Uploaded by JASON MARANDI GAU-C-15/059

ReportEdgeComputing

Seminar Report
on
Edge Computing
(Subject Code: MCO2S01)
By
JASON MARANDI
M.Tech 1st Sem
Roll No: 32013211
COMPUTER ENGINEERING(Cyber Security)
December 7,2020
DEPARTMENT OF COMPUTER ENGINEERING
NATIONAL INSTITUTE OF TECHNOLOGY
KURUKSHETRA,HARYANA - 136119
ABSTRACT
The centralized cloud computing infrastructure provides storage and processing capabilities
for large volume of data produced by different devices. These data need to be transferred
through communication channel from the place it is generated to the Cloud. Thus it
result in high bandwidth bandwidth requirement and latency. Thus the concept of Edge
Computing arises which is, to bring the computation of the data closer to the point of
its origin. In many real time applications such as financial transactions, health care,
gaming, autonomous vehicles etc. a very low levels of latency can be tolerated. Thus to
reduce latency and and save bandwidth Edge Computation can be used at the edge of the
network. Some of the implementations of edge computing such as Fog Computing, Mobile
Edge Computing(MEC) and Cloudlets has been discussed in this report.
Keywords: Cloud Computing, Edge Computing, Fog Computing, Mobile Edge Computing, Cloudlet.
i
Contents
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
i
List of Figures
iii
List of Tables
iv
1 Introduction
1.1 Origin and Background . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.2 Concept . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
1
2
2 Cloud Computing
3
3 Edge Computing
3.1 Proprties of Edge Computing [3] . . . . . . . . . . . . . . . . . . . . . . . .
3.2 Layered components of Edge Computing . . . . . . . . . . . . . . . . . . .
5
6
7
4 Implementations of Edge Computing
4.1 Mobile Edge Computing(MEC) . . .
4.2 Fog Computing . . . . . . . . . . . .
4.3 Cloudlets . . . . . . . . . . . . . . .
4.4 Comparison of Inmplementations . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
9
9
10
10
11
5 Use Cases
12
6 Challenges
13
7 Conclusion
14
Bibliography
15
ii
List of Figures
2.1
Client-server model diagram . . . . . . . . . . . . . . . . . . . . . . . . . .
3
3.1
3.2
3.3
Edge in client-server model . . . . . . . . . . . . . . . . . . . . . . . . . . .
Edge Computing model diagram . . . . . . . . . . . . . . . . . . . . . . . .
General overview of edge computing components . . . . . . . . . . . . . . .
5
6
7
4.1
4.2
4.3
Mobile Edge Computing Infrastructure . . . . . . . . . . . . . . . . . . . .
Fog Computing Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . .
Fog Computing Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . .
9
10
10
iii
List of Tables
4.1
Comparision of Edge Computing Implementations . . . . . . . . . . . . . .
iv
11
Chapter 1
Introduction
The rapid development of technology has seen an increase in the number of devices in
recent years. The rise in the number of devices means the increase in the volume of
data generated. And this will continue to rise as the devices connected to the internet
increases. Managing and processing this large volume is data is a challenging task. Cloud
Computing being a resource rich solution in terms of storage and computation can deal
with such huge volume of data. But the problem rises in transmission of data. Sending
all the data generated by the edge devices to a centralized data center or to the cloud
causes bandwidth and latency issues. . In some application scenarios that require realtime feedback, such as traffic monitoring, financial transaction, autonomous vehicles etc.,
cloud computing will not be able to meet business real-time requirements. Here the Edge
Computing comes to play. Edge computing offers a more efficient alternative i.e. data is
to be processed and analyzed closer to the point where it’s created. Therefore data does
not have to traverse over a network to a cloud or data center to be processed, and hence
we can save bandwidth and decrease the latency.
1.1
Origin and Background
The roots of edge computing reach back to the late 1990s, when Akamai introduced content
delivery networks (CDNs) to accelerate web performance. A CDN uses nodes at the edge
close to users to prefetch and cache web content. CDNs are especially valuable for video
content, because the bandwidth savings from caching can be substantial.
In 1997, Brian Noble and his colleagues first demonstrated edge computing’s potential
value to mobile computing. They showed how speech recognition could be implemented
with acceptable performance on a resource-limited mobile device by offloading computation
to a nearby server.
Edge computing generalizes and extends the CDN concept by leveraging cloud computing
infrastructure. As with CDNs, the proximity of cloudlets to end users is crucial. However,
instead of being limited to caching web content, a cloudlet can run arbitrary code just as
in cloud computing. This code is typically encapsulated in a virtual machine (VM) or a
1
lighter-weight container for isolation, safety, resource management, and metering.
