►
Description
Get your espresso ready for another EMEA OpenShift Coffee Break! This week our special guest is Kevin Dubois, Solution Architect at Red Hat, and we’re going to talk about how Serverless , Camel Quarkus, and Camel K are perfectly complementary to build simple, standardized and highly scalable integrations between different applications.
Twitch: https://red.ht/twitch
A
A
Good
morning,
good
morning,
welcome
back
to
the
openshift
coffee
break.
I
hope
you
got
your
espresso
shot,
because
today
we
have
a
very
very
cool
topic.
We
have
a
super
special
guest
which
is
kevin
before
introducing
kevin.
I
would
like
to
say
hello
to
my
friends
and
co-host,
andrea
and
fabio.
How
are
you.
B
A
C
A
D
D
Kevin
dubois,
I'm
I'm
from
belgium,
and
and
that's
where
I
am
right
now
I
have-
I
may
have
a
slight
american
accent,
so
I
did
spend
a
good
chunk
of
my
career
in
the
in
the
states
as
a
software
engineer,
but
yeah
I'm
happy
to
be
back
in
belgium
at
least
for
now.
It's
sunny.
So
that's
that's
a
rare
day
in
belgium
yeah.
So
thank
you
for
inviting
me.
D
So
yes,
I
I
live
in
brussels,
I'm
originally
from
from
antwerp.
So
there's
a
saying
in
in
belgium,
if
you're,
if
you're
from
antwerp
you
have
like
antwerp,
is
basically
the
center
of
the
earth
kind
of
like
romans,
they
also
think
this,
and
but
so
they
called
the
rest
of
the
rest
of
belgium.
The
parking
to
answer.
A
D
All
right
yeah,
so
that's
yeah.
I
I
figured
this
this.
This
is
an
interesting
topic,
especially
you
know
a
few
weeks
ago.
I
guess
it
was
in
december.
Alex
was
on
your
show
and
I
was
talking
about
serverless
and
he
showed
a
cool
demo
about
websites
where
you
know,
based
on
the
demand
they
would
scale
up
and
and
then
scale
down
to
zero.
If
no
demand
was
coming
and
actually
that's
a
really
good
use
case
for
for
serverless
integration
as
well.
D
You
know
so
integration
where
you're,
integrating
with
external
or
internal
other
components-
and
you
you
basically
don't
have
control
over
what's
going
on
in
terms
of
demand
right.
So
if
you
can
scale
based
on
that
demand,
that's
that's
that's
better!
So
what
is
integration?
Let's
start
there
right.
So
what
is
the
problem
that
you
need
to
solve
with
with
integration,
so
there's
kind
of
two
main
use
cases?
D
Sometimes
it's
kind
of
a
micro
problem
where
you
get
a
request,
and
maybe
you
get
a
an
xml
request
from
you
know
an
old
batch
process
or
you
know
from
an
from
a
soap
service,
and
you
need
to
translate
that
to
to
a
json
request.
You
know
that's
it's
a
really
kind
of
basic
but
very
typical
example.
So
that's
kind
of
one
of
the
use
cases
of
integration
right,
so
you
transform
from
one
thing
to
another
and
then
the
other
use
case
is
more
of
a
let's
say,
macro
problem.
D
So
people
are,
you
know
that
are
familiar
with
the
old
school
esbs
are
are
probably
familiar
with
this
as
well
right.
You
have
to
integrate
between
different
components,
and
you
know
also
maybe
a
couple-
a
twitter
feed
to
a
sales
force
right
to
to
to
ingest
all
the
messages
that
you're
getting
for
your
marketing
team
or
your
your
you're,
connecting
with
with
an
external
party
or
you
know,
there's
many
many
use
cases.
That's
also
integration.
D
So
it's
both
about
connecting
different
applications
and
also
about
the
transformation
of
the
information
that
that
you're
getting.
And
then
you
know
when
it
comes
to
integration,
you
have
kind
of
good
integration
and,
and
maybe
not
so
good
integration.
