Week 9

Introduction to Internet of Things
Week 9
Presented By: Riya Tapwal

Under the supervision of
Prof. Sudip Misra
Indian Institute of Technology, Kharagpur, India
IIT Kharagpur Introduction to Internet of Things March, 2024 1

OpenStack
 A software to create a cloud infrastructure
 Launched as a joint project of Rackspace Hosting and NASA in 2010
 Opensource
 Presently many companies are contributing to openstack
 Eg. IBM, CISCO, HP, Dell, Vmware, Redhat, suse, Rackspace hosting
 It has a very large community
 Can be used to develop private cloud or public cloud
 Versions: Austin, Bexar, Cactus, Diablo, Essex, Folsom, Grizzly, Havana, Icehouse, Juno, Kilo, Liberty, Mitaka, Newton, Ocata…,
Zed (2022).
Bobcat is the recent , Antelop is developed. Caracal is next release date is april 3 2024
IIT Kharagpur Introduction to Internet of Things Week 9 2

OpenStack: Components
 Keystone  Nova
 Identity service  Compute service

 Provides authentication and authorization  Where you launch your instances
 Horizon  Glance
 Dashboard  Image service

 GUI of the software  Discovering, registering, retrieving the VM
 Provides overview of the other components  Snapshots
 Swift
 Neutron
 Object storage
 Provides networking service
 Helps in storing data safely, cheaply and efficiently
 Enables the other services to communicate with each
other
 Make your own network

OpenStack: Components
 Cinder
 Block storage
 Virtualizes the management of block service
 Heat
 Orchestration
 Ceilometer
 Billing
 What service you are using
 How long are you using

Sensor Cloud
 It is not mere integration of sensors and cloud computing
 It is not only “dumping the sensor data into cloud”
 Concept of virtualization of sensor node

 Pay-per-use
 One sensor node/network appears as many
 A stratum between sensor nodes and end-users

Sensor Cloud: Virtualization Concept
 One computer host appears as many
computers-concept of Virtual Machine
(VM)
 Improve IT throughput and costs by
using physical resources as a pool from
which virtual resources can be allocated.
 Benefit
 Sharing of resources: Same resource can be
shared, in turn cost reduction
 Encapsulation: A complete computing
environment
 Independence: Runs independently of
underlying hardware
 Portability: VM Migration

Sensor Cloud: Difference with WSN

Sensor Cloud: Actors
 End-users
 Unknown about what and which physical sensor is/are allocated to serve the application
 Sensor-owner
 Plays a role from business perspective.
 They purchase physical sensor devices, deployed over different geographical locations, and lend these devices to the
sensor-cloud
 Sensor-Cloud Service Provider (SCSP)
 A business actor.
 SCSP charges price from the end-users as per their usage of Se-aaS.
Some examples of Sensor-Cloud Service Providers include AWS IoT, Microsoft Azure IoT, Google Cloud IoT, and
IBM Watson IoT.

Sensor Cloud: Architecture
 End-users: Registered themselves, selects templates, and
request for application(s)
 Sensor-owner: Deploy heterogeneous/
homogeneous physical sensor nodes over
different geographical location
 SCSP: Plays managerial role
Management Issues in Sensor-Cloud

 Optimal Composition of virtual sensor nodes
 Data Caching
 Optimal Pricing

Sensor Cloud: View

Virtual Sensor
 Efficient virtualization of the physical sensor nodes Why Composition of Virtual Sensor?
 An optimal composition of VSs  Resource-constrained sensor nodes
 Consider same geographic region: CoV-I  Dynamic change in sensor conditions
 Spanning across multiple regions: CoV-II  The composition of virtual sensors are non-traditional
CoV-I: Formation of Virtual Sensor
 Optimal formation of Virtual
Sensor (VS)
 Homogeneous sensor nodes
within same geographical
boundary
CoV-II: Formation of Virtual Sensor

Group
 Formation of Virtual Sensor
Group (VSG)
 Heterogeneous physical sensor
nodes across different
geographical locations
Dynamic And Adaptive Caching Mechanism
 Introduces internal and external caching mechanisms
 Ensures efficiency in resource utilization
 Flexible with the varied rate of change of the physical environment
Why Caching in Sensor-Cloud?
 End-users request for the sensed information through a Web-interface
 Allocation of physical sensor nodes and virtualization takes place
 Physical sensor nodes continuously sense and transmit data to sensor-cloud.
 Practically, in some cases, the change in environmental condition are significantly slow
 Due to the slow change in environment, the sensed data of physical sensors unaltered
 In such a situation, unnecessary sensing causes energy consumption
External and Internal Caching Mechanism
Internal Cache (IC)
 Handles requests from end-user
 Takes decision whether the data should be provided directly to the end user or is it required to re-cache the data from
external cache
External Cache (EC)
 After every certain interval data are required to re-cache
 Initially, few data are used to be transmitted to IC
Architecture of Caching

Dynamic Optimal Pricing for Sensor-Cloud Infrastructure
 Existing schemes consider homogeneity of service (e.g. for IaaS, SaaS)
 No scheme for SeaaS.
 The proposed pricing scheme comprises of two components:
 Pricing attributed to hardware (pH)

