Modern Education Society’s

College of Engineering, Pune

19, Late Prin. V. K. Joag Path, Wadia College Campus, Pune-411001
Affiliated to Savitribai Phule Pune University & Approved by AICTE, New Delhi.

Prof. S. B. Shinde
Asst Professor, MESCOE Pune
 Introduction
 In the computing system (web and business applications),
there are enormous data that comes out every day from the
web. A large section of these data is handled by Relational
database management systems (RDBMS).

 Classical relation database follow the ACID Rules:

 A database transaction must be atomic, consistent, isolated and
durable.

Prof. S. B. Shinde
 Introduction

Prof. S. B. Shinde
 Distributed Systems
 A distributed system consists of multiple computers and
software components that communicate through a computer
network (a local network or by a wide area network).

 A distributed system can consist of any number of possible

configurations, such as mainframes, workstations, personal
computers, and so on.

 The computers interact with each other and share the

resources of the system to achieve a common goal.
Prof. S. B. Shinde
 Distributed Computing
Advantages: Disadvantages:
 Reliability (fault tolerance)  Troubleshooting
 Scalability  Less Software Support
 Sharing of Resources  Network infrastructure
 Flexibility  Security
 Speed
 Open System
 Performance

Prof. S. B. Shinde
 What is NoSQL?
 Stands for Not Only SQL.

 NoSQL is a non-relational database management systems.

 NoSQL database were developed in response to a rise in the

volume of data stored about users, objects and products, the
frequency in which this data is accessed, and performance and
processing needs.

 It is designed for distributed data stores where very large scale

of data storing needs (for example Google or Facebook which
collects TB of data every day for their users).

 These type of data storing may not require fixed schema, avoid
join operations and typically scale horizontally.
Prof. S. B. Shinde
 Why NoSQL ?
 In today’s time data is becoming easier to access and capture
through third parties such as Facebook, Google+ and others.

 Personal user information, social graphs, geo location data,

user-generated content and machine logging data are just a few
examples where the data has been increasing exponentially.

 To avail the above service properly, it is required to process

huge amount of data. Which SQL databases were never
designed.

 Instead of using structured tables to store multiple related

attributes in a row, NoSQL databases use the concept of a
key/value store.
Prof. S. B. Shinde
 Why NoSQL ?

Unstructured and Semi-

structured data

Structured data

Prof. S. B. Shinde
 RDBMS vs NoSQL
RDBMS NoSQL
 Structured and organized data  Stands for Not Only SQL
 Structured query language (SQL)  No declarative query language
 Data and its relationships are  No predefined schema
stored in separate tables  Key-Value pair storage, Column
 DDL,DML Store, Document Store, Graph
 Tight Consistency databases
 ACID Transaction  Eventual consistency rather ACID
property
 Unstructured and unpredictable data
 CAP Theorem
 Prioritizes high performance, high
availability and scalability
 BASE Transaction

Prof. S. B. Shinde
Prof. S. B. Shinde
 Brief History of NoSQL
 The term NoSQL was coined by Carlo Strozzi in the year
1998. He used this term to name his Open Source, Light
Weight, DataBase which did not have an SQL interface.
 Three Eras of Databases

 RDBMS for transactions, Data Warehouse for analytics

and NoSQL for …?

Prof. S. B. Shinde
 Before NoSQL

Prof. S. B. Shinde
 After NoSQL

Prof. S. B. Shinde
 Type of NoSQL
 Document Oriented Databases:
 Document oriented database stores data in the form of documents.
 A collection of documents.
 A document can be in a JSON, BSON, XML, YAML, etc format.
 Data in this model is stored inside documents.
 A document is a key value collection where the key allows access to its
value.
 Documents are stored into collections in order to group different kinds of
data.
Relational model Document model

Tables Collections

Rows Documents

Columns Key/value pairs

Joins not available

 Example: MongoDB, Elasticsearch, Couchbase Server, CouchDB, RethinkDB,

Terrastore, MarkLogic Server etc.
Prof. S. B. Shinde
 Type of NoSQL
 Document Oriented Databases:

Prof. S. B. Shinde
 Type of NoSQL
 Column Oriented Databases ::
 Column-oriented databases primarily work on columns
and every column is treated individually.
 Column stores can improve the performance of queries as
it can access specific column data.
 High performance on aggregation queries (e.g. COUNT,
SUM, AVG, MIN, MAX).
 Works on data warehouses and business intelligence,
customer relationship management (CRM), Library card
catalogs etc.