So
typically
what
what
a
lot
of
organizations
start
with-
and
you
know
in
my
career
as
well-
we
would
start
with
you
know
just
kind
of
doing
some
custom
things
and
cobble
them
together,
and
that
would
work.
D
But,
as
you
can
kind
of
see
from
from
that
picture
on
the
left,
it
doesn't
really
scale
very
well
and
then
afterwards,
if
you
need
to,
you,
know,
replace
something
you
need
to
find.
You
know
that
tiny
little
custom
thing
that
that
that's
not
standard.
So
if
you
can
use
reusable
blocks
first
of
all,
you
can
go
faster
right
to
build
with
lego,
it's
pretty
easy
and
fast,
and
it
also
means
that
you
can
swap
out
components
much
more
easily.
D
So
obviously
we
prefer
the
the
standard
integration
and
that's
where
apache
camel
the
apache
camel
process
comes
in.
So
it's
a
project
that
was
kind
of
started
to
standardize.
The
integration
with
you
know
with
different
some
capabilities,
so
they
call
it
kind
of
the
swiss
knife
of
integration.
So,
on
the
one
hand,
it's
based
on
you
know
well
established
patterns.
So
if
you're
familiar
with
the
the
enterprise
integration
patterns
book,
it's
it's
pretty
much
focused
on
that
right.
D
D
There's
lightweight
run
times
that
you
can
use
with
it,
so
you
can
really
do
integration
kind
of
where
you
want
it
on
the
spot.
You
know,
even
even
in
in
terms
of
microservices,
so
you're
not
really
tied
to
the
classic
esb
approach.
Where
you
have
this
big
kind
of
monolithic
integration
component,
where
you
know
it's
kind
of
a
bottleneck
for
everything
else
and
in
terms
of
data
formats.
It
also
knows
all
the
different
data
formats
or
a.
D
So
you
don't
have
to
reinvent
the
wheel
there
either
apache
camel
is
based
on
a
kind
of
a
domain
specific
language.
It's
very
easy
to
learn.
So
here
you
can
see,
for
example,
in
that
little
example
from
a
kafka
topic
right
topic,
name
to
a
grpc
endpoint.
I
mean
it's
really
can
be
that
simple
and
in
fact,
I'll
show
you
in
the
demo
that
it
really
can
be
that
simple.
So
a
couple
more
things
about
apache
camel.
So
it's
a
reactive,
it
has
a
reactive
routing
engine.
D
A
D
It's
kind
of
interesting
how
it's
you
know
the
project
has
been
around
for
a
long
time,
but,
as
you
can
see
kind
of
more
recently,
there's
a
spike
in
in
in
the
amount
of
contributions.
So
this
project
is
actually
not
dying
out.
It's
it's
it's
picking
up
even
more,
and
that
has
something
to
do
with
some.
D
Of
apache
camel
and
we'll
we'll
jump
into
that
a
little
bit
more
so
in
terms
of
integration,
there's
also
been
kind
of
this
evolution
of
you
know.
Initially,
you
had
point-to-point
integration
right,
you
had
like
classic
it
systems
and
you
would
build.
You
know
an
integration
from
this
system
to
that
system,
and
you
know
a
lot
of
it
would
be
basically
hard
coded
to
communicate
with
another
service.
C
D
D
D
You
know
very
small
where,
where
it's
needed,
we
can
also
scale
very
fast
based
on
on
the
demand
and
scale
down
to
zero,
which
you
know
is
is
interesting.
When
you
know,
if
you're
not
getting
requests
or
are
you
getting
very
uneven
requests,
maybe
once
a
day
or
something
you
don't
need
to
be
running
a
service
the
whole
time
so
apache
camel
3.
C
D
D
Of
of
camel,
camel,
corkus
and
then
a
couple
more
camel
crafts,
spring
boots
and
and
kafka
connectors
and
and
we'll
not
talk
about
that
very
much
today.
My
main
focus
is
apache,
camel
extensions
for
quarks,
that's
the
the
official
name.
I
just
call
it
camel
cork.
