Computers, materials & continua/Computers, materials & continua (Print),
Journal Year:
2024,
Volume and Issue:
80(1), P. 747 - 800
Published: Jan. 1, 2024
Software
Defined
Networking
(SDN)
is
programmable
by
separation
of
forwarding
control
through
the
centralization
controller.
The
controller
plays
role
'brain'
that
dictates
intelligent
part
SDN
technology.
Various
versions
controllers
exist
as
a
response
to
diverse
demands
and
functions
expected
them.
There
are
several
available
in
open
market
besides
large
number
commercial
controllers;
some
developed
meet
carrier-grade
service
levels
one
recent
trends
open-source
Open
Network
Operating
System
(ONOS).
This
paper
presents
comparative
study
between
source
controllers,
which
known
Controller
Platform
(NOX),
Python-based
(POX),
component-based
framework
(Ryu),
Java-based
OpenFlow
(Floodlight),
OpenDayLight
(ODL)
ONOS.
discussion
further
extended
into
ONOS
architecture,
well
as,
evolution
controllers.
article
will
review
use
cases
based
on
application
deployments.
Moreover,
opportunities
challenges
be
discussed,
exploring
for
future
real-world
deployments,
unique
features
identifying
suitable
choice
providers.
In
addition,
we
attempt
provide
answers
critical
questions
relating
implications
nature
regarding
vendor
lock-in,
interoperability,
standards
compliance,
Similarly,
organizations
using
highlighted
how
community
contributes
development
Furthermore,
faced
projects,
considerations
when
choosing
an
underscored.
Then
Artificial
Intelligence
(AI)
Machine
Learning
(ML)
light
research
indicated.
limitations
associated
with
deploying
production
networks,
can
they
mitigated,
finally
handle
network
security
ensure
configurations
policies
robust
resilient
presented.
Potential
deployment
outlined
conclude
article.
Sensors,
Journal Year:
2025,
Volume and Issue:
25(6), P. 1905 - 1905
Published: March 19, 2025
The
Internet
has
been
vulnerable
to
several
attacks
as
it
expanded,
including
spoofing,
viruses,
malicious
code
attacks,
and
Distributed
Denial
of
Service
(DDoS).
three
main
types
most
frequently
reported
in
the
current
period
are
DoS
DDoS
attacks.
Advanced
too
complex
for
traditional
security
solutions,
such
intrusion
detection
systems
firewalls,
detect.
combination
machine
learning
methods
with
AI-based
led
introduction
novel
attack
systems.
Due
their
remarkable
performance,
models,
particular,
have
essential
identifying
However,
there
is
a
considerable
gap
work
on
real-time
This
study
uses
Mininet
POX
Controller
simulate
an
environment
detect
settings.
CICDDoS2019
dataset
identifies
classifies
simulated
environment.
In
addition,
virtual
software-defined
network
(SDN)
used
collect
information
from
surrounding
area.
When
occurs,
pre-trained
models
analyze
traffic
predict
real-time.
performance
proposed
methodology
evaluated
based
two
metrics:
accuracy
time.
results
reveal
that
model
achieves
99%
within
1
s
FUDMA Journal of Sciences,
Journal Year:
2025,
Volume and Issue:
9(3), P. 154 - 163
Published: March 31, 2025
In
today's
ever-evolving
networking
landscape,
Software
Defined
Networking
(SDN)
has
emerged
as
a
paradigm-shifting
technology
that
promises
greater
flexibility,
agility,
and
control
over
network
infrastructures.
However,
how
easy
are
the
configuration,
extensibility,
programmability
of
these
SDN
controllers,
considering
practical
implications
for
administrators
developers?
Despite
growing
adoption
SDN,
there
is
limited
research
on
comparative
performance
controllers
across
multiple
simulation
environments.
This
study
aims
to
explore
implementation
evaluation
different
Controllers
within
diverse
simulated
Prominent
such
POX
Faucet
meticulously
configured
deployed
in
environments
created
using
GNS3
(Graphical
Network
Simulator-3),
OPNET
(Optimized
Engineering
Tools),
NS3
(Network
OMNET++
(Objective
Modular
Testbed
C++)
MININET
platforms.
Furthermore.
employs
range
metrics
like
controller
latency,
throughput,
packet
loss,
CPU
(Central
Processing
Unit)
memory
utilization
assess
efficacy
efficiency
each
Controller.
The
Results
indicated
Mininet
provided
lowest
whereas
demonstrated
better
scalability
large-scale
networks.
NS3,
though
useful
design
visualization,
exhibited
higher
might
limit
its
simulations.
While
offered
balanced
resource
utilization,
making
it
suitable
choice
prioritizes
realistic
modelling,
OMNET++,
other
hand,
moderate
with
efficient
simulations
requiring
balance
between
performance...
Computers,
Journal Year:
2024,
Volume and Issue:
13(4), P. 85 - 85
Published: March 25, 2024
An
alternative
networking
approach
called
Software
Defined
Networking
(SDN)
enables
dynamic,
programmatically
efficient
network
construction,
hence
enhancing
performance.
It
splits
a
traditional
into
centralized
control
plane
and
configurable
data
plane.
Because
the
core
component
overseeing
every
action
is
controller
in
plane,
which
may
contain
one
or
more
controllers
thought
of
as
brains
SDN
network,
functionality
performance
are
crucial
to
achieve
optimal
performances.
There
much
research
available
existing
literature.
Nevertheless,
no
qualitative
comparison
study
OpenFlow-enabled
distributed
but
logically
exists.
This
paper
includes
quantitative
investigation
several
custom
scenarios
using
Mininet,
well
thorough
them.
More
precisely,
we
give
evaluation
their
attributes
classify
categorize
13
according
capabilities.
Additionally,
offer
comprehensive
emulation
tool,
Mininet-based
assessment,
this
study.
Using
six
metrics—bandwidth,
round-trip
time,
delay,
jitter,
packet
loss,
throughput—this
work
also
assesses
five
within
two
(uniform
non-uniform
host
distribution).
Our
analysis
reveals
that
Ryu
outperforms
OpenDayLight
terms
latency,
while
performs
throughput,
bandwidth,
jitter.
Throughout
entire
experiment,
HyperFlow
ONOS
performed
worst
all
metrics.
Finally,
discuss
detailed
findings
on
These
experimental
results
provide
decision-making
guidelines
when
selecting
controller.