 Pricing attributed to Infrastructure (pI)
 Goal of the proposed pricing scheme:
 Maximizing profit of SCSP

 Maximizing profit of sensor owner
 End users’ satisfaction
Focus on:
 Maximizing the profit made by SCSP
 Optimal pricing to the end-users
 End users satisfaction
 Pricing attributed to hardware (pH): Deals with usage of physical sensor nodes
 Pricing attribute to infrastructure (pI): Deals with the price associated with infrastructure of sensor-cloud

Fog Computing
 Fog computing or fogging is a term coined by CISCO.
 The idea of fog computing is to extend the cloud nearer to the IoT
devices.
 The primary aim: solve the problems faced by cloud computing during
IoT data processing.
 an intermediate layer between cloud and devices.
 40% of the whole worlds data came from sensors alone by 2020.
 90% of the world’s data were generated only during the period of last
two years.
 2.5 quintillion bytes of data is generated per day.
 total expenditure on IoT devices will be $1.7 Trillion by 2020
 the total number of connected vehicles worldwide will be 250 millions by
2020.
 there will be more than 30 billion IoT devices
 The amount of data generated by IoT devices is simply huge.

Why Fog Computing
Why Fog Computing
 The ability of the current cloud model is insufficient to handle the requirements
of IoT.
 Issues are:
 Volume
 Latency
 Bandwidth
 Data Volume:
 By 2020, about 50 billion devices were online.
 Presently billions of devices produce exabytes of data
everyday.
 Device density is still increasing everyday.
 Current cloud model is unable to process this amount of data.
 Private firms, Factories, airplane companies produces
colossus amount of data everyday
 Current cloud model cannot store all these data
 Data need to be filtered

Why Fog Computing: Contd…
 Latency
 Time taken by a data packet for a round trip
 An important aspect for handing a time sensitive data.
 If edge devices send time sensitive data to cloud for analysis and wait for the cloud to give a proper action, then it can
lead to many unwanted results.
 While handling time sensitive data, a millisecond can make a huge differences.
 Sending time-sensitive data to cloud for analysis
 Latency will be increased

 When the action reaches the device, accident may have already occured

Why Fog Computing: Contd…
 Bandwidth:
 Bit-rate of data during transmission
 If all the data generated by IoT devices are sent to cloud for storage and analysis, then, the traffic generated by these
devices will be simply gigantic.
 consumes almost all the bandwidths.
 Handling this kind of traffic will be simply a very hard task.

Requirements of IoT
 Reduce latency of data:
 Appropriate actions at the right time prevents major accidents machine failure etc.
 A minute delay while taking a decision makes a huge difference
 Latency can be reduced by analyzing the data close to the data source
 Data security:
 IoT data must be secured and protected from the intruders.
 Data are required to be monitored 24x7
 An appropriate action should be taken before the attack causes major harm to the network
 Operation reliability:
 The data generated from IoT devices are used to solve real time problem
 Integrity and availability of the data must be guaranteed
 Unavailability and tampering of data can be hazardous
 Processing of data at respective suitable place
 Extremely time sensitive data should be analyzed very near to the data source
 Data which are not time sensitive will be analyzed in the cloud.

Requirements of IoT
 Monitor data across large geographical area:
 The location of connected IoT devices can be spread across a large geographical region
 E.g. monitoring the railway track of a country or a state
 the devices are exposed to the harsh environments condition
When should we use fog

 If the data should be analyze with fraction of second
 If there are huge number of devices
 If the devices are separated by a large geographical distance
 If the devices are needed to be subjected to extreme
conditions

Architecture of Fog
 Cloud services are extended to IoT devices
through fog
 Fog is a layer between cloud and IoT devices
 many fog nodes can be present
 Sensor data are processed in the fog before it
is sent to the cloud
 Reduces latency, save bandwidth and save the
storage of the cloud
Fog nodes
 Characteristics for a fog node:
 Storage - To give transient storage
 Computing facility
- To process the data before it is sent to cloud
- To take quick decisions
 Network connectivity - To connect with IoT devices, other fog nodes
and cloud
 E.g. - routers, embedded servers, switches, video surveillance etc.
 deployable anywhere inside the network.
 Each fog nodes have their aggregate fog node.

Working of Fog
 Three types of data
 Very time-sensitive data

 Less time-sensitive data
 Data which are not time-sensitive
 Fog nodes works according to the type of data they receive.

 An IoT application should be installed to each fog nodes.
 The nearest fog node ingest the data from the devices.
 Most time-sensitive data
 Data which should be analyzed within fraction of a second
 Analyze at the nearest node itself  Less time-sensitive data
 Sends the decision or action to the devices  Data which can be analyzed after seconds or minutes
 Sends and stores the summary to cloud for future analysis  Are sent to the aggregate node for analysis
 Non-time-sensitive data  After analysis, the aggregate node send the decision or action
 Data which can be wait for hours, days, weeks to the device through the nearest node
 Sent to cloud for storage and future analysis.  The aggregate node sends the summary to cloud for storage
 Those summaries from fog nodes can be considered as less and future analysis
time sensitive data.

