INTERNET OF THINGS

(DESIGN PRINCIPLES FOR

WEB CONNECTIVITY)

By: Dr. Raj Kamal

Copyright © 2017 McGraw Hill Education, All Rights Reserved.

PROPRIETARY MATERIAL © 2017 The McGraw Hill Education, Inc. All rights reserved. No part of this PowerPoint slide may be displayed, reproduced or distributed in any form or
by any means, without the prior written permission of the publisher, or used beyond the limited distribution to teachers and educators permitted by McGraw Hill for their individual course
preparation. If you are a student using this PowerPoint slide, you are using it without permission.
Chapter 5

Data Acquiring, Organising,

Processing and Analytics
Application

 Application software
 A collection of software components for an application
 Enables a user perform a group of coordinated activities, functions and tasks

Examples
 Streetlights control and monitoring Application
 Application for tracking and inventory control using RFID device as tags on the parcels
 Application for enabling a user withdrawal of cash using an ATM (Automatic Teller
Machine)
 Automobile maintenance service
 An application for The waste container management
 Service
 Enables use by interested application or clients
 A mechanism which enables the provisioning of access to one or more capabilities
 An Interface provisioning for the the access to capabilities
 The access consistent with service description for the constraints and policies
 Enable application registration or subscription after a service level agreement

Examples:
 Security provisioning, bank transactions
 Device IDs management or tracking
 Devices location detection and tracking service
 Service Oriented Architecture (SOA)

 A software architecture model which consists of Services, Messages,

Operations and Processes

SOA Components
 Distributed over a network or Internet in a high level business entity
 Used for developing for new business applications
 Used for developing an Applications Integration Architecture in an
Enterprise
 Process
 A composition of a group of structured activities or tasks or services that
leads to a particular goal
Examples:
 Streetlights control process
 The purchase process of airline ticket
 Specification of activities with relevance rules based on data or inputs in
the process

Process Matrix
 A multi-element entity
 Each element relating to a set of data or inputs to an activity (or subset of
activities).
 Passive Devices Data
Passive Device
 A contactless card have or may not have an associated microcontroller,
Memory and transceiver
 Level or barcode not have an associated microcontroller

Passive Devices Data

 Data generates at the device or system
 Data following the result of interactions
 A passive device does not have own power source
 An external source makes that device to send data
 RFID or an ATM debit card
 Active Devices Data
 Data generates at the device or system
 Data following the result of interactions
 Active device own power source
 Examples: Active RFIDs, streetlight sensor, wireless sensor node.
 Active device associated microcontroller, memory and transceiver.
 (An) Event Data from Device

 Generating data on an event once only

 Detection of the traffic or on dark ambient conditions, that signals an event.
Then event communicates a need for the lighting up a group of streetlights.
 A system consisting of security cameras generating data on an event of
security breach or on intrusion detection
 A waste container with associate circuit generating data in the event of
getting it 90% or above filled up.
 The components and devices in an automobile generate data of their
performance and functioning, and communicate to Internet as and when the
automobile reaches near a Wi-Fi access point
 Data Acquisition

 Data acquisition means acquiring the data from IOT/M2M devices

 The data communicate after the interactions with a Data acquisition system
(Application)

 The Application interacts and communicates with number of devices for acquiring
the needed data

 The devices send data on demand or at the programmed intervals

 Data of devices communicate using the network, transport and security layers.
 Data Validation

 Data needs validation checks

 Data validation software do the validation checks
 Validation software applies logic, rules and semantic annotations.
 Must as the Applications/services/Processes depend on valid data
 Then only the analytics, predictions/ prescriptions/ diagnosis/decisions
acceptable
 Data Storage

Three Categories
 On-line or real time or streaming data needing the processing, and only the results of
processing and analysis need storage
 Data called once, only the results of processing at a later time and of analysis store,
 Data needing repeated calls store for reference or audit in future.,

