Advance

Computing

Fog Computing
Contents :

• Introduction to Fog Computing

• Characteristics
• Application Scenarios
• Issues and challenges
• Fog Computing Architecture
• Communication and Network Model
• Fog Architecture for smart cities, healthcare and vehicles

2
Introduction to Fog Computing

• Cisco invented the phrase "Fog Computing," which refers to

extending cloud computing to an enterprise's network's edge.

• It makes computation, storage, and networking services more

accessible between end devices and computing data centers.
• Fog computing is the computing, storage, and communication
architecture that employs EDGE devices to perform a
significant portion of computation, storage, and communication
locally before routing it over the Internet backbone.
Introduction to Fog Computing
Introduction to Fog Computing

• Fog computing is a type of distributed computing that connects a cloud to a

number of "peripheral" devices. (The term "fog" refers to the edge or
perimeter of a cloud.)

• Rather than sending all of this data to cloud-based servers to be processed,

many of these devices will create large amounts of raw data (for example,
via sensors).

• The goal of fog computing is to conduct as much processing as possible

using computing units that are co-located with data-generating devices so
that processed data rather than raw data is sent and bandwidth needs are
decreased.

• Another advantage of processing locally rather than remotely is that the

processed data is more needed by the same devices that created the data, and
the latency between input and response is minimized.
Introduction to Fog Computing
Differences between cloud computing and
fog computing
Specialty Cloud Computing fog computing
Delay Cloud computing has higher latency than Fog computing has low latency
fog computing
Capacity Cloud computing does not provide any Fog computing reduces the amount of data
reduction in data while sending or sent to cloud computing.
converting data.
Responsiveness The response time of the system is low. The response time of the system is high.
Security Cloud computing has less Security Fog computing has high Security.
compared to Fog Computing
Speed Access speed is high depending on the High even more compared to Cloud
VM connectivity. Computing.
Data Integration Multiple data sources can be integrated. Multiple Data sources and devices can be
integrated.
Mobility In cloud computing, mobility is Limited. Mobility is supported in fog computing.
Location Awareness Partially Supported in Cloud computing. Supported in fog computing.
Number of Server Nodes Cloud computing has Few numbers server Fog computing has a Large number of server
nodes. nodes.
Geographical It is centralized. It is decentralized and distributed.
Distribution
Location of service Services provided within the Internet. Services are provided at the edge of the local
network.
Working environment Specific data center building with air Outdoor (streets, base stations, etc.) or indoor
conditioning systems (houses, cafes, etc.)
Communication mode IP network Wireless communication: WLAN, WiFi, 3G, 4G,
ZigBee, etc. or wired communication (part of
the IP networks)
Dependence on the Requires strong network core. It can also work in a Weak network core.
quality of core network
Comparison between Fog Computing
and Edge Computing.
EDGE COMPUTING FOG COMPUTING

Less scalable than fog computing. Highly scalable when compared to edge computing.

Billions of nodes are present. Millions of nodes are present.

Nodes in this computing are installed closer to the

Nodes are installed far away from the cloud.
cloud(remote database where data is stored).

Edge computing is a subdivision of fog computing. Fog computing is a subdivision of cloud computing.

The bandwidth requirement is very low. Because data The bandwidth requirement is high. Data originating from
comes from the edge nodes themselves. edge nodes is transferred to the cloud.

Operational cost is higher. Operational cost is comparatively lower.

High privacy. Attacks on data are very low. The probability of data attacks is higher.

Edge devices are the inclusion of the IoT devices or

Fog is an extended layer of cloud.
client’s network.

The power consumption of nodes filter important

The power consumption of nodes is low. information from the massive amount of data collected
from the device and saves it in the filter high.

Fog computing helps in filtering important information

Edge computing helps devices to get faster results by
from the massive amount of data collected from the
processing the data simultaneously received from the
device and saves it in the cloud by sending the filtered
devices.
data.
Application of Fog Computing

• It's utilized when only a small amount of data has to be sent to

the cloud. This data is chosen for long-term storage and is
accessed by the host less frequently.

