Renewable and Sustainable Energy Reviews,
Год журнала:
2022,
Номер
167, С. 112782 - 112782
Опубликована: Июль 27, 2022
Cloud
Computing
services
can
be
accessed
anytime,
anywhere
via
the
Internet.
The
overwhelming
growth
of
cloud
data
centers
over
past
decade
has
increased
their
costs
as
energy
demands
have
risen.
As
a
result,
higher
carbon
dioxide
emissions
and
other
greenhouse
gasses
are
putting
strain
on
our
ecosystem.
main
objective
this
study
is
to
reduce
power
consumption
in
computing
with
no
or
negligible
trade-offs
quality
service.
This
paper
presents
new
algorithm
called
efficiency
heuristic
using
virtual
machine
consolidation
minimize
high
cloud.
By
setting
two
thresholds,
hosts
classified
into
three
classes.
designed
model
reallocates
machines
from
one
physical
host
another
consumption.
results
proposed
been
obtained
terms
migrations,
performance
degradation
caused
by
migration,
service
level
agreement
violations,
execution
time,
showing
significant
improvement
state-of-the-art
techniques.
Software Practice and Experience,
Год журнала:
2021,
Номер
52(1), С. 194 - 235
Опубликована: Июнь 28, 2021
Abstract
Cloud
systems
have
become
an
essential
part
of
our
daily
lives
owing
to
various
Internet‐based
services.
Consequently,
their
energy
utilization
has
also
a
necessary
concern
in
cloud
computing
increasingly.
Live
migration,
including
several
virtual
machines
(VMs)
packed
on
minimal
physical
(PMs)
as
consolidation
(VMC)
technique,
is
approach
optimize
power
consumption.
In
this
article,
we
proposed
energy‐aware
method
for
the
VMC
problem,
which
called
(EVMC),
consumption
regarding
quality
service
guarantee,
comprises:
(1)
support
vector
machine
classification
based
rate
all
resource
PMs
that
used
PM
detection
terms
amount'
load;
(2)
modified
minimization
migration
VM
selection;
(3)
particle
swarm
optimization
implemented
placement.
Also,
evaluation
functional
requirements
presented
by
formal
and
non‐functional
simulation.
Finally,
contrast
standard
greedy
algorithms
such
best
fit
decreasing,
EVMC
decreases
active
VMs,
respectively,
30%,
50%
average.
it
more
efficient
30%
average,
resources
balance
degree
15%
average
cloud.
IEEE Access,
Год журнала:
2021,
Номер
9, С. 18625 - 18648
Опубликована: Янв. 1, 2021
Resource
allocation
is
an
important
problem
for
cloud
environments.
This
paper
introduces
energy-aware
combinatorial
auction-based
model
the
resource
in
clouds.
The
proposed
allows
users
of
a
to
submit
their
virtual
requests
as
bids
using
provided
bidding
language
which
complementarities
and
substitutabilities
among
those
resources
be
declared.
finds
most
profitable
mutually
satisfiable
set
winning
bids,
corresponding
while
considering
placement
available
physical
by
executing
optimization
problem.
During
optimization,
also
takes
account
non-linear
energy
requirements
based
on
utilization
levels
find
with
lowest
cost,
thus,
providing
solution
associated
formally
defined
formulated
integer
programming.
Since
intractable,
four
heuristic
methods
are
proposed.
To
evaluate
performance
methods,
several
experiments
conducted
comprehensive
test
suite.
results
demonstrate
benefits
model,
high-quality
solutions
methods.
IEEE Access,
Год журнала:
2022,
Номер
10, С. 81787 - 81804
Опубликована: Янв. 1, 2022
Increasing
demand
for
computational
resource
as
services
over
the
internet
has
led
to
expansion
of
datacenter
infrastructures.
Thus,
authorities
are
striving
adopt
optimal
power
usage
schemes
minimize
costs,
emissions
and
Service
Level
Agreement
(SLA)
violations
in
their
task
scheduling
heterogeneous
computation
centers.
One
most
effective
strategies
reduce
energy
consumption
is
maximize
utilization
physical
machines
shut
down
idle
ones.
This
can
be
realized
through
two
main
algorithms,
namely
virtual
machine
placement
consolidation.
The
VM
method
a
dynamic
process
put
these
devices
on
machines.
consolidation
technique,
however,
tries
improve
efficiency
grouping
live
migration
dispersed
lower
number
active
machine.
In
this
paper,
novel
approach
proposed
improving
efficiency.
employs
heuristics
meta-heuristic
algorithms
with
eight
performance
criteria
implemented
small
medium
scale
data
centers
using
simulated
cloud
module.
results
indicates
that
showed
up
10.3%,
5.3%,
12.5%
more
significant
rather
best
previous
respectively,
terms
consumption,
SLA
violation
VMs
migration.
Abstract
Cloud
computing
faces
challenges
in
energy
consumption
and
quality
of
service
(QoS).
Virtual
machine
(VM)
consolidation,
involving
relocation
between
hosts,
helps
reduce
power
usage
enhance
QoS.
OpenStack
Neat,
a
leading
VM
consolidation
framework,
uses
the
modified
best‐fit
decreasing
(MBFD)
strategy
but
QoS
issues.
To
address
these,
we
present
secure
efficient
(SEEVMC)
method,
introducing
unique
host
selection
criterion
based
on
incurred
loss
during
placement.
We
evaluated
SEEVMC
with
real‐time
workload
data
from
PlanetLab
Materna
over
ten
days
using
CloudSim.
For
PlanetLab,
reduced
by
78.33%,
57.74%,
19.57%,
6.30%
system‐level
agreement
(SLA)
violations
92.49%,
92.78%,
45.16%,
15.67%,
compared
MBFD,
power‐aware
best
fit
decreasing,
medium
power‐efficient
bit
decreasing.
Materna,
14.12%,
59.5%,
3.92%,
3.80%
fewer
SLA
74.85%,
86.95%,
11.40%,
46.60%.
also
migrations
time
per
active
host,
improving
cloud
efficiency.