1.2
Concept
The increase of devices at the edge of the network is producing a massive amount of
data to be computed at data centers, pushing network bandwidth requirements to the
limit. Despite the improvements of network technology, data centers cannot guarantee
acceptable transfer rates and response times, which could be a critical requirement for
many applications. Furthermore, devices at the edge constantly consume data coming
from the cloud, forcing companies to build content delivery networks to decentralize data
and service provisioning, leveraging physical proximity to the end user.
In a similar way, the aim of Edge Computing is to move the computation away from
data centers towards the edge of the network, exploiting smart objects, mobile phones or
network gateways to perform tasks and provide services on behalf of the cloud. By moving
services to the edge, it is possible to provide content caching, service delivery, storage and
computation, resulting in better response times and transfer rates.
2
Chapter 2
Cloud Computing
It is the practice of using a network of remote servers hosted on the internet to store,
manage, and process data, rather than a local server or a personal computer. Cloud also
know as data center. A data center traditionally refers to server hardware on your premises
to store and access data through your local network.
Figure 2.1: Client-server model diagram
In Figure 2.1 we can see the view of Cloud Computing model.In these types of model
most of the work is done by the central server and the clients only interact as needed..
If you’ve ever used an email client or a web browser, this should be familiar. While you
may compose your email or read a web page on your device, the server is what handles
the background routing and formatting of your emails or the rendering of the web content,
3
usually coupled with resources pulled from a database.
Services provided by Cloud Computing:
ˆ Infrastructure as a Service (IaaS)
ˆ Platform as a Service (PaaS)
ˆ Software as a Service (SaaS)
Cloud is capable of dealing with large volume data which require storage and computation power to process it. Now, to use the services of cloud computing the data must
first be transferred to the cloud which is generally far from where the data is generated
or needed. Nowadays data generated by edge device are generally huge in volume. Thus,
transmission of such large volume of data over the internet give rise to issue of bandwidth
and latency.
Therefore centralized cloud computing suffers from bandwidth issue and latency.
Introducing Edge Computing solve the issue of latency and save bandwidth, as computation is done locally at the edge of network.
4
Chapter 3
Edge Computing
To define Edge Computing let us first define what is an edge. An Edge is any computing and network resources along the path between data sources and cloud data centers as
shown in Figure 2.2.
Figure 3.1: Edge in client-server model
Thus Edge Computing can be defined broadly as all computing outside the cloud
happening at the edge of the network. Generally near the source of the data. More formal
definition of edge computing is ”Edge computing is a distributed computing paradigm that
brings computation and data storage closer to the location where it is needed, to improve
response times and save bandwidth”. Edge can relate to data processing as well as local
processing of the real time data. The various edge components that can be counted upon are
Data processing and storing. Edge computing enables data stream acceleration, including
real-time data processing without latency. It allows smart applications and devices to
respond to data almost instantaneously as it’s being created, eliminating lag time.
Edge computing allows for efficient data processing in that large amounts of data can be
processed near the source, reducing Internet bandwidth usage. This both eliminates costs
and ensures that applications can be used effectively in remote locations. In addition, the
ability to process data without ever putting it into a public cloud adds a useful property
for securing of sensitive data.
5
Figure 3.2: Edge Computing model diagram
3.1
Proprties of Edge Computing [3]
1. Proximity: In Edge computing, computation resources and services are available in
the proximity of the users that can improve their experience.
2. Dense Geographical Distribution: Edge computing brings the Cloud services
closer to the user by deploying numerous computing platforms at the edge the network. The platforms/devices are densely distributed across the geographical region.
3. Mobility Support: As the number of mobile devices is rapidly growing, Edge
computing also supports mobility, such as the Locator ID Separation Protocol (LISP),
to communicate directly with mobile devices.
4. Location Awareness: The location awareness attribute of Edge computing allows
the mobile users to access services from that Edge server closest to their physical
location.
5. Low Latency: The low latency property of Edge computing enables the users to
execute their resource-intensive and delay-sensitive applications on the resource-rich
Edge devices (e.g. router, access point, base station, or dedicated server).
Some terms related to Edge Computing:
6
ˆ Edge Node:An edge node is a generic way of referring to any edge device, edge
server, or edge gateway on which edge computing can be performed.