Is
that
just
like
flows
a
little
better?
So
let's
go
with
that,
and
so
you
know
it's
based
on
quarkus.
So
natalia,
I'm
not
sure
if,
if
you've
had
any
guests
that
that
have
talked
about
quarkus
yet
well,.
C
A
I
catch
this
opportunity
to
say
hello
to
zineb
in
the
chat.
Zinab
is
a
pmc
at
apache
camel,
so
you
know
we
have
also
the
community
that
joined
the
discussion
and
please,
if
you
have
any
question
the
people
attending,
send
the
question
in
the
chat.
So
we
we
will
help
and
and
send
the
question
to
kevin
thanks
excellent.
D
D
A
A
D
D
So
this
is
where
you
know
the
quarkus
developers
came
well
came
up
with
with
with
quarkus,
I
guess,
as
a
java
runtime
that
uses
a
lot
less
resources
starts
up
a
lot
faster
and
you
know
takes
advantage
of
those
smaller
footprints
but
still
having
you
know
the
the
same
knowledge
of
the
java
framework
for
the
developers.
So,
as
you
can
see
here,
you
can
fit
a
lot
more
applications
onto
a
platform.
You
can
increase
that
density
to
be
able
to
use
the
full.
D
You
know
the
full
size
of
of
your
platform,
and
so
you
know,
kind
of
circling
back
to
camel
camel
together
with
quarkus,
makes
a
lot
of
sense
right.
So
you
can
have
your
integration,
but
do
it
in
a
scalable
way
in
a
quick
way,
and
you
know
have
all
those
advantages
camel
extensions
for
quarkus
camel
corkus
you
can
deploy
it.
D
D
D
Those
are
all
events
and
they
trigger
an
action
right
in
serverless
that
your
application
sees.
It
gets
a
request
and
then
it
scales
up
or
or
maybe
it's
already
running-
handles
the
request
and
then
produces
a
result
and
that
result
can
be
also
an
event
it
can
be.
You
know
it
can
also
be
you
know,
making
a
file
and
and
saving
it
somewhere.
It
can
be,
you
know
whatever
you
want,
but
that's.
D
Pattern
of
serverless,
so
in
a
nutshell
again,
you
know,
I
know
that
alex
talked
about
this
in
december
and
explained
it
much
better
than
I
did,
but
you
have
you
know
over
to
serverless
based
on
containers
and
making
containers
easy
right.
So
it
has
some
capabilities
to
deploy
and
use
containers
more
easily.
It's
ready
for
the
hybrid
cloud
because
you
can
use
it.
D
You
know
on-prem
in
in
the
public
cloud
private
cloud
wherever
you
want,
it
has
its
concept
of
immutable
revisions,
and
so
basically,
every
time
you
change
the
configuration
of
your
of
your
serverless
service.
It
creates
a
new
revision
and
that's
kind
of
easy,
because
then,
if
you
need
to
roll
back
or
you
need
to
you
know,
you
need
to
do
any
kind
of
canary
deployments
or
a
b
or
whatever
it
makes
it.
D
To
to
work
with
that,
openshift
serverless,
because
it's
container
based
you
can
use
any
programming
language
right,
so
java
python
go
whatever,
but
what's
really
interesting
for
for
integration!
Is
that
those
last
two
parts
to
automatic
scaling
right?
So
we
can
scale
down
to
zero,
wait
for
requests
to
come
in
and
then
scale
up
to
whatever
we
need
to
do
right
if
it's
a
thousand
pods
or
one
pod
or
anything
in
between
you
know
whatever
is
needed,
it
can
scale
really
fast
and
then
you
know,
of
course,
event
driven
architectures.
D
We're
basically
talking
a
lot
about
events
in
in
all
those
examples
that
we've
given
right.
So
somebody
logs
on
to
a
website,
that's
an
event,
somebody
posts,
something
on
a
kafka
stream,
that's
an
event
right.