• It's utilized when a large number of services must be delivered

over a broad region and at various places.

• Fog computing is required for devices that are subjected to

demanding calculations and processing.

• Fog computing is utilized in IoT devices (for example, the Car-

to-Car Consortium in Europe), Devices with Sensors and
Cameras (IIoT-Industrial Internet of Things), and other
applications.
When to use fog computing?

• It is used when only selected data is required to send to the

cloud. This selected data is chosen for long-term storage and is
less frequently accessed by the host.

• It is used when the data should be analyzed within a fraction of

seconds i.e Latency should be low.

• It is used whenever a large number of services need to be

provided over a large area at different geographical locations.

• Devices that are subjected to rigorous computations and

processings must use fog computing.
Advantages of Fog Computing

• This approach reduces the amount of data that needs to be sent

to the cloud.

• Since the distance to be traveled by the data is reduced, it

results in saving network bandwidth.

• Reduces the response time of the system.

• It improves the overall security of the system as the data resides

close to the host.

• It provides better privacy as industries can perform analysis on

their data locally.
Disadvantages of Fog Computing

• Congestion may occur between the host and the fog node due
to increased traffic (heavy data flow).

• Power consumption increases when another layer is placed

between the host and the cloud.

• Scheduling tasks between host and fog nodes along with fog
nodes and the cloud is difficult.

• Data management becomes tedious as along with the data

stored and computed, the transmission of data involves
encryption-decryption too which in turn release data.
Fog Computing Architecture
Fog Computing Components

1. Physical & virtual nodes (end devices)

• End devices serve as the points of contact to the real world, be

it application servers, edge routers, end devices such as mobile
phones and smartwatches, or sensors.

• These devices are data generators and can span a large

spectrum of technology.

• This means they may have varying storage and processing

capacities and different underlying software and hardware.
Fog Computing Components

2. Fog nodes
• Fog nodes are independent devices that pick up the generated
information.
• Fog nodes fall under three categories: fog devices, fog servers,
and gateways.
• These devices store necessary data while fog servers also
compute this data to decide the course of action.
• Fog devices are usually linked to fog servers.
• Fog gateways redirect the information between the various fog
devices and servers.
• This layer is important because it governs the speed of
processing and the flow of information.
• Setting up fog nodes requires knowledge of varied hardware
configurations, the devices they directly control, and network
connectivity.
Fog Computing Components

3. Monitoring services
• Monitoring services usually include application programming
interfaces (APIs) that keep track of the system’s performance
and resource availability.

• Monitoring systems ensure that all end devices and fog nodes
are up and communication isn’t stalled.

• Sometimes, waiting for a node to free up may be more

expensive than hitting the cloud server.

• The monitor takes care of such scenarios. Monitors can be used

to audit the current system and predict future resource
requirements based on usage.
Fog Computing Components

4. Data processors
• Data processors are programs that run on fog nodes. They filter,
trim, and sometimes even reconstruct faulty data that flows
from end devices.
• Data processors are in charge of deciding what to do with the
data — whether it should be stored locally on a fog server or
sent for long-term storage in the cloud.
• Information from varied sources is homogenized for easy
transportation and communication by these processors.
• This is done by exposing a uniform and programmable
interface to the other components in the system.
• Some processors are intelligent enough to fill the information
based on historical data if one or more sensors fail.
• This prevents any kind of application failure.
Fog Computing Components

5. Resource manager
• Fog computing consists of independent nodes that must work in
a synchronized manner.
• The resource manager allocates and deallocates resources to
various nodes and schedules data transfer between nodes and
the cloud.
• It also takes care of data backup, ensuring zero data loss.
• Since fog components take up some of the SLA commitments
of the cloud, high availability is a must.
• The resource manager works with the monitor to determine
when and where the demand is high.
• This ensures that there is no redundancy of data as well as fog
servers.
Fog Computing Components

6. Security tools
• Since fog components directly interact with raw data sources,
security must be built into the system even at the ground level.
•
• Encryption is a must since all communication tends to happen
over wireless networks.