Features
 Objects in a Data Store model using Classes which the database schemas define.
 Data Store may be distributed over multiple nodes, (Apache Cassandra is example of
distributed Data Store.)
 Ways to Organise Data
(Collect + Assemble +Manage) for analytics and Acquire, Organize and Analyse
Intelligence) Manage applications, services, enterprise and
Cloud Infrastructure Device business processes and Intelligence

Informix Time Data Centre Server Mgmt Identity Mgmt &

series service Mgmt. access Data Centre Server Mgmt
Assemble events from the IoT into logic flows in Mgmt. Mgmt.
Device
Register and Analytics In-memory
connect and Intelligence Analytics
and Business
Big Big
Database Firewall

Intelligence
Data Store Data Store

Internet Firewall

Database Firewall
Device
Internet Firewall

Collect from Key Value

NoSQL Access Data Store
things.
Mgmt
Device Database
Manage time- Relational Time Identity Mgmt RDBMS Complex
series view of Series service
data Enterprise Applicatio
Device ns
Mgmt. Integration
Spatial Storage and Integration
Manage connections Real Time Analytics and SoA
Data Routing &
and Subscriptions Mgmt. Analysis

SoA: Service Oriented Architecture

Consumer/Business
(a) (b)

Connect + Collect + Assemble +Manage IBM conceptual framework (b) Manage,

13
Acquire, Organise and Analyse Oracle’s framework in the IoT Architecture
 Query and Query Processing

 An Application/Service/Process seeking a specific data set from a

database/relational database
 Query processing means using a process
 Getting the results of the query made from a DB

Structured Query Language (SQL)

 A language for data access control, schema creation and modifications.
 Language for managing the RDBMS.
 Language for data definition, data manipulation and data control
instructions
 Data Transactions Functions

Transaction
 A collection of operations that form a single logical unit of database

 A collection of operations that form a single logical unit of database.

 Consist of all operations executed between begin and end of transaction

 Operations such as connect, insertion, append, deletion or modification in

a unit of database
 Types of Transaction Processing

Interactive Transactions Real time transaction Batch Transactions

Processing processing Processing
• Transactions which involve Transactions processing at the • Transactions processes in
continual exchange of same time as the data arrives from batches and in non interactive
information between the the data sources and Data Stores way
computer and a user • When one set of transactions
• Processing just the opposite of finish the results are stored
batch processing and next batch is taken up
• Examples: user interactions
during e-shopping and e-
banking.
Business Process

 A series of activities
 A collection of interrelated structured activities or tasks or processes which
follow a logical sequence.
 One that serves a particular specific result.
 Example: Business Process Tracking of RFID labeled

Process Matrix
 Number of elements
 Each element may represent series of operations and activities on given set
of inputs that perform a specific task leading to a decision point in the
process
 Distributed Business Process (DBPs)

 A collection of logically interrelated business processes in an Enterprise network

 A software system that manages the distributed BPs
Features
 A collection of logically related business processes,
 Cooperation between the processes in a transparent manner
 Each user within the system may access all of the process decisions within all of the
business processes as if they were a single business process
Location independence
 Enterprise business Intelligence unaware of where the processes are located
 Possible to move the results of analytics and knowledge from one physical location to
another without affecting the user.
 Interrelationships in the DBPs

 EBP1 interacts with device data, Adaptation, Data Integration and

Enrichment layer for a specific automobile.
 EBP1 analytics with the non-historical data.
 Network Layer Business Process

 Network layer Business process 2 (NBP2) interrelates directly with the

Enterprise layer Business Process 3 (EBP3).
NBP2
 Uses the Data Access, SQL, Query Processing, R- Descriptive Statistics,
and Component Specific Historical Databases.
 Has access to data of number of automobile of same model as one sending
data to EBP1.
EPB3
 EBP3 analytics interacts with the other automobiles databases and historical
databases
 Enables predictive and prescriptive Analytics

 NBP2 interrelates also directly with the DGBP4 updating the database for
the NBP2.
 Business Intelligence

 A process which enables a business service to extract new facts and

knowledge and then undertake the better decisions
 The new facts and knowledge follows from the earlier results of data
processing, aggregation and analysis of those results
 Extract new facts and knowledge from Business Processes at Application
layer
 Example: Automobile Service Centre Business Processes

Enterprise business Intelligence

 Processes move the results of analytics and knowledge from one physical
node to another in distributed business processes
 Application Integration
 Application software may consists of number of applications
 Applications integration: a process of deploying number of applications and
services into one application
 Example: Weather forecasting consisting of location, satellite maps and
predicative analytics
 Complex Application integration
 Integration of heterogeneous application architectures and number of
processes

 An Enterprise has number of Applications/Services and Processes.

