The
integration
of
Edge
AI
within
hybrid
IT
systems
presents
significant
challenges,
particularly
in
terms
scalability,
security,
and
data
integrity.
This
review
explores
the
complexities
deploying
environments,
emphasizing
role
containerization
blockchain-based
provenance
solutions
mitigating
these
challenges.
Containerization
enhances
portability
scalability
models
across
diverse
edge
devices
cloud
infrastructures,
while
blockchain
ensures
secure
verifiable
lineage,
addressing
concerns
related
to
authenticity
regulatory
compliance.
paper
examines
key
deployment
barriers,
including
resource
constraints,
interoperability
issues,
latency
considerations,
alongside
strategies
for
optimizing
model
efficiency
distributed
computing
environments.
Additionally,
it
evaluates
real
world
use
cases,
technological
frameworks,
best
practices
integrating
containerized
with
blockchain-driven
mechanisms.
By
bridging
gaps
operational
efficiency,
trust,
this
highlights
a
pathway
toward
resilient
transparent
deployments
ecosystems.
Advances in computational intelligence and robotics book series,
Journal Year:
2025,
Volume and Issue:
unknown, P. 271 - 304
Published: March 28, 2025
Swarm
intelligence
is
transforming
drone
tech
to
enable
autonomous
air
systems
collaborate
and
adapt
readily
real-world
conditions.
By
flying
in
coordination,
drones
can
perform
sophisticated
tasks
that
would
be
difficult
or
impossible
for
a
single
unit
accomplish.
Whether
search
rescue
large-scale
agricultural
surveillance,
coordinated
improve
speed,
coverage,
decision-making.
Edge
AI
significant
it
allows
process
information
real
time,
cutting
down
on
reliance
remote
cloud
servers.
This
enables
swarms
react
quickly
changing
environments,
such
as
navigating
through
disaster
scenes
tracking
moving
objects.
Unlike
using
central
controller
all
commands,
communicate
with
each
other
collectively
decide,
birds
flock
ants
colony.
coordination
supplemented
advanced
technologies
like
5G
connectivity
sensor
fusion
the
smooth
sharing
of
data
among
drones.
International Journal of Network Management,
Journal Year:
2024,
Volume and Issue:
35(1)
Published: Aug. 18, 2024
ABSTRACT
Deep
neural
network
(DNN)
and
machine
learning
(ML)
models/
inferences
produce
highly
accurate
results
demanding
enormous
computational
resources.
The
limited
capacity
of
end‐user
smart
gadgets
drives
companies
to
exploit
resources
in
an
edge‐to‐cloud
continuum
host
applications
at
user‐facing
locations
with
users
requiring
fast
responses.
Kubernetes
hosted
poor
resource
request
estimation
service
level
agreement
(SLA)
violation
terms
latency
below
par
performance
higher
end‐to‐end
(E2E)
delays.
Lifetime
static
provisioning
either
hurts
user
experience
for
under‐resource
or
incurs
cost
over‐provisioning.
Dynamic
scaling
offers
remedy
delay
by
upscaling
leading
additional
whereas
a
simple
migration
another
location
offering
SLA
bounds
can
reduce
minimize
cost.
To
address
this
challenges
ML
the
inherent
heterogeneous,
resource‐constrained,
distributed
edge
environment,
we
propose
ProKube,
which
is
proactive
container
orchestrator
dynamically
adjust
fair
balance
between
delay.
ProKube
developed
conjunction
Google
Engine
(GKE)
enabling
cross‐cluster
and/
dynamic
scaling.
It
further
supports
regular
addition
freshly
collected
logs
into
scheduling
decisions
handle
unpredictable
behavior.
Experiments
conducted
heterogeneous
settings
show
efficacy
its
counterparts
greedy
(CG),
(LG),
GeKube
(GK).
68%,
7%,
64%
reduction
CG,
LG,
GK,
respectively,
it
improves
4.77
cores
LG
more
3.94
CG
GK.
The
integration
of
Edge
AI
within
hybrid
IT
systems
presents
significant
challenges,
particularly
in
terms
scalability,
security,
and
data
integrity.
This
review
explores
the
complexities
deploying
environments,
emphasizing
role
containerization
blockchain-based
provenance
solutions
mitigating
these
challenges.
Containerization
enhances
portability
scalability
models
across
diverse
edge
devices
cloud
infrastructures,
while
blockchain
ensures
secure
verifiable
lineage,
addressing
concerns
related
to
authenticity
regulatory
compliance.
paper
examines
key
deployment
barriers,
including
resource
constraints,
interoperability
issues,
latency
considerations,
alongside
strategies
for
optimizing
model
efficiency
distributed
computing
environments.
Additionally,
it
evaluates
real
world
use
cases,
technological
frameworks,
best
practices
integrating
containerized
with
blockchain-driven
mechanisms.
By
bridging
gaps
operational
efficiency,
trust,
this
highlights
a
pathway
toward
resilient
transparent
deployments
ecosystems.