ˆ Edge Devices:Any device which produces data like sensors, or any embedded systems through which data is collected and delivered.
ˆ Edge Clusters/Servers:An edge cluster/server is a general-purpose IT computer
that is located in a remote operations facility and act as a sever.
ˆ Edge Gateways: An edge gateway are devices that serves as a network entry point
for other edge devices that are connected to the cloud services.
ˆ Edge cloud: Edge clouds are micro-data centers that can store, analyze, and process
data near to the edge devices.
3.2
Layered components of Edge Computing
Edge computing involves multiple layers of abstraction that organize tasks and resources
to allow for heterogeneous, distributed tasks to be concurrently processed on heterogenous, distributed hardware in a managed fashion. This may vary from implementation to
implementation as Edge Computing is a new and emerging field.
Figure 3.3: General overview of edge computing components
1. Distributed data and infrastructure: This layer deals with how devices and data
sources are organized into networks and hierarchies that structure how they can and
will be used by the layers above them.This abstraction can reorganize, for instance,
devices into contained groups by device class, department of use, or ownership, with
needing to relocate the physical device to allow for segmentation to take place at the
network level.
7
2. Data transport and integration service bus: The data transport and information service bus, or data layer for short, handles direct communication paths between
hosts in your distributed infrastructure. These paths can serve a couple of purposes:
to handle framework control messages, as in the integration service bus, or to pass
data messages, as in the data transport.
3. Semantic data layer: A semantic data layer is a layer of abstraction that provides
a consistent way of interpreting data. It deals with what kind of data is being used.
4. Application layer: The application layer is the most visible to the user of an
edge computing framework. In this layer, one or more functions, or small pieces of
computation to be performed, are distributed over the framework, either statically,
automatically, or dynamically (if efficient resource scheduling is provided). These
functions perform predefined tasks, usually on incoming data, and, if necessary, provide the results of their computation to other functions. End users can either use
existing functions compatible with their chosen edge computing framework or write
their own as the software allows.
8
Chapter 4
Implementations of Edge Computing
Edge Computing is implemented in following 3 ways [5]:
4.1
Mobile Edge Computing(MEC)
MEC can be defined as an implementation of Edge Computing to bring computational
and storage capacities to the edge of the network within the Radio Access Network to
reduce latency and improve context awareness. The MEC nodes or servers are usually
co-located with the Radio Network Controller or a macro base-station. Figure 4.1 depicts
the infrastructure of MEC.
Figure 4.1: Mobile Edge Computing Infrastructure
9
4.2
Fog Computing
The Fog Computing implementation is a decentralized Computing infrastructure based on
Fog Computing Nodes (FCNs) placed at any point of the architecture between the end
devices and the cloud. The FCNs are heterogeneous in nature and thus can be based on
different kinds of elements including devices such as gateways, switches, routers, servers,
etc.
Figure 4.2: Fog Computing Infrastructure
4.3
Cloudlets
A Cloudlet can be defined as a trusted cluster of computers, well connected to the Internet,
with resources available to use for nearby mobile devices. A cloudlet is viewed as a “data
center in a box” whose goal is to “bring the cloud closer”. They have lighter computing
capabilities than clouds.
Figure 4.3: Fog Computing Infrastructure
10
4.4
Comparison of Inmplementations
In table 4.1 comparison of edge computing implementations is given based on how they
work,where they are located and in which devices they could be implemented.
EC Elements
Node Device
Node location
Software Architecture
Context awareness
Proximity
Access Mechanisms
Internode Communication
Fog Computing
Mobile Edge Computing
Routers,Switches,Access Servers running in
points,Gateways
base stations
Varying between End Radio Network ConDevices and Cloud
troller/Macro
Base
Station
Fog Abstraction Layer Mobile Orchestrator
based
based
Medium
High
One or Multiple Hops One Hop
Bluetooth,WiMobile Networks
Fi,Mobile
Networks,Wired Cable
Supported
Partial
Cloudlets
Data Center in a box
Local/Outdoor installation
Cloudlet Agent based
Low
One Hop
Wi-Fi,Wired Cable
Partial
Table 4.1: Comparision of Edge Computing Implementations
11
Chapter 5
Use Cases
Following are few use cases of Edge Computing:
1. Automotive: While driving all the information coming through sound, vision, movement, and all our other senses are processed to give us a sense of the environment
and what we need to do to maintain the vehicle’s desired course. In autonomous
vehicles, this information is ingested through sensors, and needs to be processed on
the fly too. The processing should be close to vehicle for making real time decision.
