Multi‐Criteria Optimization of Scientific Workflow Schedules for Improved Energy Efficiency in Cloud Infrastructures
Concurrency and Computation Practice and Experience,
Год журнала:
2025,
Номер
37(9-11)
Опубликована: Апрель 9, 2025
ABSTRACT
Rising
global
dependence
on
cloud
services
has
become
crucial
for
enterprises,
aiming
to
guarantee
continuous
data
accessibility
while
pursuing
enhanced
energy
efficiency
and
minimized
carbon
emissions
from
centers.
However,
the
persistent
challenge
of
high‐energy
consumption
in
these
facilities
necessitates
a
concentrated
approach
toward
reduction.
This
paper
introduces
an
innovative
multi‐objective
scheduling
strategy
scientific
workflows,
tailored
heterogeneous
computing
environments.
Our
method
employs
hybrid
genetic
algorithm,
incorporating
Hill
Climbing
generate
initial
population
chromosomes.
Subsequently,
algorithm
optimizes
task
assignments
most
suitable
virtual
machines,
utilizing
meticulously
designed
fitness
function
evaluate
each
chromosome's
suitability
solving
problem.
Through
extensive
experimentation,
we
demonstrate
that
our
proposed
outperforms
other
techniques
terms
solution
quality,
contributing
reduced
consumption,
processing
duration,
cost.
We
contend
this
holds
substantial
potential
mitigating
footprint
associated
with
centers,
offering
sustainable
environmentally
conscious
workflow
scheduling.
Язык: Английский
The mapping trick: leveraging RoboSoccer obstacle avoidance and navigation for advanced task scheduling solutions in foggy IoE ecosystems
The Journal of Supercomputing,
Год журнала:
2025,
Номер
81(5)
Опубликована: Апрель 15, 2025
Язык: Английский
An Efficient Load Distribution Approach for Optimizing Resources in SDN‐Based Edge Computing Environment
Concurrency and Computation Practice and Experience,
Год журнала:
2025,
Номер
37(12-14)
Опубликована: Май 14, 2025
ABSTRACT
In
the
rapidly
evolving
networking
and
communication
technology
era,
emergence
of
novel
edge
computing
paradigms
helps
reduce
latency
improve
efficiency.
The
advancements
bring
data
processing
closer
to
its
source,
reducing
distance.
Moreover,
integrating
Software‐Defined
Networking
(SDN)
in
enhances
network
management
by
decoupling
control
plane
from
plane,
enabling
more
flexible
efficient
resource
allocation
distributed
environments.
However,
scheduling,
allocation,
load
balancing
are
significant
obstacles
enhancing
resources'
performance.
Besides,
help
use
all
resources
optimize
system's
performance
effectively.
To
address
these
issues,
this
paper
proposed
an
Average‐Based
Resource
Allocation
Load
Balancing
(ABRL)
algorithm
for
task
balancing,
which
aims
minimize
task's
completion
time
enhance
utilization.
A
three‐layer
SDN‐based
architecture
is
designed
implement
that
improves
simulation
studies
have
been
conducted
using
OpenDaylight
(ODL)
controller
implemented
Python.
Experimental
results
demonstrate
strategy
optimizes
makespan,
average
utilization,
level
under
consideration
exhibits
better
than
existing
state‐of‐the‐art
techniques.
Язык: Английский