It's
all
events
that
allow
us
to
build
more
loosely
coupled
architectures,
distributed,
applications
and
stuff,
like
that,
so
just
to
kind
of
show
what
the
advantage
is
here
in
terms
of
integration
as
well
again,
if
you
have
a
request
coming
in
and
you
need
to
be
able
to
handle
them
without
serverless.
B
Yes,
it
looks
like
there
is
more
than
a
good
question,
a
feedback
from
andre
again
andrew
good
morning.
Feedback
around
quarkos
is
already
trying
parkour's
apps
in
production
and
looking
at
his
experience
seems
that
everything
is
great
and
obviously,
as
already
someone
suggested
in
in
the
chat
tina
suggested.
B
B
D
A
D
B
D
See
with
the
with
the
slide
you
know
with
serverless
on
on
the
right,
you
can
really
scale
based
on
the
actual
need,
but
you
need
a
framework
that
that
can
support
that
right.
So
traditional
java
application
is
not
going
to
be
able
to
do
that
because
it
needs
time
to
start
up
and
by
then
maybe
the
demand
is
already
down,
and
you
know
the
quarkus.
A
C
D
D
D
D
Again,
your
application
starts
very
quickly.
You
make
some
changes,
the
the
you
know.
Hopefully
your
ide
saves
automatically
right
most
ids
these
days.
Do
you
don't
have
to
like
control
s
or
whatever
anymore
and
immediately
the
you
know
the
changes
are
taken
are
are
redeployed
by
by
quarkx,
so
you
don't
need
to
recompile
or
anything.
So
that
makes
it
a
really
nice
developer
experience
just
just.
B
D
All
right
there's
a
reason
why
it's
called
dev
mode
right,
I
mean
that's,
that's
kind
of
the
the
the
trade-off
with
with
quarkus.
Is
that
because
it's
because
it's
so
fast
it
needs
to,
you
know,
build
everything
ahead
of
time.
It
needs
to
compile
everything
ahead
of
time.
So,
if
you're
building
a
corcus
application-
and
I
guess.
D
Mention
the
the
native
build
of
of
quarkx.
So
basically
you
can
build
quarkus
application,
as
you
know,
as
a
native
application,
so
that
you
don't
have
to
have
a
jvm
running
in
your
container.
That's
one
of
the
ways
that
they
can
make
it
a
lot
faster,
but
it
does
take
a
while
to
to
build
a
native
application.
D
Time,
you're
you're
wanting
to
test
a
little
change.
That's
that's
not
ideal,
so
this
dev
mode
is
kind
of
a
great
compromise
where,
as
you're
developing,
you
can
make
all
those
changes
and
reload
really
fast
and
then
once
you're
ready
to
to
deploy
it,
especially
to
production
yeah.
Please
do
go
through
a
cic
pipeline
that
you
know,
does
all
your
tests
and
all
your
verifications
and
runs
the
integrations
and
all
that
stuff.
A
A
D
Yes,
so
actually
maybe
I'll
I'll
show
the
slide
first,
so
this
demo
is
really
simple.
Basically,
we're
gonna
be
looking
at
posts
on
twitter
and
the
two
tags
that
it's
looking
for
is
camel
corkus.
So
if
you're,
if
you're
on
twitter-
and
you
want
to
see
your
your
your
actual
tweet
coming
up
in
in
this
demo,
you
know
now's
the
time
to
post
something
on
twitter.
Just
do
oh
tag,
camel
hashtag,
quercus
and
then
hopefully
we'll
we'll
see
it
come
up
in
the
demo.
D
Next
to
each
other,
all
right,
so
I'm
just
using
the
developer
sandbox,
because
it's
super
easy
to
use
and
it's
and
it's
free
to
to
do
little
demos
and
little
proof
of
concepts
and
so
on.
So
you
know
you
can
actually
just
do
this
the
same
way
that
I'm
doing
it
and
I'll
you
know,
maybe
in
the
comments
or
something
I
can
share
the
the
github
and
then,
if
you
want
to
do
it,
please
do
so.