Working of Fog

Advantages of Fog
 Security  Business agility
 Provides better security  Fog application can be easily developed according to tools
 Fog nodes can use the same security policy available
 Can be deployed anywhere we need
 Low operation cost  Can be programed according to the customer’s need
 Data are processed in the fog nodes before sending to cloud
 Reduces the bandwidth consumption  Support mobility
 Nodes can be mobile
 Reduces unwanted accidents  Nodes can join and leave the network anytime
 Latency will be reduce during decision making
 Quick decision making  Deployable in remote places
 Can be deployed in remote places
 Better privacy  Can be subjected to harsh environmental conditions
 Every industry can analyze their data locally  Under sea, railway tracks, vehicles, factory floor etc
 Store confidential data in their local servers
 Send only those data which can be shared to the cloud  Better data handling
 Can operate with less bandwidth
 Data can be analyzed locally
 Reduce the risk of latency

Applications of Fog
 Real time health analysis  Real time rail monitoring
 Patients with chronic illness can be monitored in real time  Fog nodes can be deployed to railway tracks
 Stroke patients  Real time monitoring of the track conditions
 Analyze the data real time  For high speed train, sending the data in cloud for
 During emergency, alerts the respective doctors immediately analysis is inefficient
 Historical data analysis can predict future dangers of the patient  Fog nodes provide fast data analysis
 Improve safety and reliability
 Intelligence power efficient system
 Power efficient  Pipeline optimization
 Reports detail power consumption report everyday  Gas and oils are transported through pipelines
 Suggest economical power usage plan  Real time monitoring of pressure, flow, compressor is
necessary
 Terabytes of data are created
 Real time wind mill and turbine analysis  Sending all this data to cloud for analysis and storage is
 Wind direction and speed analysis can increase output not efficient
 Data can be monitored real time  Network latency is not acceptable
 Fog is a solution

Challenges of Fog
 Power consumption  Programming architecture
 Fog use addition nodes  Fog nodes may be mobile
 Power consumption is higher than centralized cloud  Nodes can connect and leave the network when necessary
 Many data processing frameworks are statically
configured
 Data Security  These frameworks cannot provide proper scalability and
 Data generating nodes are distributed flexibility
 Providing authentication and authorization system for
the whole nodes is not an easy task
 Real time analysis

 Reliability  Real time analysis is a primary requirement for minimizing
 Maintaining data integrity and availability for millions of latency
nodes is difficult  Dynamic analysis and decision making reduces danger and
 failure of a node cannot affect the network increase output
 Monitor huge number of nodes is not easy
 Fault tolerance
 Failure of a node should be immediately fixed
 Individual failure should not affect the whole scenario
Question No: 1
OpenStack is proprietary and a closed source software that you have to pay and buy.
a. True
b. False

Question No: 1
OpenStack is proprietary and a closed source software that you have to pay and buy.
a. True
b. False
OpenStack is open source and requires no payment.

Question No: 2
In OpenStack which of the following component takes care of user authentication?
a. Nova
b. Horizon
c. Keystone
d. None of these

Question No: 2
In OpenStack which of the following component takes care of user authentication?
a. Nova
b. Horizon
c. Keystone
d. None of these
Keystone component in OpenStack provides authentication and user authorization features

Question No: 3
You cannot ping your OpenStack instance from an outside network unless you connect your instance with
the public network through a ____________
a. Router
b. Firewall
c. Repeater
d. Load balancer

Question No: 3
You cannot ping your OpenStack instance from an outside network unless you connect your instance with
the public network through a ____________
a. Router
b. Firewall
c. Repeater
d. Load balancer
You are supposed to connect your virtual OpenStack instance with the public network through a router

Question No: 4
What is the primary service model for a sensor cloud?
a. Hardware-as-a-Service
b. Platform-as-a-Service
c. Data-as-a-Service
d. Sensor-as-a-Service

Question No: 4
What is the primary service model for a sensor cloud?
a. Hardware-as-a-Service
b. Platform-as-a-Service
c. Data-as-a-Service
d. Sensor-as-a-Service
The primary functionality of a sensor cloud is to provide Sensor-as-aService.

Question No: 5
You can run two logical computers, one Linux machine and one Windows machine on top of the same
physical hardware through the use of ________________.
a. Sharing
b. Scaling
c. Virtualization
d. Inheritance.

Question No: 5
You can run two logical computers, one Linux machine and one Windows machine on top of the same
physical hardware through the use of ________________.
a. Sharing
b. Scaling
c. Virtualization
d. Inheritance.
Virtualization supports multiple logical instances of different computers on the same hardware.

Question No: 6
In Sensor Cloud, one physical sensor can be abstracted as multiple virtual sensors.
a. True
b. False

Question No: 6
In Sensor Cloud, one physical sensor can be abstracted as multiple virtual sensors.
a. True
b. False
Virtualization allows the abstraction of one physical sensors into multiple virtual sensors.

Question No: 7
In a Sensor cloud framework, the end-users have to continuously worry about owning and managing the
physical sensor devices.
a. True
b. False

Question No: 7
In a Sensor cloud framework, the end-users have to continuously worry about owning and managing the
physical sensor devices.
a. True
b. False
Sensor cloud supports the existence of multiple actors with end-users concerned only with the application data and not
internal details.

Question No: 8
Virtual sensors within a sensor cloud have communication interfaces with _______________.
a. Only the physical sensors below them

b. Only the end-user applications above them.
c. Both physical sensors below and applications above.
d. Neither the physical sensors, nor the applications above.

