1.

TYPES OF DATABASES

Modern day computer systems are capable of systematically storing useful, relevant information
in a well-organized and easily accessible manner. A collection of such intelligible information is
called Database. Database is a dynamic entity which can be categorized into various types
depending on how the data is intended to be utilized, how it’s designed and how or where it’s
stored.

 Hierarchal Database:
A strict parent-child model to store data. You can visualize it as a family tree, a rigid one-to-
many format where a child record can’t correspond to more than one parent record. Here parent
record is the commonality at top with many entities branching down from it. In trade for this
highly structuredness and rigidity, it offers quicker query time and higher efficiency.

 Relational Database:

Structured Query Language is the most obvious and common application program for relational
databases. Data is essentially stored structurally on a table, with rows(tuple) and columns
(attributes). This database stores data in discrete tables, which can be joined together by fields
known as foreign keys. For example, there could be a table with customer details such as (name,
username, address, customer contact) and another table with purchase information (customer
contact, item purchased, price paid). Customer contact here is the key.

 NoSQL database/ non=relational database:

The broadly distributed un-structured data, lacking in structure, allows for a speedy process of
larger data sets due to the flexibility. This caters to many complex modern day applications’
expanding demands that are now building on cloud servers.
  Centralized Database:

The information(data) is stored at a single, centralized location and accessible securely from
remote location. The central computer at a local office in this case would work as the “Centre”,
running various authentication procedures; verification and validation of end-users to track and
record data usage. This allows us to fetch data from a distance meanwhile keeping data both
secure and efficient.
 Distributed Database:

The data is not in one place and is spread out to various physical sites; operating on multiple
devices, communicated within a singularly connected network and organization. This way even
if data is breached or compromised on one device others can still be up and running.

 Personal Database:

Single-user data. Generally small and simple to manage data; stored on personal computers to be
accessed by a small group of people, i.e. a small department of a larger infrastructure.

 Commercial Database:
Subject specific databases purposefully designed for end-users (clients) to access through
commercial links once they have paid for it.

 Open-Source Database:
A database designed with source code for free-public access. End users can freely make
changes to database without paying but with lesser/limited features as commercial
databases.

 End-user database:
Database designed to be accessed by a user of the product, who is not necessarily concerned with
how the data is managed or stored with the organization.
  Operational database.

Operational databases, unlike conventional databases, don’t require batch processing and allow
users to modify and manipulate data in real-time to handle daily business transactions.

 Cloud database:
Database on cloud is optimized for a “virtual environment”. This essentially means that even
though data is stored on physical location, the information is transmitted online. Database can be
on a private, public or a hybrid cloud platform. Users can either build themselves or pay a
service provider on per-user basis to store and administer their data for them. This allows for
high availability, flexibility, and scalability on demand.

 Document Database:
Stores data in JSON format rather than rows and columns.

 Time Series Database:

 Graph database:
A type of NoSQL database that is designed to identify and connect relationships between data
points.

 Key-Value Databases:

Every element is stored as a key-value pair consisting of an attribute name ("key") and a value.
It’s a NoSQL type of database. Most used for shopping carts.
 Autonomous Databases:
A much newer kind of cloud database that machine learning to administer tasks once
traditionally used by “Database Administrators”.
 2. FLAWS OR PROBLEMS IN THE FILE SYSTEM

The file-based system suffers from poor data integrity: inconsistent and redundant data. And
unlike DBMS it requires maintaining distinct files of every department. Hence, it takes up more
space and proves to be highly cost in-effective.
1. There is no control over the structure or organization of files.
2. Flat files system limits user access by not allowing multiple parties to access data
simultaneously causing concurrent access anomaly.
3. Complex data transactions are impossible.
4. Data is saved in different files, and each file can be in a significantly different format.
This makes it extremely hard to relate data and extract data from multiple files
causing “Data Isolation”.
5. Programmers used programming codes to deploy constraints. In the File system it’s
difficult to add new constraints, which results in data getting integrated, making it
incorrect and inconsistent.
6. It’s difficult to ensure atomicity in the File systems. Meaning if the data is entered is not
complete, the system will fail in the middle of transaction. This causes “Data
Atomicity” because it’s impossible to make changes later.
7. The data is easily modifiable and hence raises serious security concerns.