Database = Data + Base, the actual storage of all the information that are
interested
▫ Level 2: Database Management System (DBMS)
 The software tool package that helps gatekeeper and manage data storage,
access and maintenances. It can be either in personal usage scope (MS Access,
SQLite) or enterprise level scope (Oracle, MySQL, MS SQL, etc).
▫ Level 3: Database Application
 All the possible applications built upon the data stored in databases (web site,
BI application, ERP etc).
 Research Computing

Examples at each level

• Level 1: data collection
 text files in certain format: such as many bioinformatic databases
 the actual data files of databases that stored through certain DBMS, i.e.
MySQL, SQL server, Oracle, Postgresql, etc.
• Level 2: Database Management (DBMS)
 SQL Server, Oracle, MySQL, SQLite, MS Access, etc.
• Level 3: Database Application
 Web/Mobile/Desktop standalone application - e-commerce, online banking,
online registration, etc.
 Research Computing

Examples at each level

• Level 1: data collection
 text files in certain format: such as many bioinformatic databases
 the actual data files of databases that stored through certain DBMS, i.e.
MySQL, SQL server, Oracle, Postgresql, etc.
• Level 2: Database Management System (DBMS)
 SQL Server, Oracle, MySQL, SQLite, MS Access, etc.
• Level 3: Database Application
 Web/Mobile/Desktop standalone application - e-commerce, online banking,
online registration, Wikipedia, etc.
 Research Computing

Database Types
• Flat Model
• Navigational databases
▫ Hierarchical (tree) database model
▫ Network/Graph model
• Relational Model
• Object model
• Document model
• Entity–attribute–value model
• Star schema
 Research Computing

Typical Database Application Architecture

app

DBMS DB
 Research Computing

Data File Structure

• Demo :
▫ Take a look at the following file directories:

 MySQL -- C:\ProgramData\MySQL\MySQL Server 8.0\Data\

 Access -- C:\ARCS_dbtutorial\db\access\

 Postgresql -- /project/scv/examples/db/tutorial/data/postgresql/testdb/
 Research Computing

Data File Structure - MySQL

 Research Computing

Data File Structure - Access

 Research Computing

Data File Structure - PostgreSQL

 Research Computing

ATTENTION

!!!
NO database files can be accessed directly,
but only through the database engine, called “DBMS”
 Research Computing

Typical Database Application Architecture

app

DBMS DB
Three Common Acronyms
• SQL – Structured Query Language

• CRUD – Create, Read, Update, Delete

• ACID – Atomicity, Concurrency, Integrity and Durability (transaction)
 Research Computing

Disadvantage of conventional flat file

▫ Redundancy - same data may store in many different copies
▫ Inconsistency – data regarding to same business entity may appear in
different forms, for example, state name, phone number, etc. This made it
hard to modify data and keep clean track of change; even loss of data
▫ Mixture of all data together – not clear in logical relationships between the
data columns, thus hard to understand and manage once the data
structure gets complex
▫ Hard to maintain and manage
▫ No concurrency support (only one can operate on the file)
 Research Computing

First Acronym - ACID

Atomicity – transactions are either all or none (commit/rollback)
Consistency – only valid data is saved
Isolation – transactions would not affect each other
Durability – written data will not be lost

Good example : bank transaction

Most of challenges of ACID compliance come from multiple users/concurrent

using of database
 Research Computing

How Databases solves the problem?

• Self-describing data collection of related records (meta data, data about
data) , detail explanation as below:
- Self-describing means:
▫ Database not just contains data, but also contains definition of the
structure of data, that can be considered ‘Meta data’; It includes many
related info : table column definition, index and key info, constraints, etc
▫ At database application level, database can also store other application
related meta data as well, it makes personalization and customization of
the application according to user profile much easier to handle. The typical
example could be the user preference for those common social media sites
or e-commerce sites, etc.
 Research Computing

Database Content
Typical Database • User Data: tables to store user data

• Meta data: keep the structure (schema) of

the data, including table name, column
name and type and contraints over the
column(s)
• User data
• Medadata
• Application metadata
• Application meta data: application specific
• Index and other
meta data regarding to user settings or
overhead
functions of the application

• Index and other overhead data: used for

improving performance and maintenance,
such as logs, track, security, etc.
 Research Computing

Terminology and Concept – Tables (Relations)

The very central concepts of relational databases are Tables (Relations),
Relationships.

Table (formally called ‘relation’) – is the building block of relational database. It

stores data in 2D, with its row reflects one instance of record (tuple), and each
of its column reflects one aspect of the attributes of all instances, column may
also be called ‘field’.