D
Logged
in
to
into
openshift,
and
if
my
user
interface
wants
to
load
there,
we
go
so
I
don't
have
anything
running
right.
So
if
I
want
to
add
my
application,
so
let's
first
import
it
from
git-
and
this
is
my
repository
and
so
open
shift
just
automatically
validates
it
and
sees
okay.
This
is
a
java
application
and
that'll
be
good
enough
for
for
for
this.
For
this
example,
what.
D
D
D
A
A
D
D
D
Okay,
this
is
the
job
application
and
it's
going
to
start
a
build
of
of
the
application
and,
as
you
can
see
here,
the
build
is
running
and
it's
going
to
you
know,
run
a
maven
package
and
it's
going
to
download
half
the
internet
and
stuff
like
that.
So
that's
going
to
take
a
while
and
we're
not
going
to
wait
for
that.
But
what's
cool
here
as
well
is
that
now
I
can
go
look
at
the
source
code
directly
in
open
shift
in
in
the
code,
ready
workspaces,
and
then
you
know
if
that
loads.
D
Maybe
I
can
also
just
deploy
this
application
from
a
container
image
that
I've
already
built,
and
this
one
is
one
that
I
built
as
a
as
a
native
build.
So
we're
gonna
see
that
this
one's
gonna
start
a
lot
faster.
I
mean,
obviously
because
it's
already
built,
but
also
because,
because
of
the
native
part,
it's
going
to
be
going
a
lot
faster.
D
D
B
A
Please
send
your
tweets
in
live.
We
will
see
in
the
live
demo
your
tweets
coming
up.
I
I
sent
one
tweet,
but
please
send
the
other
tweet
with
a
quarkus
and
camel
hashtag
they're
gonna.
The
this
demo
will
show
them
so
talking
about
in
the
while.
Having
talking
about
the
architecture
of
this
now
how
the
data
is
retrieved
retrieved,
it's
the
application
using
the
the
library,
the
camera
library,
what
what
is
the
where?
How
is
the
flow.
D
C
D
D
D
D
D
D
A
D
Completely
automatically
created
so
I
didn't,
I
didn't,
create
any
dev
files
or
anything.
So
this
is
basically
just
you
know
what
I
showed
you
here.
I
just
clicked
on
the
little
icon
here
to
edit
the
source
code,
and
that
opens
a
code
ready
workspace
for
me,
where
I
can
immediately
start
changing
start
changing
the
code.
D
D
D
D
A
D
Configured
a
little
rest
endpoint,
but
what
we
can
do
with
serverless2
is.
We
can
couple
our
applications
with
event
sources
right,
so
it
can
be
a
kafka
stream.
It
can
be
a
cloud
event,
it
can
be
other.
You
know
camel
k,
components
or
whatever
in
this
case,
I'm
going
to
keep
it
really
simple,
because.
D
A
D
I'm
just
going
to
do
ping
source,
so
you
know,
and
if
you,
if
you
noticed,
I
set
it
to
ping
every
minute,
so
every
minute
is
going
to
wake
up
the
application.
The
application
is
going
to
look
for
twitter
messages
or
yes,
twitter
messages
and
then
going
to
send
them
on
to
our
telegram.
So
we'll
see
you
know
in
less
than
a
minute
now,
if
that's
actually
going
to
happen
or
not,
the
topology
is
small.
Can
we
increase
it?
Well,
let's
see,
of
course
we
can.
D
And
I
don't
think
you
heard
this,
but
I
did.
I
got
twitter
notifications,
so
here
we
can
see
that
you
know
zenith,
for
example,
posted
a
nice
talk
by
kevin
dubois.
Thank
you,
and
so
it
works
right.
I
mean
it's
fairly
simple.
This
is
really
a
basic
example
of
what
it
could
be
and,
as
you
can
see
now
it
did
its
job.
D
It
searched
for
the
twitter
messages,
sent
them
on
to
telegram
and
then
terminating,
and
it's
just
going
to
lay
quiet
until
the
next
time
it's
summoned
upon
which
in
this
case
is
every
minute
because
of
the
pig
source,
but
you
know
a
picture,
maybe
where
you
have
another
event,
source
that
comes
in
sporadically
through
kafka
stream
or
whatever,
and
then
you
know
the
application
scales
up.