Question No: 8

Virtual sensor layer sits in the middle having interface to both physical sensors as well as applications.

Question No: 9
Which of the following is true?
a. Traditional WSN: Virtualization support, Sensor Cloud: Virtualization support

b. Traditional WSN: No virtualization, Sensor Cloud: Virtualization Support
c. Traditional WSN: Virtualization support, Sensor Cloud: No Virtualization
d. Traditional WSN: No virtualization, Sensor Cloud: No Virtualization

Question No: 9
Which of the following is true?
a. Traditional WSN: Virtualization support, Sensor Cloud: Virtualization support

b. Traditional WSN: No virtualization, Sensor Cloud: Virtualization Support
c. Traditional WSN: Virtualization support, Sensor Cloud: No Virtualization
d. Traditional WSN: No virtualization, Sensor Cloud: No Virtualization
Sensor Cloud supports complete virtualization

Question No: 10
An organization ABC wishes to have environmental monitoring parameters of a particular region, but
does not have the resources to deploy its own physical sensors. However, another organization B has
some physical sensors that monitor the environment deployed in that region. Can sensor cloud help in
achieving A’s objectives?
a. Yes
b. No

Question No: 10
An organization ABC wishes to have environmental monitoring parameters of a particular region, but
does not have the resources to deploy its own physical sensors. However, another organization B has
some physical sensors that monitor the environment deployed in that region. Can sensor cloud help in
achieving A’s objectives?
a. Yes
b. No
Sensor cloud can provide virtual sensor services to A

Question No: 11
Referring back to Question 10 (above) and based on your answer, determine the roles that organization A
and B play in the given scenario.
a. A: Sensor owner, B: End-user (service consumer)

b. A: End-User (service consumer), B: Sensor owner
c. Both A and B are sensor owners
d. Both A and B are end-users

Question No: 11
Referring back to Question 10 (above) and based on your answer, determine the roles that organization A
and B play in the given scenario.
a. A: Sensor owner, B: End-user (service consumer)

b. A: End-User (service consumer), B: Sensor owner
c. Both A and B are sensor owners
d. Both A and B are end-users
As B owns the physical sensors and has deployed them, B is the sensorowner and as A requires the sensor services, it is
the end-user

Question No: 12
Which among the following is a limitation of cloud computing?
a. On-demand services
b. Virtualization orchestration
c. Large amount of computing and storage resources.
d. Additional communication latency for critical application

Question No: 12
Which among the following is a limitation of cloud computing?
a. On-demand services
b. Virtualization orchestration
c. Large amount of computing and storage resources.
d. Additional communication latency for critical application
Communication latency to and from cloud is one of the major drawback for clouds, especially for
critical applications

Question No: 13
In Fog computing, the Fog layer acts as an intermediary between the data source and the cloud for
immediate processing of time critical data.
a. True
b. False

Question No: 13
In Fog computing, the Fog layer acts as an intermediary between the data source and the cloud for
immediate processing of time critical data.
a. True
b. False
Fog actually sits in between the device and cloud to provide immediate services

Question No: 14
Suppose the one-way communication latency from a device to a Fog layer is 20 ms and that from the
device to the cloud layer is 50ms. Data processing time at Fog is 10ms and that at Cloud is 5 ms, after
which (i.e completion of data processing) a reply is sent back to the device by the Fog or Cloud. What is
the difference between total communication latency (two-way round-trip communication plus processing
time at Fog/Cloud) between the device and the Cloud and Fog respectively?
a. 50ms
b. 60ms
c. 55ms
d. 100ms

Question No: 14
Suppose the one-way communication latency from a device to a Fog layer is 20 ms and that from the
device to the cloud layer is 50ms. Data processing time at Fog is 10ms and that at Cloud is 5 ms, after
which (i.e completion of data processing) a reply is sent back to the device by the Fog or Cloud. What is
the difference between total communication latency (two-way round-trip communication plus processing
time at Fog/Cloud) between the device and the Cloud and Fog respectively?
a. 50ms
b. 60ms
c. 55ms
d. 100ms
Total communication latency for cloud = 50 + 5 + 50 = 105ms.

Total communication latency for Fog = 20 + 10 + 20 = 50ms.
Therefore difference = 105-50 = 55 ms

Question No: 15
Which among the following is true?
a. Fog computing acts as a complement to cloud computing.

b. Fog computing is a replacement for cloud computing.
c. Fog computing and cloud computing are the same.
d. Fog computing is more powerful than cloud computing (with respect to resources).

Question No: 15

Fog computing and cloud computing are complementary technologies.

Fog helps in bringing the cloud closer to the IoT devices.

