Heliyon,
Год журнала:
2024,
Номер
10(19), С. e37912 - e37912
Опубликована: Сен. 13, 2024
The
convenience
and
cost-effectiveness
offered
by
cloud
computing
have
attracted
a
large
customer
base.
In
environment,
the
inclusion
of
concept
virtualization
requires
careful
management
resource
utilization
energy
consumption.
With
rapidly
increasing
consumer
base
data
centers,
it
faces
an
overwhelming
influx
Virtual
Machine
(VM)
requests.
technology,
mapping
these
requests
onto
actual
hardware
is
known
as
VM
placement
which
significant
area
research.
article
presents
Dragonfly
Algorithm
integrated
with
Modified
Best
Fit
Decreasing
(DA-MBFD)
proposed
to
minimize
overall
power
consumption
migration
count.
DA-MBFD
uses
MBFD
for
ranking
VMs
based
on
their
requirement,
then
Minimization
Migration
(MM)
algorithm
hotspot
detection
followed
DA
optimize
replacement
from
overutilized
hosts.
compared
few
other
existing
techniques
show
its
efficiency.
comparative
analysis
against
E-ABC,
E-MBFD,
MBFD-MM
shows
%improvement
reflecting
reduction
in
8.21
%,
8.6
6.77
violations
service
level
agreement
9.25
6.98
%-7.86
%
number
migrations
6.65
8.92
7.02
respectively.
Research Square (Research Square),
Год журнала:
2024,
Номер
unknown
Опубликована: Янв. 31, 2024
Abstract
Maintaining
data
confidentiality
and
integrity
during
the
large
VM
migration
is
quite
challenging.
Simultaneously,
use
of
complex
encryption
or
steganography
for
managing
them
increases
time
overheads.
These
may
cause
loss.
The
transportation
VMs
further
consumes
significant
bandwidth
causes
page
faults.
However,
these
issues
aren't
dealt
with
in
modern
literature,
despite
many
research
attempts.
Moreover,
unlawful
intrusions
various
transmission
errors
make
matters
worse.
Hence,
this
work
proposes
an
efficient
technique
that
addresses
such
outstanding
a
unified
way.
suggested
solution
has
special
compression
method
reduces
big
sizes
to
53.9%,
new
enhance
integrity,
smart
split
stop
faults
as
well
lower
loss
0.0009%.
results
show
it
cuts
down
on
downtime
by
10%
more
than
existing
methods.
obtained
justify
its
efficiencies
over
other
ones
distinct
dimensions.
IEEE Transactions on Sustainable Computing,
Год журнала:
2024,
Номер
unknown, С. 1 - 13
Опубликована: Янв. 1, 2024
The
rapid
growth
and
widespread
adoption
of
cloud
computing
have
led
to
significant
electricity
costs
environmental
impacts.Traditional
approaches
that
rely
on
static
utilization
thresholds
are
ineffective
in
dynamic
environments,
simply
consolidating
virtual
machines
(VMs)
minimize
energy
does
not
necessarily
result
the
lowest
carbon
footprints.In
this
paper,
a
deep
reinforcement
learning
(DRL)
based
framework
called
CFWS
is
proposed
enhance
efficiency
renewable
sources
(RES)
supplied
data
centers
(DCs).CFWS
incorporates
an
adaptive
adjustment
method
TCN-MAD
by
evaluating
predicted
probability
physical
machine
(PM)
being
overloaded
prevent
unnecessary
VM
migrations
mitigate
service
level
agreement
(SLA)
violations
due
imbalanced
workload
distribution.Additionally,
introduces
novel
action
space
DRL
algorithm
representing
among
geo-distributed
as
flattened
indices
accelerate
its
execution
efficiency.Simulation
results
demonstrate
can
achieve
superior
optimization
footprints,
saving
5.67%
13.22%
brown
with
maximized
RES
utilization.Furthermore,
reduces
up
86.53%
maintains
SLA
within
suboptimal
time
comparison
state-of-art
algorithms.
Heliyon,
Год журнала:
2024,
Номер
10(19), С. e37912 - e37912
Опубликована: Сен. 13, 2024
The
convenience
and
cost-effectiveness
offered
by
cloud
computing
have
attracted
a
large
customer
base.
In
environment,
the
inclusion
of
concept
virtualization
requires
careful
management
resource
utilization
energy
consumption.
With
rapidly
increasing
consumer
base
data
centers,
it
faces
an
overwhelming
influx
Virtual
Machine
(VM)
requests.
technology,
mapping
these
requests
onto
actual
hardware
is
known
as
VM
placement
which
significant
area
research.
article
presents
Dragonfly
Algorithm
integrated
with
Modified
Best
Fit
Decreasing
(DA-MBFD)
proposed
to
minimize
overall
power
consumption
migration
count.
DA-MBFD
uses
MBFD
for
ranking
VMs
based
on
their
requirement,
then
Minimization
Migration
(MM)
algorithm
hotspot
detection
followed
DA
optimize
replacement
from
overutilized
hosts.
compared
few
other
existing
techniques
show
its
efficiency.
comparative
analysis
against
E-ABC,
E-MBFD,
MBFD-MM
shows
%improvement
reflecting
reduction
in
8.21
%,
8.6
6.77
violations
service
level
agreement
9.25
6.98
%-7.86
%
number
migrations
6.65
8.92
7.02
respectively.