Does
its
job
runs
and
then
scales
back
down
to
to
zero.
D
A
D
D
A
D
Yeah
exactly
yeah
I
mean
you
know
so
it's
part
of
the
live
coding
capability
of
quarkus.
But
yes,
indeed,
you
can
also
do
this
for
for
your
actual
testing.
So
you
can
do
continuous
testing
continuous,
making
sure
that
you
know
as
you're
writing
code
that
your
tests
don't
fail
and
that
you
don't
have
to
wait
until
you're
going
through
your
cic
pipeline
to
realize,
like.
A
D
B
B
It
probably
yeah
it's
an
italian
mistake,
so
please
apologize
again
so
sebi
pointed
out
to
the
as
he
called
it.
The
perfect
combos
manager
cafe
native
cafe
it's
something
that
you
ever
probably
know
not
highlighted
service
binding
a
feature
that
makes
probably
the
integration
effort,
far
easier
than
just
doing
everything
by
hands.
C
D
D
Yeah,
I
mean
that's,
that's
what
we're
saying
right.
I
mean
that
you
know,
for
example,
kafka
is
a
great
use
case
for
event,
driven
architectures,
where
you're
decoupling,
different
applications
and
with
with
the
offerings
you
know
like
natalia,
was
saying
too.
We
have
a
managed
kafka
offering
where
you
don't
have
to
deal
with,
setting
up
a
whole
kafka
cluster,
and
you
know
configuring
zookeepers
and
all
that
stuff.
You
can
just
kind
of
get
going
with
it
and
it
integrates
really
well.
B
D
D
B
And
in
the
meantime,
sebi
also
pointed
out
to
one
more
great
feature.
Probably
we
will
not
have
time
to
today
to
speak
about
it
and
probably
during
more
during
the
next
session,
but
kamikaze
was
camelets
camelets,
obviously
that
you
could
do
almost
the
same
demo
as
sebi
is
telling
in
his
comment
just
applying
some
yaml
file
and
accommodate
the
resource.
That's
right.
D
Yeah
absolutely
so,
I
have
a
few
more
slides
if,
if
you
want
to
show
them
real,
quick,
I'm
real
real
high
level.
So
there's
this
you
know,
so
we
were
talking
about
camel
quark.
Is
this
whole
time
right
where
we're
actually
in
the
code
integrating
with
camel
there's,
also
camel
k,
and
then
that
uses
camel
camelets
as
well,
which
which
I'll
talk
about
in
just
a
second,
so
camouflage.
D
D
Manifest
that's
that
you'll
deploy
on
kubernetes
this
camel
k
operator.
That's
gonna!
Take
you
know
your
basically,
your
your
little
integration
file
and
it's
going
to
deploy
it
in
a
serverless
way
and
and
take
all
that
kind
of
stuff
that
I
was
doing
a
little
more
manually
is
gonna
automate
that
so
that
makes
it
makes
it
pretty
cool
too
bruno.
C
C
D
D
D
You
know:
do
you
need
to
do
a
little
bit
more
complex
integration,
then
you
know
camel.
Corkus
is
probably
the
way
to
go.
If
you're
like
a
developer,
who
likes
to
be
in
his
ide,
you
can
use
camel
corkus,
but
if
you
want
to
do
just
like
basic
integrations
and
and
not
have
to
deal
with
with
the
whole
code
part,
then
then
this
is
a
pretty
good
use
case
to
use
to
do
integration
without
even
really
knowing
any
java
or
anything.
D
B
Yeah
thanks
thanks
for
introduce
the
next
session
kevin
and
one
thing
just
a
very
small
question.
Looking
at
your
presentation,
someone
can
think
that
camo
quarkus
should
be
used
only
or
camel,
with
quarkx
accession
could
be
used
or
should
be
used
only
when
you
have
to
deal
with
serverless
services
or
serverless
implementation,
but
actually
is
not.