• End users directly ask the fog nodes for data in some cases.

• As such, user and access management is part of the security

efforts in fog computing.
Fog Computing Components

7. Applications
• Applications provide actual services to end-users.

• They use the data provided by the fog computing system to

provide quality service while ensuring cost-effectiveness.

• It is important to note that these components must be governed

by an abstraction layer that exposes a common interface and a
common set of protocols for communication.

• This is usually achieved using web services such as APIs.

CHARACTERISTICS OF THE FOG COMPUTING

Heterogeneity: Fog Computing is a highly virtualized platform that yields

compute, storage, and networking services between end devices and traditional
Cloud Computing Data Centers, typically, but not elite located at the edge of
network. Compute, storage, and networking resources are the building blocks of
both the Cloud and the Fog.

Edge location: The origins of the Fog can be traced to early proposals to
support endpoints with rich services at the edge of the network, including
applications with low latency requirements (e.g. gaming, video streaming,
augmented reality.

Geographical distribution: In sharp contrast to the more centralized Cloud,

the services and applications targeted by the Fog demand widely distributed
deployments. The Fog, will play an active role in delivering high quality
streaming to moving vehicles, through proxies along highways and tracks.
CHARACTERISTICS OF THE FOG COMPUTING

Large-scale sensor networks: To monitor the environment and the Smart

Grid are other examples of inherently distributed systems, requiring distributed
computing and storage resources.

Very large number of nodes, as a consequence of the wide geo-distribution,

as evidenced in sensor networks in general and the Smart Grid in particular.

Support for mobility. It is essential for many Fog applications to

communicate directly with mobile devices, and therefore support mobility
techniques, such as the LISP protocol, that decouple host identity from location
identity, and require a distributed directory system.
CHARACTERISTICS OF THE FOG COMPUTING

Real-time interactions. Important Fog applications involve real-time

interactions rather than batch processing. Interoperability and federation.
Seamless support of certain services (streaming is a good example) requires the
cooperation of providers. Hence, Fog components must be able to interoperate,
and services must be federated across domains.
CHARACTERISTICS OF THE FOG COMPUTING

Real-time interactions. Important Fog applications involve real-time

interactions rather than batch processing. Interoperability and federation.
Seamless support of certain services (streaming is a good example) requires the
cooperation of providers. Hence, Fog components must be able to interoperate,
and services must be federated across domains.
Major Issues with Fog Computing

Authentication and Trust issues

• Authentication is one of the most concerning issues of fog computing since
these services are offered at a large scale.

• Fog service providers can be different parties like cloud service providers,
internet service providers, and end-users.

• This flexibility complicates the whole structure and trust situation of fog. A
rouge fog node is a fog device which pretends to be legal and coaxes end
user to connect to it.

• Once a user connects to it, it can manipulate the signals coming to and from
the user to the cloud and can easily launch attacks.
Major Issues with Fog Computing

Privacy
• Privacy concern is always there when there are many networks involved.
Since fog computing is based on wireless technology, there is a huge
concern regarding network privacy.

• There are so many fog nodes that each end-user is accessible to them and
because of this more sensitive information passes from end-users to the fog
nodes.

Security
Fog computing security issues arise as there are many devices connected to fog
nodes and at different gateways.

Each device has a different IP address, and any hacker can fake your IP address
to gain access to your personal information that is stored in that particular fog
node.
Major Issues with Fog Computing

Fog Servers
• The right placement of fog servers should be there so that it can deliver its
maximum service.

• The company should analyze the demand and work done by the fog node
before placing it will help in reducing the maintenance cost.

Energy consumption
• Energy consumption is very high in fog computing as the number of fog
nodes present in the fog environment are high and require energy to work.

• Companies should try to minimize the energy requirement by the fog nodes
so that they should become more energy-efficient and save costs.
Fog Architecture for smart cities
Fog Architecture for Healthcare
Fog Architecture for Vehicle Connections
Thank you