Question No: 16
The OpenStack cloud simulation framework provides an interactive Graphical User Interface (GUI)
a. True
b. False

Question No: 16
The OpenStack cloud simulation framework provides an interactive Graphical User Interface (GUI)
a. True
b. False

Question No: 17
Virtualized resources within the OpenStack simulator that you can define, set parameters of and deploy
within OpenStack are also known as
a. Instances
b. Files
c. Hypervisors
d. Sketches

Question No: 17
a. Instances
b. Files
c. Hypervisors
d. Sketches

Question No: 18
If you want to connect your OpenStack instance with the external public network (like the Internet), you
would need to connect the following between your instance and the public network
a. Database
b. File
c. Memory
d. Router

Question No: 18
If you want to connect your OpenStack instance with the external public network (like the Internet), you
would need to connect the following between your instance and the public network
a. Database
b. File
c. Memory
d. Router

Question No: 19
Which among the following is NOT a part of a sensor node connected to a wireless sensor network.
a. Sensing unit
b. Gaming unit
c. Processing unit
d. Communication unit

Question No: 19
Which among the following is NOT a part of a sensor node connected to a wireless sensor network.
a. Sensing unit
b. Gaming unit
c. Processing unit
d. Communication unit

Question No: 20
Consider that you have two types of sensors, soil monitoring sensor for smart farming and temperature sensor for smart
industry. The concept that the correct (right) type of sensor must be deployed only at their correct and appropriate
physical location suitable as per their functionalities is known as
a. Right way of deployment

b. Right place of deployment
c. Right time of deployment
d. None of these

Question No: 20
Consider that you have two types of sensors, soil monitoring sensor for smart farming and temperature sensor for smart
industry. The concept that the correct (right) type of sensor must be deployed only at their correct and appropriate
physical location suitable as per their functionalities is known as
a. Right way of deployment

b. Right place of deployment
c. Right time of deployment
d. None of these

Question No: 21
Sensor-as-a-Service (Se-aaS) is an exciting new concept that brings the service models of cloud
computing to traditional IoT sensor networks. In this aspect, which among the following forms an
essential component of a Se-aaS architecture.
a. Sensor marketing
b. Sensor division
c. Sensor virtualization
d. Sensor manufacturing

Question No: 21
Sensor-as-a-Service (Se-aaS) is an exciting new concept that brings the service models of cloud
computing to traditional IoT sensor networks. In this aspect, which among the following forms an
essential component of a Se-aaS architecture.
a. Sensor marketing
b. Sensor division
d. Sensor manufacturing

Question No: 22
Considering the end-to-end architecture of a WSN, from the physical sensor nodes right up to the users, sensor-cloud
supports different entities to have ownership of the different layers and components simultaneously.
a. True
b. False

Question No: 22
Considering the end-to-end architecture of a WSN, from the physical sensor nodes right up to the users, sensor-cloud
supports different entities to have ownership of the different layers and components simultaneously.
a. True
b. False

Question No: 23
In sensor cloud architecture the Sensor Cloud Service Provider (SCSP) is logically situated at the
following position of the architecture
a. In the same level as the physical sensors

b. In between the physical sensors and the user layer
c. In the same level as the users
d. None of these

Question No: 23
In sensor cloud architecture the Sensor Cloud Service Provider (SCSP) is logically situated at the
following position of the architecture
a. In the same level as the physical sensors

b. In between the physical sensors and the user layer
c. In the same level as the users
d. None of these

Question No: 24
In sensor cloud, the association between virtual sensor instances and the corresponding physical sensors follows
a. one to one mapping

b. one to many mapping
c. many to one mapping
d. many to many mapping

Question No: 24
In sensor cloud, the association between virtual sensor instances and the corresponding physical sensors follows
a. one to one mapping

b. one to many mapping
c. many to one mapping
d. many to many mapping

Question No: 25
Dynamic caching mechanism improves the flexibility and efficiency of sensor cloud.
a. True
b. False

Question No: 25
Dynamic caching mechanism improves the flexibility and efficiency of sensor cloud.
a. True
b. False

Question No: 26
With respect to the caching based architecture of a sensor cloud, the External Cache (EC) has a direct data connection
with which among the following
a. The Internal Cache (IC)

b. The user applications
c. Physical sensors
d. Both physical sensors and Internal Cache (IC)

Question No: 26
With respect to the caching based architecture of a sensor cloud, the External Cache (EC) has a direct data connection
with which among the following
a. The Internal Cache (IC)

b. The user applications
c. Physical sensors
d. Both physical sensors and Internal Cache (IC)

Question No: 27
Data from an IoT device is transferred to cloud via a network, which is then processed at the cloud and then a response is
sent back to the IoT device from the cloud after processing. The time it takes for one- way data transfer between the node
and cloud is 10s and the data processing time at the cloud is ‘x’ seconds. It takes a total of 25s for the entire to and fro
transfer of data between the sensor and cloud along
with processing at the cloud. What is the value of x?
a. 10s
b. 5s
c. 15s
d. 20s

Question No: 27
transfer of data between the sensor and cloud along
with processing at the cloud. What is the value of x?
a. 10s
b. 5s
c. 15s
d. 20s

Question No: 28
There are two types of sensor data, A and B. A is time sensitive that is required to be processed immediately, while B is not
time sensitive and can tolerate longer time for processing. As per the standard utilities of cloud and fog computing, which
among the following options show the correct processing locations for A and B.
a. A:Cloud, B: Fog
b. A: Fog, B: Cloud
c. None of these
d. Both of these

Question No: 28
There are two types of sensor data, A and B. A is time sensitive that is required to be processed immediately, while B is not
time sensitive and can tolerate longer time for processing. As per the standard utilities of cloud and fog computing, which
among the following options show the correct processing locations for A and B.
a. A:Cloud, B: Fog
b. A: Fog, B: Cloud
c. None of these
d. Both of these