 Example: Hadoop/Hbase, Cassandra, Amazon SimpleDB,

HPCC, Cloudera etc.

Prof. S. B. Shinde
 Type of NoSQL
 Column Oriented Databases

Prof. S. B. Shinde
 Type of NoSQL
 Key-Value Databases:
 In key-value database each item in the database is stored as
an attribute name (or “key”), together with its value.
 Key-value stores are most basic types of NoSQL databases.
 Designed to handle huge amounts of data.
 In the key-value storage, database stores data as hash table
where each key is unique and the value can be string, JSON,
BLOB (basic large object) etc.
 Key-Value stores follows the 'Availability' and 'Partition'
aspects of CAP theorem.
 Key-Value stores can be used as collections, dictionaries,
associative arrays etc.
 Example: Redis, Riak, Azure Table Storage, DynamoDB,
Berkeley DB, LevelDB, FoundationDB etc.
Prof. S. B. Shinde
 Type of NoSQL
 Key-Value Databases:

Prof. S. B. Shinde
 Type of NoSQL

 Graph database:
 A graph database uses graph structures with nodes, edges,
and properties to represent and store data in database.
 A graph databases is faster for associative data sets and
hence it’s gaining popularity these days.
 Graph stores are used to store information about networks,
such as social connections.
 Each node represents an entity (such as a student or
business) and each edge represents a connection or
relationship between two node

 Example: HyperGraphDB, GraphBase Neo4J, WhiteDB,

Infinite Graph, BrightstarDB etc.

Prof. S. B. Shinde
 Type of NoSQL

 Graph database:

Prof. S. B. Shinde
 SQL v/s NoSQL
SQL Databases NoSQL Databases
Types One type (SQL database) with Many different types including
minor variations key-value stores, document
databases, wide-column stores,
and graph databases.
Development Developed in 1970s to deal with Developed in 2000s to deal with
History first wave of data storage limitations of SQL databases,
applications. particularly concerning scale,
replication and unstructured data
storage.
Examples MySQL, Postgres, Oracle Database MongoDB, Cassandra, HBase,
Neo4j
Data Specific language using Select, Through object-oriented APIs
Manipulation Insert, and Update statements,
Consistency Can be configured for strong Depends on product. Some
consistency provide strong consistency (e.g.,
MongoDB) whereas others offer
eventual consistency (e.g.,
Cassandra)

Prof. S. B. Shinde
 SQL v/s NoSQL
SQL Databases NoSQL Databases
Scaling Vertically, meaning a single server Horizontally, meaning that to add
must be made increasingly powerful capacity, a database administrator
in order to deal with increased can simply add more commodity
demand. servers or cloud instances.
Development Mix of open-source (e.g., Postgres, Open-source
Model MySQL) and closed source (e.g.,
Oracle Database)
Supports Yes, updates can be configured to In certain circumstances and at
Transactions complete entirely or not at all certain levels (e.g., document level
vs. database level)
Data Storage Individual records are stored as rows Varies based on database type.
Model in tables, with each column storing
a specific piece of data about that
record much like a spreadsheet.
Schemas Structure and data types are fixed in Typically dynamic. Records can
advance. To store information about add new information on the fly,
a new data item, the entire database and unlike SQL table rows,
must be altered, during which time dissimilar data can be stored
the database must be taken offline. together as necessary.
Prof. S. B. Shinde
 CAP Theorem (Brewer’s Theorem)
CAP Theorem states that there are three basic
requirements which exist in a special relation when
designing applications for a distributed architecture.

