International Journal of Electrical and Electronic Engineering & Telecommunications,
Journal Year:
2024,
Volume and Issue:
13(3), P. 184 - 199
Published: Jan. 1, 2024
The
data
generated
by
the
IoT
needs
a
powerful
platform
such
as
cloud
computing
for
processing.
However,
faces
challenges
when
dealing
with
various
types
of
resources,
high
delay,
and
cost,
this
represents
substantial
challenge
in
scheduling
tasks.
Therefore,
need
appeared
to
introduce
concept
fog.
To
address
these
limitations,
optimization
algorithms
PSO
were
used.
In
traditional
PSO,
all
particles
swarm
are
influenced
single
global
best
particle
(Gbest),
if
it
becomes
stuck
local
optimum,
will
move
closer
it,
thus,
may
easily
get
trapped
premature
convergence.
This
paper
proposed
an
adaptive
cloud-fog
integrated
approach
based
on
modified
called
Optimized
Leader
(PSO-OL).
These
modifications
four
stages:
Firstly,
method
ensure
diversity
initialization
phase
is
introduced.
Secondly,
reduce
chance
population
getting
farthest-best
Third,
addition
primary
Gbest,
second
Gbest
different
good
presented
explore
multiple
promising
regions.
Finally
new
crossover
operator
optimized
leader.
PSO-OL
was
evaluated
results
show
effectiveness
enhanced
leader
40%
farthest-best,
45%
second-Gbest
compared
standard
where
outperforms
other
minimizing
makespan
34%,
cost
14%,
increasing
throughput
75%,
comparison
existing
load
balancing
methods:
RR,
BLA,
MPSO,
ETS,
TCaS.
Sensors,
Journal Year:
2023,
Volume and Issue:
23(5), P. 2445 - 2445
Published: Feb. 22, 2023
Cloud-fog
computing
is
a
wide
range
of
service
environments
created
to
provide
quick,
flexible
services
customers,
and
the
phenomenal
growth
Internet
Things
(IoT)
has
produced
an
immense
amount
data
on
daily
basis.
To
complete
tasks
meet
service-level
agreement
(SLA)
commitments,
provider
assigns
appropriate
resources
employs
scheduling
techniques
efficiently
manage
execution
received
IoT
in
fog
or
cloud
systems.
The
effectiveness
directly
impacted
by
some
other
important
criteria,
such
as
energy
usage
cost,
which
are
not
taken
into
account
many
existing
methodologies.
resolve
aforementioned
problems,
effective
algorithm
required
schedule
heterogeneous
workload
enhance
quality
(QoS).
Therefore,
nature-inspired
multi-objective
task
called
electric
earthworm
optimization
(EEOA)
proposed
this
paper
for
requests
cloud-fog
framework.
This
method
was
using
combination
(EOA)
fish
(EFO)
improve
EFO's
potential
be
exploited
while
looking
best
solution
problem
at
hand.
Concerning
time,
makespan,
consumption,
suggested
technique's
performance
assessed
significant
instances
real-world
workloads
CEA-CURIE
HPC2N.
Based
simulation
results,
our
approach
improves
efficiency
89%,
consumption
94%,
total
cost
87%
over
algorithms
scenarios
considered
different
benchmarks.
Detailed
simulations
demonstrate
that
provides
superior
scheme
with
better
results
than
techniques.
PeerJ Computer Science,
Journal Year:
2024,
Volume and Issue:
10, P. e2128 - e2128
Published: June 17, 2024
Fog
computing
has
emerged
as
a
prospective
paradigm
to
address
the
computational
requirements
of
IoT
applications,
extending
capabilities
cloud
network
edge.
Task
scheduling
is
pivotal
in
enhancing
energy
efficiency,
optimizing
resource
utilization
and
ensuring
timely
execution
tasks
within
fog
environments.
This
article
presents
comprehensive
review
advancements
task
methodologies
for
systems,
covering
priority-based,
greedy
heuristics,
metaheuristics,
learning-based,
hybrid
nature-inspired
heuristic
approaches.
Through
systematic
analysis
relevant
literature,
we
highlight
strengths
limitations
each
approach
identify
key
challenges
facing
scheduling,
including
dynamic
environments,
heterogeneity,
scalability,
constraints,
security
concerns,
algorithm
transparency.
Furthermore,
propose
future
research
directions
these
challenges,
integration
machine
learning
techniques
real-time
adaptation,
leveraging
federated
collaborative
developing
resource-aware
energy-efficient
algorithms,
incorporating
security-aware
techniques,
advancing
explainable
AI
methodologies.
By
addressing
pursuing
directions,
aim
facilitate
development
more
robust,
adaptable,
efficient
task-scheduling
solutions
ultimately
fostering
trust,
security,
sustainability
systems
facilitating
their
widespread
adoption
across
diverse
applications
domains.
IEEE Access,
Journal Year:
2023,
Volume and Issue:
11, P. 143417 - 143445
Published: Jan. 1, 2023
Efficient
task
scheduling
on
the
cloud
is
critical
for
optimal
utilization
of
resources
in
data
centers.
It
became
even
more
challenging
with
emergence
5G
and
IoT
applications
that
generate
massive
number
tasks
stringent
latency
requirements.
This
gives
birth
to
fog/edge
computing
-
a
complementary
layer
cloud.
Tasks
fog
can
be
reduced
as
processing
network
done
closer
end
devices
users,
but
every
cannot
scheduled
due
limited
availability.
Conventional
algorithms
often
fail
exploit
heterogeneous
resources;
therefore,
specially
designed
well-tuned
are
desired
achieving
better
quality
service.
In
this
study,
state-of-the-art
environments
investigated
diverse
set
dimensions.
Among
relevant
studies
published
between
2018–2022
indexed
Web-of-Science
(WOS),
SCOPUS,
Google
Scholar
databases,
eighteen
selected
both
domains
from
WOS
Scopus,
while
seventeen
chosen
Scholar.
Thus,
total
106
included
survey
detail
investigation.
The
broadly
classified
into
three
categories
such
heuristic,
meta-heuristic,
hybrid
meta-heuristic
followed
by
detailed
analysis.
has
been
observed
most
dynamic
non-preemptive
nature,
higher
fraction
independent
comparison
bag
workflows.
Similarly,
97%
focus
multiple
objectives
68%
techniques
non-deterministic.
Further,
twenty
different
identified
makespan,
resource
utilization,
delay,
load
balancing,
energy
consumption
significant
metrics.
evaluation
methods
including
simulations
(51%),
real
experiments
(4%),
analytical
equations
(2%),
datasets
(43%)
etc.
surveyed.
At
end,
open
issues,
challenges,
future
directions
argued.