Question No: 29
There can be multiple fog nodes in between the physical sensor layer at the bottom and the cloud layer at the top.
a: True
b. False

Question No: 29
There can be multiple fog nodes in between the physical sensor layer at the bottom and the cloud layer at the top.
a: True
b. False

Question No: 29
Which among the following is/are a potential problem and challenge in fog computing?
a. Power consumption
b. Data security
c. Reliability
d. All of these

Question No: 29
Which among the following is/are a potential problem and challenge in fog computing?
a. Power consumption
b. Data security
c. Reliability
d. All of these

Question No: 30
Openstack is a free open source software for cloud framework simulation and experimentation with various cloud
applications.
a. True
b. False

Question No: 30
applications.
a. True
b. False

Question No: 31
Which among the following is NOT a component of OpenStack.
a. Horizon
b. Heat
c. Plasma
d. Neutron

Question No: 31
Which among the following is NOT a component of OpenStack.
a. Horizon
b. Heat
c. Plasma
d. Neutron

Question No: 32
You cannot ping your OpenStack instance from an outside network unless you connect your instance with the public
network through a ____________
a. Router
b. Firewall
c. Repeater
d. Load balancer

Question No: 32
You cannot ping your OpenStack instance from an outside network unless you connect your instance with the public
network through a ____________
a. Router
b. Firewall
c. Repeater
d. Load balancer

Question No: 33
a. Instances
b. Files
c. Hypervisors
d. Sketches

Question No: 33
a. Instances
b. Files
c. Hypervisors
d. Sketches

Question No: 34
The SCSP in sensor clouds is responsible for caching the data in the databases.
a. True
b. False

Question No: 34
The SCSP in sensor clouds is responsible for caching the data in the databases.
a. True
b. False

Question No: 35
Select the correct option for replacing the box (in red) in the following architecture for sensor cloud user
organization view.
a. User credentials
b. Template display
c. Template view
d. User view

Question No: 35
Select the correct option for replacing the box (in red) in the following architecture for sensor cloud user
organization view.
a. User credentials
b. Template display
c. Template view
d. User view

Question No: 36
Which among the following is the principal feature of sensor clouds, with respect to sensor nodes?
a. Sensor monitoring
b. Sensor instantiation
d. Sensor collection

Question No: 36
Which among the following is the principal feature of sensor clouds, with respect to sensor nodes?
a. Sensor monitoring
b. Sensor instantiation
d. Sensor collection
IIT Kharagpur Week 9 10

0
Question No: 37
Caching in sensor cloud provide no benefit over regions with slow environmental monitoring rate.
a. True
b. False

1
Question No: 37
Caching in sensor cloud provide no benefit over regions with slow environmental monitoring rate.
a. True
b. False

2
Question No: 38
How many different types of caching mechanism are there in sensor cloud?
a. 1
b. 4
c. 2
d. 3

3
Question No: 38
How many different types of caching mechanism are there in sensor cloud?
a. 1
b. 4
c. 2
d. 3

4
Question No: 39


5
Question No: 39


6
Question No: 40
transfer of data between the sensor and cloud along with processing at the cloud. What is the value of x?
a. 10s
b. 5s
c. 15s
d. 20s

7
Question No: 40
transfer of data between the sensor and cloud along with processing at the cloud. What is the value of x?
a. 10s
b. 5s
c. 15s
d. 20s

8
Question No: 41
In IoT, temporal sensitivity of data plays an important role
a. True
b. False

9
Question No: 41
In IoT, temporal sensitivity of data plays an important role
a. True
b. False

0
Question No: 42
Extremely time sensitive data in IoT should be processed and analyzed ____________.
a. Farthest away from source

b. Closest to the source
c. Distance between source and place of analysis does not matter
d. None of these

1
Question No: 42
Extremely time sensitive data in IoT should be processed and analyzed ____________.
a. Farthest away from source

b. Closest to the source
c. Distance between source and place of analysis does not matter
d. None of these

2
Question No: 43
Which among the following is a view of a Fog Computing Architecture?
a. Node View
b. System View
c. Software View
d. All of these

3
Question No: 43
Which among the following is a view of a Fog Computing Architecture?
a. Node View
b. System View
c. Software View
d. All of these

4
Question No: 44


5
Question No: 44


6
Question No: 45
applications.
a. True
b. False

7
Question No: 45
applications.
a. True
b. False

8
Question No: 46
Which component of OpenStack do you use to access all the other components?
a. Horizon
b. Glance
c. Neutron
d. None of these

9
Question No: 46
Which component of OpenStack do you use to access all the other components?
a. Horizon
b. Glance
c. Neutron
d. None of these

0
Question No: 47
Which among the following is NOT a component of OpenStack?
a. Horizon
b. Heat
c. Plasma
d. Neutron

1
Question No: 47
Which among the following is NOT a component of OpenStack?
a. Horizon
b. Heat
c. Plasma
d. Neutron

2
Question No: 48
Sensor cloud is simply dumping and organizing of sensor data on cloud computing platforms.
a. True
b. False

3
Question No: 48
Sensor cloud is simply dumping and organizing of sensor data on cloud computing platforms.
a. True
b. False