 Consistency:

 Availability:

 Partition tolerance:

Prof. S. B. Shinde
 CAP Theorem
Consistency:
 Consistency: (all nodes see the same data at the same
time)

 Consistency - the data in the database remains

consistent after the execution of an operation.

 For example after an update operation all clients see

the same data.

Prof. S. B. Shinde
 CAP Theorem
 Availability:

 Availability: (a guarantee that every request receives a

response about whether it was successful or failed)

 Availability - the system is always on (service

guarantee availability), no downtime.

Prof. S. B. Shinde
 CAP Theorem
 Partition Tolerance
 Partition Tolerance (the system continues to operate
despite arbitrary message loss or failure of part of the
system)

 Partition Tolerance - the system continues to function

even the communication among the servers is unreliable,
i.e. the servers may be partitioned into multiple groups
that cannot communicate with one another.

Prof. S. B. Shinde
 CAP Theorem
 Theoretically it is impossible to fulfill all 3 requirements.
 A distributed system can support only 2 out of the 3
characteristics:
 CA - Single site cluster, therefore all nodes are
always in contact. When a partition occurs, the
system blocks.
 CP - Some data may not be accessible, but the
rest is still consistent/accurate.
 AP - System is still available under partitioning,
but some of the data returned may be inaccurate.

Prof. S. B. Shinde
 CAP Theorem

Prof. S. B. Shinde
 The BASE System
 The BASE acronym was defined by Eric Brewer, who
is also known for formulating the CAP theorem.

 The CAP theorem states that a distributed computer

system cannot guarantee all of the following three
properties at the same time:
 Consistency:
 Availability:
 Partition tolerance:

 A BASE system gives up on consistency.

Prof. S. B. Shinde
 The BASE System Contd..
 Basically Available indicates that the system does guarantee
availability, in terms of the CAP theorem.

 Soft state indicates that the state of the system may change
over time, even without input. This is because of the eventual
consistency model.

 Eventual consistency indicates that the system will become

consistent over time, given that the system doesn't receive
input during that time.
ACID BASE
Atomic Basically Available
Consistency Soft state
Isolation Eventual consistency
Durable
Prof. S. B. Shinde
 NoSQL pros/cons

Advantages: Disadvantages:
 High scalability  Maturity
 Dynamic Schemas  Enterprise Support
 Replication  Transaction Support
 Auto-sharding  Expertise (Highly Skilled
 Integrated Caching Programmers)
 Distributed Computing
 Low Cost
 No complicated Relationships

Prof. S. B. Shinde
 Advantages of NoSQL
 Scalability: NoSQL database can be scaled up easily and with minimum
effort and hence it’s well suited for today’s every increasing database need
(bit data). NoSQL database have scalable architecture, so it can efficiently
manage data and can scale up to many machines instead of costly machines
that are required while scaling using of SQL DBMS.
 With dynamic schema, if we want to change the length of column, or add
new column we don’t need to change whole table data instead the new data
will be stored with the new structure without affecting the previous data
/structure. In NoSQL databases we can insertion data without a predefined
schema.
 Replication provides redundancy and increases data availability. With
multiple copies of data on different database servers, replication protects a
database from the loss of a single server.
 To use replication with sharding, deploy each shard as a replica set.
 Sharding is the process of storing data records across multiple machines and
is MongoDB’s approach to meeting the demands of data growth.
 Many NoSQL database have integrated caching mechanism, hence
frequently used data are stored in system memory as much as possible and
discarding the need for a separate caching layer.
 Disadvantages of NoSQL
 Maturity – NoSQL database are new and emerging technologies. Since its
under heavy development bugs, new features, keep on arising.

 Enterprise Support – If system fails company must be able to get timely

support. In case of NoSQL, there are very few companies which know the
technology and hence can be a deciding factor before using NOSQL.

 Transaction Support – NOSQL doesn’t support SQL transaction features

and hence for financial application SQL are still one the best in industry.

 Expertise (Highly Skilled Programmers) – There are millions of people

around the world who knows RDBMS while very few people are aware of
such technology hence getting a NOSQL programmer can be difficult.

Prof. S. B. Shinde
Thank
You