 Heterogeneous systems complexity when integrating them in the

Enterprise.
 Oracle Application Integration Architecture

 Collaboration/Knowledge Management
 Enterprise Architecture/SOA
 e-Commerce
 External Customer Services
 Supply Chain Automation/Visualisation
 Data Centre Optimisation 
 Service
 A mechanism enabling the provisioning of access to one or more capabilities
described in service description
 An Interface for the service providing the access to capabilities
 Has a service description about the capabilities
 Applications or enterprise can subscribe on selection among number of services
 A service level agreement (SLA) binds the enterprise application and service
 The access to each capability consistent with the constraints and policies
 A collection of self contained, distinct and reusable components
 Providing the logically grouped and encapsulated functionalities. Example: Traffic
lights synchronizing service
Service-Oriented Architecture (SOA)
 Software architecture model consisting of Services, Messages, Operations
and Processes (Fig. 5.4)
 SOA components distribute over a network or Internet in a high level
business entity
 SOA enables development of new business Applications and Applications
integration architecture in an Enterprise
 Models the number of services and interrelationships
 Each service initiates on messages from a process or service
 Service Discovery, Selection and
Orchestration Layer
 Communicates with
 Web services
 IoT services
 Cloud of things
 Cloud
 Components select the services for Applications Integration
 Service Orchestration software coordinates the execution of the number of services,
cloud services, cloud IOT services and web services. Services run in parallel and a
number of processes in sequences.
Sublayer
 Enables Services run in parallel as well as a number of processes in sequences.
 Data Analytics for IoT /M2M Data
 Analytics enables finding new facts, taking decisions on those facts
 Organised Data after acquiring from the devices used for multiple purposes

 Phases of Analytics
1. Descriptive analytics: Enable deriving the additional value, from visualizations and
reports
2. Predictive analytics: Advanced Analytics which enables extraction of new facts and
knowledge, and then predict/forecast
3. Prescriptive analytics: Enable deriving the additional value, and undertake the better
decisions for new option(s) to maximize the profits.
 Descriptive Analytics Methods
1. Spreadsheet Based Reports and data visualisations : Spreadsheet enables user visualization of
what if.
 A value in cell CjRi (cell at jth column and ith row) can be related to another cell or a set of
cells through a formula or Boolean relation or statistically analysed value
2. Descriptive Statistics Based Reports and data visualisations
 Finding peak, minima, variance, probabilities, and statistical parameters
 Formulae used for the data sets to enable the understanding of variations in data shown
3. Descriptive analytics, statistics, data mining and machine learning analytics tools.
 Data Mining : Use of algorithms which extract hidden or unknown information or patterns
from large amount of data
 Machine Learning means modeling of the specific tasks
4. Online Analytical Processing (OLAP) for Analytics
 Advanced Analytics, OLAP and Data
Visualisation

OLAP
 Enables viewing of analysed data up to desired granularity
 Enables view of rollup (finer granulites data to coarse granulites data) or drill down (coarser
granulites data to finer granulites data
 Enables obtaining summarized information and automated reports from large volume
database
 A significant improvement over query systems
 An interactive system to show different summaries of multidimensional data by interactively
selecting the attributes in a multidimensional data cube
 Enables analysing data in multiple dimensions in a structure called data cube
 Each dimension represents a hierarchy
 Each dimension has a dimension attribute which defines the dimension and summary of
measure attribute
 A Slice of Data-cube
 Can be viewed when values of multiple dimensions are fixed.
 A slice of data relationship between two attributes can be individually
visualized
 Six different cross referenced tables can be created during OLAP for three-
dimensional structure for analysing data.