For example, A ‘student’ table may contains (student id, first name, last name,
grade, school name, home address, …), and each row may represent one
student’s information, and each column of the table represents one piece of
information of all students. And this is called a ‘relation’.
 Research Computing

Primary Key and Foreign Key

• Primary key: Unique Identifier made of one or more columns to
uniquely identify rows in a table. If the primary key contains more than
one column, it can be called ‘composite key’ as well.
• Foreign Key: is the primary key of another table, which is referenced in
the current table. It’s the key to establish the relationship between the
two tables, and through DBMS, to impose referential integrity.
 Research Computing

Surrogate Key
• Surrogate key is a unique column
• added to a relation to use as the primary key when lack of natural
column serves as primary key, or when composite key needs to be
replaced for various reasons.
• Surrogate key is usually in form of auto increment numeric value, and
of no meaning to the user, and thus cd often hidden in the table, or
form or other entity for the internal use.
• Surrogate keys are often used in the place of composite key to add more
flexibility to the table.
 Research Computing

Terminology and Concept – E-R model

E-R Model: Entity-Relationship data model is the common technique used
in database design. It captures the relationships between database tables
and represent them in a graphical way. The relationships between two
entities can be 1:1, 1:N, or M:N. And it is usually established through
‘foreign key’ constraint.

Examples:
1:1 Employee – Locker
1:N Customer – Order, Order – Order Detail
M:N Student – Course
 Research Computing

Sample E-R
diagram #1
 Research Computing

Sample E-R
diagram #2
 Research Computing

Sample E-R
diagram #3
 Research Computing

One more concept - Normalization

Wikipedia definition:
• Database normalization, or simply normalization, is the process of
restructuring a relational database in accordance with a series of so-
called normal forms in order to reduce data redundancy and improve data
integrity. It was first proposed by Edgar F. Codd as an integral part of
his relational model.
• Normalization entails organizing the columns (attributes)
and tables (relations) of a database to ensure that their dependencies are
properly enforced by database integrity constraints. It is accomplished by
applying some formal rules either by a process of synthesis (creating a new
database design) or decomposition (improving an existing database design).
 Research Computing

Unnormalized Form (UNF) –

same attributes can be contained in one row

Example: Students take courses

Id Name Courses Id Name Course1 Course2

Mathematics
1. Jack 1. Jack Mathematics Chemistry
Chemistry
or
2. Tim Chemistry 2. Tim Chemistry
Physics
3. Ana 3. Ana Physics Chemistry
Chemistry
 Research Computing

First Normal Form (1NF) –

each attributes can only have one single value in one row; no duplicated row; no row/column order

Example: Students take courses

Id Name Course
1. Jack Mathematics
4. Jack Chemistry
2. Tim Chemistry
3. Ana Physics
5. Ana Chemistry
 Research Computing

Second Normal Form (2NF) –

1NF + no partial dependency (non-key attributes may not depend on any of candidate keys)

Example: Students take courses

Id Name Course CourseID Department

1. Jack Mathematics M-1 Math
4. Jack Chemistry C-1 Chemistry
2. Tim Chemistry C-1 Chemistry
3. Ana Physics P-1 Physics
5. Ana Chemistry C-1 Chemistry
 Research Computing

Second Normal Form (2NF) – continue

1NF + no partial dependency (non-key attributes may depend on any of candidate keys)

(Course) = Func(CourseID) (Department) = Func(CourseID)

Id Name CourseID
Course
Course Department
1. Jack M-1 ID
4. Jack C-1 M-1 Mathematics Math
2. Tim C-1 C-1 Chemistry Chemistry
3. Ana P-1 P-1 Physics Physics
5. Ana C-1
 Research Computing

Third Normal Form (3NF) –

2NF + no transitive dependency (non-key attributes may not depend on each other)

Example: Students take courses

Id Name Course CourseID Department Building_no building

1. Jack Mathematics M-1 Math 31 D
4. Jack Chemistry C-1 Chemistry 22 A
2. Tim Chemistry C-1 Chemistry 22 A
3. Ana Physics P-1 Physics 18 S
5. Ana Chemistry C-1 Chemistry 22 A
 Research Computing

Third Normal Form (3NF) – continue

very similar to 2NF, but more strict; no functional dependency between non-key attributes at all.

Id Name CourseID Building_no Course

Course Department
ID
1. Jack M-1 31
M-1 Mathematics Math
4. Jack C-1 22
C-1 Chemistry Chemistry
2. Tim C-1 22
P-1 Physics Physics
3. Ana P-1 18
5. Ana C-1 22
Building_no Building
18 S
22 A
31 D
 Research Computing

Boyce–Codd Normal Form (BCNF) –

3NF + non dependency between all candidate keys

Example: 3NF, but not BCNF - Today's Court Bookings

Court Start Time End Time Rate Type

1 09:30 10:30 SAVER
1 11:00 12:00 SAVER
1 14:00 15:30 STANDARD
2 10:00 11:30 PREMIUM-B
2 11:30 13:30 PREMIUM-B
2 15:00 16:30 PREMIUM-A
 Research Computing