4
Question No: 49
Which among the following are limitations of traditional Wireless Sensor Networks (WSNs)?
a. Procurement Issues
b. Deployment Issues
c. Maintenance Issues
d. All of the given

5
Question No: 49
Which among the following are limitations of traditional Wireless Sensor Networks (WSNs)?
a. Procurement Issues
b. Deployment Issues
c. Maintenance Issues
d. All of the given

6
Question No: 50
In a typical sensor cloud architecture, the sensor cloud infrastructure that provides the virtualization lies _____________
______________________________
a. At the same layer as physical sensor devices
b. At the application layer
c. In between the physical sensor layer and the application layer
d. Sensor cloud does not support virtualization

7
Question No: 50
In a typical sensor cloud architecture, the sensor cloud infrastructure that provides the virtualization lies _____________
______________________________
a. At the same layer as physical sensor devices
b. At the application layer
c. In between the physical sensor layer and the application layer
d. Sensor cloud does not support virtualization

8
Question No: 51
Generally speaking, Sensor Cloud Service Providers (SCSPs) are also always the owners of the physical
sensors
a. Yes
b. No

9
Question No: 51
Generally speaking, Sensor Cloud Service Providers (SCSPs) are also always the owners of the physical
sensors
a. Yes
b. No

0
Question No: 52
In a typical sensor cloud architecture with virtualization, one virtual sensor can be associated with
how many physical sensors?
a. Only one
b. One or more than one
c. None
d. Only two

1
Question No: 52
In a typical sensor cloud architecture with virtualization, one virtual sensor can be associated with
how many physical sensors?
a. Only one
b. One or more than one
c. None
d. Only two

2
Question No: 53
Sensor virtualization aims to achieve more _________________________ in providing sensor
based services
a. Complexity
b. Flexibility

3
Question No: 53
Sensor virtualization aims to achieve more _________________________ in providing sensor
based services
a. Complexity
b. Flexibility

4
Question No: 54
Suppose that you want to start a business to provide some IoT based application, but you do not have the means to
develop the application layer logic, neither you have the means to purchase and deploy physical sensors. You can
however, rent cloud servers for use and write interface logic for interfacing with other modules. Which among the
following actors will be the most suitable for you?
a. Physical sensor owner

b. Application layer developer
c. Sensor Cloud Service Provider (SCSP)
d. You cannot start the business

5
Question No: 54
Suppose that you want to start a business to provide some IoT based application, but you do not have the means to
develop the application layer logic, neither you have the means to purchase and deploy physical sensors. You can
however, rent cloud servers for use and write interface logic for interfacing with other modules. Which among the
following actors will be the most suitable for you?
a. Physical sensor owner

b. Application layer developer
c. Sensor Cloud Service Provider (SCSP)
d. You cannot start the business

6
Question No: 55
Which among the following is implemented along with sensor cloud to make its services and performance
better?
a. Cashing
b. Caching
c. Casing
d. Calling

7
Question No: 55
Which among the following is implemented along with sensor cloud to make its services and performance
better?
a. Cashing
b. Caching
c. Casing
d. Calling

8
Question No: 56
Which among the following cases is most likely to reduce the overall price of sensor-cloud implementation provided that
the data traverses through multiple sensor hops starting from the origin sensor to the sink node, and provided that all
owners are honest and charge for only what is required?
a. If physical sensors are owned by multiple owners with high profit margin
b. If all physical sensors are owned by a single owner with uniform profit margin

9
Question No: 56
Which among the following cases is most likely to reduce the overall price of sensor-cloud implementation provided that
the data traverses through multiple sensor hops starting from the origin sensor to the sink node, and provided that all
owners are honest and charge for only what is required?
a. If physical sensors are owned by multiple owners with high profit margin
b. If all physical sensors are owned by a single owner with uniform profit margin

0
Question No: 57
Fog computing is aimed to replace cloud computing completely and has no scope for integration with
cloud
a. True
b. False

1
Question No: 57
Fog computing is aimed to replace cloud computing completely and has no scope for integration with
cloud
a. True
b. False

2
Question No: 58
Suppose data from an IoT device first goes to Fog layer for some basic processing, after which it goes to Cloud layer for
advanced processing, then the processed data comes back to the Fog layer and then, it finally comes back to the origin
sensor node. In the return journey no processing of data takes place anywhere, just transfer of data takes place. If ‘Tf’ is
the time taken by the data to travel from sensor to fog and vice versa, and ‘Tc’ is the time taken by the data to travel from
fog to cloud and vice versa, ‘Tfp’ is the data processing time at fog and ‘Tcp’ is the data processing time at cloud, what is
the total round trip time ‘T’ taken by data starting from the origin sensor node, processing the data and then back to the
sensor node after being processed.
a. T = Tf + Tc + Tfp + Tcp
b. b. T = Tf + Tc
c. T = 2(Tf + Tc) + Tfp + Tcp
d. T = 4(Tf + Tc + Tfp + Tcp)

