AN OVERVIEW ON
EDGE COMPUTING
MD BIDDUT HOSSAIN
e-mail: biddut2j8@[Link]
HTTP://[Link]/
LEARNING OBJECTIVE
In this study, the objective is to find the importance of computation
nearest the user network using edge computation (EC).
Cloud computing (CC) and its limitations
Overview of edge computing (EC) and relation on CC and EC.
Characterized and application of EC
Technologies used in EC
Architecture and components of EC
Challenges and future research
HTTP://[Link]/
INTRODUCTION
At present the devices around the world are more connected and moving
to internet-based services. But the communication on multiple devices,
gathering and sharing data across heterogeneous networks, variety of
mechanisms and protocols has an obstacle in the development of internet-
oriented applications.
This trend creates the concept of Cloud Computing.
[Link]
CLOUD COMPUTING
Cloud computing is a set of internet-based computing in which many
remote servers are connected to permit sharing of data-processing tasks,
centralized data storage, and online access to computer services or
resources.
The emergence of different new computing applications like smart
environments and virtual reality requires the availability of a large pool
of cloud resources and services becoming a limitation for applications
that essentially a real-time response.
[Link]
LIMITATIONS OF CLOUD COMPUTING
The cloud computing paradigm is unable to meet the requirements of low
latency, location awareness, and mobility support, real-time services and
data optimization.
How do overcome this limitation?
Edge Computing serves this purpose.
Edge computing is a mini-cloud and interplay with the cloud resources.
It transfers the utility services hosted in the centralized cloud data center
to decentralized geographical distribution edge network devices.
Real-time analysis of data at the level of local devices and edge nodes
reduces data flow and operating costs of using cloud services, traffic and
data transfer between the Edge and the cloud.
[Link]
EDGE COMPUTING
Edge computing (EC) has been emerging as one of the most promising
solutions regarding the cloud limitation, depending on the cooperation of
nodes by moving services directly to end devices and caching
information locally.
An edge computing paradigm that goals to extend the cloud resources
and services and enable them to be closer the edge of an enterprise’s
network.
EC is defined by different authors as a set of devices, sensors, computer
resources and computers that generate and collect data that are then sent
to cloud centers.
HTTP://[Link]/
COMPARISON BETWEEN CC AND EC
Edge Cloud
Reduce end-user response time Scalable
Low latency Big data processing
Advantage Distributed computation Unlimited storage capacity
Reduces bandwidth cost,
network traffic and storage.
Storage capacity is limited Centralized computation
EC needs proprietary network Response time is slow
Disadvantage Difficult to maintain the Generate network traffic
security of data. High latency
HTTP://[Link]/
APPLICATION OF EDGE COMPUTING
Internet of Things (IoT) smart
environment (Smart building,
Smart surveillance)
Virtual reality
Smart traffic, transportation
Smart health environment
Industrial automation
Agriculture
Online gaming
Real time big data analysis
HTTP://[Link]/
STRUCTURE OF EC
The Figure 1 components are described as
follows:
Devices and sensors: responsible to
produce and collect data. This group of
devices interacts directly with the end
user (sensors, smart phones, tablets,
smart bracelets or laptops).
Edge infrastructure: there are
distributed data centers to provide real-
time data processing, data visualization,
analytics, filtering, and optimizing. They
being located closer to end users, they
process, cache storage, and perform
calculations for a large volume of data.
Cloud: It provides a greater density of
computing, storage, networking
resources. Cloud servers host services for
HTTP://[Link]/
automatic learning, big data analysis and
business intelligence.
CONSIDERATION ISSUES OF EC
Reliability (Failure of individual node/ server/ end user, Lack of network
coverage, Failure of edge network and service platform)
Scalability (Adding a new point of service, Adding new services nodes,
Utilizing cloud interplay)
Security (Isolation of data paths and memory as multiple users share
edge network, Security of edge devices from malicious users)
Resource management (Resource allocation, Load balancing,
Monitoring)
Interoperability (Heterogeneous devices, protocols, and services)
HTTP://[Link]/
USED TECHNOLOGIES IN EC
Network functions virtualization
(NFV) can be applied to network
nodes to ensure security and
isolation of user domains.
Network functions virtualization
(NFV) and Software defined
networking (SDN) are potential
technologies that enable ease of
management of edge resources.
HTTP://[Link]/
CONCERNING ISSUES OF EC
Uninterrupted Communication: movement of users from one edge
network to different edge network coverage area interrupts
communication.
Privacy and Security: As the end user's data computed remotely so the
privacy and security are the major issues of end users.
Real time data processing: connectivity among many different devices
(sensors, smart phones, tablets, smart bracelets or laptops) creates in a
flood of data production. At real time to compute these data is a major
concern.
Standardization: Different vendors offer different edge devices as
service providers. Without common standardization it is impossible to
communicate between heterogeneous edge networks.
HTTP://[Link]/
FUTURE RESEARCH ISSUES
At the internet-based communication data security and response time are
the major concern of users.
Enhancement security system and ensure maximum utilization of
resources in Edge Computing.
Develop real time data processing method for load balancing and
increase the performance of EC.
HTTP://[Link]/
QUESTION
AND
ANSWER
HTTP://[Link]/
THANK YOU
HTTP://[Link]/