Boyce–Codd Normal Form (BCNF) –

3NF + non dependency between all candidate keys

Example: convert to BCNF - Today's Court Bookings

Rate Type Today's Bookings

Member Member
Rate Type Court Court Start Time End Time
Flag Flag
SAVER 1 Yes Yes 1 09:30 10:30
STANDARD 1 No Yes 1 11:00 12:00
PREMIUM-A 2 Yes No 1 14:00 15:30
PREMIUM-B 2 No No 2 10:00 11:30
No 2 11:30 13:30
Yes 2 15:00 16:30
 Research Computing

One More Example – database anormalies

My_Anomly_ex_bank_orig
AccountNo Name Address AccountBal AccountRate
100001 Lee, Yuan 11 Main Street 23994.58 4.50
100002 Lee, Yuan 11 Main Street 100.74 0.00
100002 Tian, Fu 12 Main Street 100.74 0.00
100003 Wang, Michele 12 Main Street 2500.8 0.00
100004 Dong, Yuan 14 Main Street 32003.98 4.50
100052 Yuan, Ben 16 Main Street 37.38 0.00
 Research Computing

One More Example

My_Anomly_ex_bank_orig
AccountNo Name Address AccountBal AccountRate
100001 Lee, Yuan 11 Main Street 23994.58 4.50
100002 Lee, Yuan 11 Main Street 100.74 0.00
100002 Tian, Fu 12 Main Street 100.74 0.00
100003 Wang, Michele 12 Main Street 2500.8 0.00
100004 Dong, Yuan 14 Main Street 32003.98 4.50
100052 Yuan, Ben 16 Main Street 37.38 0.00
 Research Computing

One More Example

My_Anomly_ex_bank_orig
AccountNo Name Address AccountBal AccountRate
100001 Lee, Yuan 11 Main Street 23994.58 4.50
100002 Lee, Yuan 11 Main Street 100.74 0.00
100002 Tian, Fu 12 Main Street 100.74 0.00
100003 Wang, Michele 12 Main Street 2500.8 0.00
100004 Dong, Yuan 14 Main Street 32003.98 4.50
100052 Yuan, Ben 16 Main Street 37.38 0.00
 Research Computing

One More Example

My_Anomly_ex_bank_orig
AccountNo Name Address AccountBal AccountRate
100001 Lee, Yuan 11 Main Street 23994.58 4.50
100002 Lee, Yuan 11 Main Street 100.74 0.00
100002 Tian, Fu 12 Main Street 100.74 0.00
100003 Wang, Michele 12 Main Street 2500.8 0.00
100004 Dong, Yuan 14 Main Street 32003.98 4.50
100052 Yuan, Ben 16 Main Street 37.38 0.00
 Research Computing

One More Example

My_Anomly_ex_bank_orig
AccountNo Name Address AccountBal AccountRate
100001 Lee, Yuan 11 Main Street 23994.58 4.50
100002 Lee, Yuan 11 Main Street 100.74 0.00
100002 Tian, Fu 12 Main Street 100.74 0.00
100003 Wang, Michele 12 Main Street 2500.8 0.00
100004 Dong, Yuan 14 Main Street 32003.98 4.50
100052 Yuan, Ben 16 Main Street 37.38 0.00
Gu, Zhen (???)
 Research Computing

Advantages of Normalization
BCNF+ normalization can eliminate all anomalies :
▫ No Redundancy
▫ No Inconsistency – all changes can only be made at the same place and
keep consistent (because of the key constraints), in DB terminology – get
away with all update anormaly.
▫ Normalization is the process of decomposition, so all the business concepts
can be modeled with clear logical relationships
▫ The entire database system remains consistent over time as the database
grows with least redundancy and much durability.
▫ Strong support to be ACID compliant
 Research Computing

Advantages of Normalization
• No(less) data redundancy – means easy management, less storage, etc.
• No headache caused by data operation anomalies. Good for data
integrity and consistency.
 Research Computing

Disadvantages of Normalization
• Take effort
• May increase complexity in data structure
• Data retrieving efficiency may be discounted due to the need of join of
multiple tables; So may not be proper in read-intensive data
applications
• Sometimes the constraints may be too strict to be flexible to make some
customized change needed.
 Research Computing

Disadvantages of Normalization
• Hard to deal with complex data structures such as class, objects, rows in
a field.
• Query for comprehensive information can be costly.[6]
• Due to fixed predesigned structure, it is not flexible in terms of
restructure of data
 Research Computing

Modern applications
• Today companies like Google, Amazon and Facebook deal with loads of
data and storing that data in an efficient manner is always a big task.
They use NoSQL database which is based on the principles of
unnormalized relational model to deal with storage issue.Some of the
examples of NoSQL databases are MongoDB, Apache
Cassandra and Redis. These databases are more scalable and easy to
query with as they do not involve expensive operations like JOIN.
 Research Computing