3
Question No: 58
Suppose data from an IoT device first goes to Fog layer for some basic processing, after which it goes to Cloud layer for
advanced processing, then the processed data comes back to the Fog layer and then, it finally comes back to the origin
sensor node. In the return journey no processing of data takes place anywhere, just transfer of data takes place. If ‘Tf’ is
the time taken by the data to travel from sensor to fog and vice versa, and ‘Tc’ is the time taken by the data to travel from
fog to cloud and vice versa, ‘Tfp’ is the data processing time at fog and ‘Tcp’ is the data processing time at cloud, what is
the total round trip time ‘T’ taken by data starting from the origin sensor node, processing the data and then back to the
sensor node after being processed.
a. T = Tf + Tc + Tfp + Tcp
b. b. T = Tf + Tc
c. T = 2(Tf + Tc) + Tfp + Tcp
d. T = 4(Tf + Tc + Tfp + Tcp)

4
Question No: 59
Consider the standard Fog computing architecture. In which of the following layer will ‘very time sensitive data’ be
processed?
a. Nearest fog node

b. Distant aggregate fog node
c. Cloud
d. Does not matter

5
Question No: 59
Consider the standard Fog computing architecture. In which of the following layer will ‘very time sensitive data’ be
processed?
a. Nearest fog node

b. Distant aggregate fog node
c. Cloud
d. Does not matter

6
Question No: 60
“Network, Accelerator, Compute and Storage” constitute the part of which view of fog computing
architecture, as defined by OpenFog Consortium Architecture Working Group?
a. System View
b. Node View
c. Software View
d. None of these

7
Question No: 60
“Network, Accelerator, Compute and Storage” constitute the part of which view of fog computing
architecture, as defined by OpenFog Consortium Architecture Working Group?
a. System View
b. Node View
c. Software View
d. None of these

8
Thank You

9

Week 9

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Week 9

Uploaded by

Copyright:

Available Formats

Introduction to Internet of Things

Presented By: Riya Tapwal

Prof. Sudip Misra

Indian Institute of Technology, Kharagpur, India

IIT Kharagpur Introduction to Internet of Things March, 2024 1

IIT Kharagpur Introduction to Internet of Things Week 9 2

 Identity service  Compute service

 Dashboard  Image service

IIT Kharagpur Introduction to Internet of Things Week 9 3

IIT Kharagpur Introduction to Internet of Things Week 9 4

 Concept of virtualization of sensor node

IIT Kharagpur Introduction to Internet of Things Week 9 5

IIT Kharagpur Introduction to Internet of Things Week 9 6

IIT Kharagpur Introduction to Internet of Things Week 9 7

IIT Kharagpur Introduction to Internet of Things Week 9 8

Management Issues in Sensor-Cloud

IIT Kharagpur Introduction to Internet of Things Week 9 9

IIT Kharagpur Introduction to Internet of Things Week 9 10

CoV-II: Formation of Virtual Sensor

IIT Kharagpur Introduction to Internet of Things Week 9 13

 Pricing attributed to hardware (pH)

 Goal of the proposed pricing scheme:

 Maximizing profit of SCSP

IIT Kharagpur Introduction to Internet of Things Week 9 14

IIT Kharagpur Introduction to Internet of Things Week 9 15

IIT Kharagpur Introduction to Internet of Things Week 9 16

 Latency will be increased

IIT Kharagpur Introduction to Internet of Things Week 9 17

IIT Kharagpur Introduction to Internet of Things Week 9 18

IIT Kharagpur Introduction to Internet of Things Week 9 19

When should we use fog

IIT Kharagpur Introduction to Internet of Things Week 9 20

IIT Kharagpur Introduction to Internet of Things Week 9 21

 Very time-sensitive data

 Fog nodes works according to the type of data they receive.

IIT Kharagpur Introduction to Internet of Things Week 9 22

IIT Kharagpur Introduction to Internet of Things Week 9 23

IIT Kharagpur Introduction to Internet of Things Week 9 24

IIT Kharagpur Introduction to Internet of Things Week 9 25

 Real time analysis

IIT Kharagpur Introduction to Internet of Things Week 9 27

OpenStack is open source and requires no payment.

IIT Kharagpur Introduction to Internet of Things Week 9 28

In OpenStack which of the following component takes care of user authentication?

IIT Kharagpur Introduction to Internet of Things Week 9 29

In OpenStack which of the following component takes care of user authentication?

Keystone component in OpenStack provides authentication and user authorization features

IIT Kharagpur Introduction to Internet of Things Week 9 30

IIT Kharagpur Introduction to Internet of Things Week 9 31

IIT Kharagpur Introduction to Internet of Things Week 9 32

What is the primary service model for a sensor cloud?

IIT Kharagpur Introduction to Internet of Things Week 9 33

What is the primary service model for a sensor cloud?

The primary functionality of a sensor cloud is to provide Sensor-as-aService.

IIT Kharagpur Introduction to Internet of Things Week 9 34

IIT Kharagpur Introduction to Internet of Things Week 9 35

IIT Kharagpur Introduction to Internet of Things Week 9 36

IIT Kharagpur Introduction to Internet of Things Week 9 37

IIT Kharagpur Introduction to Internet of Things Week 9 38

IIT Kharagpur Introduction to Internet of Things Week 9 39

IIT Kharagpur Introduction to Internet of Things Week 9 40

a. Only the physical sensors below them

IIT Kharagpur Introduction to Internet of Things Week 9 41