A Dice of Data-cube
 Can be viewed with variable values in multiple dimensions
 A cubical dice has six faces, each face marked distinctly. Face 1 has one
dot, face 2 two, and so on. Sixth face has six dots.
 Six different cross referenced tables can be created during OLAP for three-
dimensional structure for analysing data.
 Slice of N-dimensional Structure
 Will have 2-n faces (tables)
 Each table and corresponding visual gives a relationship between two attributes
 The tables are cross referenced.
 A Slice of An n-dimensional structure will have 2-n faces (tables). Each table and
corresponding visual gives a relationship between two attributes. The tables are cross
referenced.

A Dice of N-dimensional Structure

 Will have 2-n faces (tables)
 Each table and corresponding visual gives a relationship between two attributes
 The cross referenced tables
 Recall RDBMS
 A relational database is a collection of data into multiple tables which
relates to each other through special fields, called keys (primary key,
foreign key and unique key)

 RDBMS: a management system for the relational databases, (Creation,

connect, insertion, append, replace)
 Recall SQL
 A Structured Query Language
 A language for viewing or changing (update, insert or append or delete)
databases
 A language for data querying the databases
 A language for data access control,
 A language for schema creation and modifications
 Also a language for managing the RDBMS
 Also a language that can embed into other languages
Recall NoSQL (also called Not-only SQL)
 A Class of non-relational data storage systems, flexible data models
multiple schemas
 Class consisting of un-interpreted key and value or ‘the big hash table’
 Class consisting of unordered keys and using the JSON, for example in
MongoDb
 Big Data: Data of High Volume

Data of high volume Data of high Variety Data of higher velocity

• ‘Big data is data of high volume, • Variety means structured as • Velocity means data received
variety and velocity, and may also well as unstructured data in with higher rates due to use of
include veracity different formats number of sources of data

• Volume means data received from • Variety of data on which no

number of sources of data SQL (Structured Query
Language) applicable
• Includes data sets with sizes
beyond the ability of commonly • The multi-structured data
used software tools to acquire, compared to RDMS which
manage and process data within a maintains more structured data
tolerable elapsed time
 Hadoop
 An open-source framework for accesses to data in sequential manner
 Performs batch processing
 A new data set results from input data set that also processes sequentially
 Hadoop file system (HDFS)
 HBase
 Database for big data
 Data access─ random access
 Provides fast look-up from large tables
 Small access latency
 Database using big hash tables
 Considered similar to Google’s BigTable.
 HiveQL for Big Data analytics

 HiveQL, a SQL like scripting language software

 Used in Hadoop ecosystem (a collection of related entities and certain
processes that link to Hadoop components)
 (Hive is word derived from structure for housing  domesticated honeybees)

Mahout

 Distributed and Scalable Library of machine learning algorithms

 MapReduce

 MapReduce is programming model

 A core of Hadoop
 Large data sets process onto a cluster of nodes using MapReduce
 Same node runs the algorithm using the data sets at HDFS and processing
is at that node itself
Knowledge Acquiring, Managing
and
Storing processes
 Knowledge and Information Relationship
 According to an English dictionary is sharable information and
understanding about a subject or context
 Knowledge comes from researching up on the information
 Knowledge is an important asset of an enterprise

Data-Information-Knowledge-Wisdom (DIKW)

 Forms a pyramid
 Information is enriched set of data values when considered in the given
context and that can be queried upon
 The answer for the information comes from processing the data and
querying
 Knowledge Discovery Tools

 Provide the knowledge at particular amount of time as more and more data
process and analyse, additional knowledge discovers

Knowledge management
 Managing the knowledge when that new knowledge
 regularly discovers, processing and storing that knowledge
 Also provisions for replacing the earlier gathered knowledge
 Managing the life cycle of stored knowledge
 Knowledge-Management Reference
Architecture
 Highest layer has knowledge acquiring, managing, storing and knowledge
life cycle management
 Sublayers for managing, storing and knowledge life cycle management.
 Knowledge acquires from the use of information access tools and
knowledge discovery tools