2. Manufacturing: With Industry 4.0 initiatives moving from the pilot stages to mass
deployments, industrial leaders are starting to experience speed limitations in data
transmissions. Indeed, to support more advanced networks of Industrial IoT devices,
and ultimately – fully automated factories, a greater level of resilience, speed and
bandwidth is required. Again, edge architecture could deliver that.
3. Augmented Reality (AR) and Virtual Reality (VR): Edge computing gaming
applications also multiplied in recent years due to the rising demand for AR/VR
gaming experiences. The current roadblock to wider adoption, however, is high
gaming hardware costs. As VR apps require significant computational power and
storage for visualizations, gamers need to purchase expensive head-mounted displays,
consoles, and PC rigs. Edge computing could majorly democratize the costs of VR
and spur its adoption in other industries such as education, hospitality, or travel.
4. Health Care: In many hospitals, there are hundreds of systems generating quite
a large amount of data in different formats and levels of quality. Each medical
practitioner has his or her own ways to label the data, making it difficult to use the
information produced by different practitioners. Edge computing makes it possible
to leverage collaborative machine learning to use all related data on multiple sites.
And because such highly personal data can’t ever leave the hospital due to privacy
concerns, this type of analysis must happen at the edge. Also, by enabling edge
computing, crucial data can be transmitted from the ambulance to the hospital in
real time, saving time and arming emergency department teams with the knowledge
they need to save lives.
12
Chapter 6
Challenges
1. General Purpose Computing on Edge Nodes: Making general purpose edge
nodes that can do the computation. In [5] theory edge computing is incorporated in
devices like routers, gateway, base stations etc. but in practical this may decrease the
efficiency of the main purpose of that particular device. So general purpose devices
are needed.
2. Discovering Edge Nodes: Discovering resources and services in a distributed computing environment means finding appropriate edge nodes that can that can be leveraged in a decentralised cloud set up.
3. Partitioning and Offloading Tasks: In a distributed environment making use of
edge nodes for offloading computations poses the challenge of partitioning computational tasks efficiently among the edge nodes.
4. Uncompromising Quality-of-Service (QoS) and Experience (QoE): Quality
delivered by the edge nodes can be captured by QoS and quality delivered to the user
by QoE. For example if edge computing is implemented in a base station then the
services provided by base station should remain uncompromised after implementing
base station.
5. Using Edge Nodes Publicly and Securely: Edge nodes will be used publicly
so securing data in the edge computing environment is a challenge on its own. Right
from the devices that generate data, then in transmission channel while transmitting
the data to edge servers and also securing the edge servers.
13
Chapter 7
Conclusion
Edge Computing try to overcome the problem of latency and bandwidth caused in the
centralized cloud computing, by doing the computing at the edge of the network i.e. where
the data is generated and needed. Implementations of Edge Computing are in the form of
Mobile edge Computing(MEC), Fog Computing and Cloudlet, all carry the same concept
that is bringing computation of cloud to the edge of the network but they differ in some
set of features. Edge computing is considered as new technology and there is a lack of
standardization.
14
Bibliography
[1] K. Dolui and S. K. Datta, ”Comparison of edge computing implementations: Fog computing, cloudlet and mobile edge computing,” 2017 Global Internet of Things Summit
(GIoTS), Geneva, 2017, pp. 1-6, doi: 10.1109/GIOTS.2017.8016213.
[2] M. Satyanarayanan, ”The Emergence of Edge Computing,” in Computer, vol. 50, no.
1, pp. 30-39, Jan. 2017, doi: 10.1109/MC.2017.9.
[3] Wazir Zada Khana , Ejaz Ahmedb , Saqib Hakakb , Ibrar Yaqoobc , Arif Ahmedd
(2019). Edge computing: A survey. Future Generation Computer Systems. 97.
10.1016/j.future.2019.02.05.
[4] Blesson Varghese, Nan Wang, Sakil Barbhuiya, Peter Kilpatrick and Dimitrios S.
Nikolopoulos. (2016). Challenges and Opportunities in Edge Computing.
[5] K. Dolui and S. K. Datta, ”Comparison of edge computing implementations: Fog computing, cloudlet and mobile edge computing,” 2017 Global Internet of Things Summit
(GIoTS), Geneva, 2017, pp. 1-6, doi: 10.1109/GIOTS.2017.8016213.
[6] Making Sense of Edge Computing by Cody Bumgardner and Caylin Hickey.
https://www.manning.com/books/making-sense-of-edge-computing
15