Denormalization
• Normalization and denormalization both have advantages and
disavantages. The best practice is always a trade off between the two.
• Denormalization will increase the risk of loss of data integrity and the
size of storage, but may gain the simplicity and intuitivity of presenting
data.
 Research Computing

Denormalization - Example
• Customer (CustomerID, Name, Address, Zip, City, State)

CustomerID [PK] Zip [PK] CustomerID [PK]

Name City Name
Address State Address
Zip [FK] Zip
City
State
 Research Computing

Denormalization - Example
• This is the normalized table design

CustomerID Name Address Zip Zip City State

02132 Boston MA
101 John 111 Main St 02132
102 Adam 17 Willow St 02451 02451 Waltham MA

103 Grace 333 Burke St 02132

 Research Computing

Denormalization - Example
• This is the denormalized table design
CustomID Name Address Zip City State

101 John 111 Main St 02132 Boston MA

102 Adam 17 Willow St 02451 Waltham MA
103 Grace 333 Burke St 02132 Boston MA
 Research Computing

Database Operation/Administration
• CRUD (Create/Read/Update/Delete) – four basic operations
• All through SQL (Structured Query Language)
▫ Sublanguage
▫ DDL (Data definition Language)
▫ DQL (Data Query Language)
▫ DML (Data Manipulate Language)
▫ DCL (Data Control Language)
▫ Scope of SQL: Query (select), Manipulate(Insert/update/delete),
Definition(Create/Modify tables/columns) , Access Control (permission)
 Research Computing

How to Learn SQL

• Same approach as many other languages, get a language reference first
▫ https://www.w3schools.com/sql/sql_syntax.asp (Standard SQL, simple)
▫ https://docs.oracle.com/cd/B28359_01/server.111/b28286/expressions.ht
m#SQLRF004 (complete, but more complicated, and may add its own
flavor)
• Start from basics (DDL, DML, DQL)
• Extend to more complicate elements (Subquery/Join/Function/Store
Procedure/Index/Programming)
 Research Computing

Demos and Exercises

• Microsoft Access Templates
• E-R schema in Microsoft Access
• Anomaly
• Data import/export via Microsoft Access (may do some real life demo
using actual data file from audience)
• Data import/export via MySQL
 Research Computing

Other Stuffs not covered by this tutorial

• Database Administration: concurrency, security, backup, recovery, and
many more
• Database Performance tuning: indexing, server configurations
• Database programming
• Database Technology Trend: BigData challenge, Data Warehouse, BI
system, NoSQL, Cloud, Hadoop/Spark, etc.
 Research Computing

BI Systems
• Reporting System
• Data Mining (Has big overlap with today’s ML/AI trend)
• Data Warehouse/Data Mart
• ETL (Extract/Transform/Load)
 Research Computing

Big Data
• 4Vs:
▫ Volume – how big in storage is need?
▫ Variety – how diverse is the data ?
▫ Veracity – can data be verified/trusted?
▫ Velocity – how fast is the data being generated?
 Research Computing

Big Data Technologies

• Predictive analytics: to discover, evaluate, optimize, and deploy predictive models by analyzing big data sources to improve business
performance or mitigate risk.
• NoSQL databases: key-value, document, and graph databases.
• Search and knowledge discovery: tools and technologies to support self-service extraction of information and new insights from large
repositories of unstructured and structured data that resides in multiple sources such as file systems, databases, streams, APIs, and
other platforms and applications.
• Stream analytics: software that can filter, aggregate, enrich, and analyze a high throughput of data from multiple disparate live data
sources and in any data format.
• In-memory data fabric: provides low-latency access and processing of large quantities of data by distributing data across the dynamic
random access memory (DRAM), Flash, or SSD of a distributed computer system.
• Distributed file stores: a computer network where data is stored on more than one node, often in a replicated fashion, for redundancy
and performance.
• Data virtualization: a technology that delivers information from various data sources, including big data sources such as Hadoop and
distributed data stores in real-time and near-real time.
• Data integration: tools for data orchestration across solutions such as Amazon Elastic MapReduce (EMR), Apache Hive, Apache Pig,
Apache Spark, MapReduce, Couchbase, Hadoop, and MongoDB.
• Data preparation: software that eases the burden of sourcing, shaping, cleansing, and sharing diverse and messy data sets to
accelerate data’s usefulness for analytics.
• Data quality: products that conduct data cleansing and enrichment on large, high-velocity data sets, using parallel operations on
distributed data stores and databases.
 Research Computing

Summary of Training
• List important points from each lesson.
• Provide resources for more information on subject.
▫ List resources on this slide.
▫ Provide handouts with additional resource material.