D
Absolutely
yeah,
so
serverless
is
a
use
good
use
case
in
terms
of
you
know,
you
can
scale
down
to
zero
and
you
can
scale
really
quickly
because
it
it
acts
on
based
on
requests
right.
So
as
soon
as
a
request
comes
in
it's
going
to
scale
up
or
down,
whereas
you
know
your
typical
kubernetes,
auto
scaling
is
based
on
cpu
and
memory,
so
it
takes
a
little
longer
to
you
know
realize
that
that
there's
a
spike
in
demand-
and
you
can
also
not
necessarily
scale
down.
D
D
A
really
easy
and
straightforward
path
to
you
know
kind
of
modernizing
approaches
that
that
we've
been
using
for
for
many
for
many
years
right.
Our
our
java
experience
our
camel
experience
and
just
kind
of
use
that
in
in
this
modern
world
of
you,
know
cloud
native
and
serverless,
and
you
know
responding
to
demand,
really
quick
and
then
you
know,
and
if
we
don't
need
it,
let's
not
use
it
and
not
let
not
waste
electricity
and
stuff
like
that
or
you
know
cloud
resources,
because
they're
they're
also
expensive.
D
A
First,
thanks
kevin
really
great
live
demo,
we've
seen
sending
it
to
it.
I
think
people
are
still
sending
the
tweets
so
that
thank
you.
We
have.
We
have
a
question
from
the
chat
from
actually
it's
a
comment
from
apple
camel
k
is
a
swiss
knife
that
you
can
use
for
all
integration
use
cases.
Well.
This
is
a
great
statement.
Thanks
abel.
If
you
see
the
from
the
live
demo,
there
are
many
tweets
so
that
worked.
A
No
live,
serverless
super
cool
one
question
kevin:
can
the
people
do
your
demo?
Is
there
any
source
code
that
you
can
share
because,
as
you
mentioned
you,
you
did
the
demo
on
developer
sandbox
for
openshift,
which
is
a
free,
openshift,
multi-tenant
environment.
So
you
can
register
for
free.
We
we
put
the
link
in
the
chat
and
and
get
your
sandbox
account
in
this
sandbox.
You
can
run
the
same
demo
that
kevin
did
so
we're
gonna,
also
kevin.
I
think
we're
gonna
share.
Oh
this
is
the
github
yeah
yeah.
D
B
C
C
D
D
I
did
see
it.
It
said
something
about
startup
in
three
seconds
or
something.
So
that's
in
jvm
mode.
Now,
if
we
go
to
to
the
one
that
the
the
native
runtime,
let's
see,
if
we
can
wake
it
up
ping,
it
real
quick
and
that
should
start
it
up
there
we
go,
you
can
see.
It
goes
pretty
quick
here
started
in
a
second
and
I
think
that's
that's
fairly
slow.
I
think.
D
Usually
it
starts
even
faster,
so
I
mean,
but
even
then
right,
that's
the
the
jvm
one
started
in
in
three
times
longer
than
than
the
quarkus
and
the
quark
is
native.
Well
they're,
both
quarkus
the
native
mode
started
in
three
times
faster
than
the
in
the
jvm
mode.
So
you
can
even
see
it
like
that
right
and
you
can
see
camel
the
camel
components
are
in
13
milliseconds.
That's
not
bad.
D
Get
metrics
on
that
as
well:
yeah
yeah
yeah,
so
I
mean
in
I'm
not
sure
if
this
is
as
easy
to
show,
but
it's
also
cool
right
in
in
openshift.
You,
you
get
your
observation
out
of
the
box.
You
can
see
immediately
all
your
memory
memory
utilization
stuff
like
that,
but
yeah
you,
you
would
be
able
to
see
it
pretty
easily.
Here
I
would
imagine.
C
A
Yeah,
we
will
have
bruno
talk
about
the
camel
k
and
in
the
shade,
if
we
look
at
the
shed,
we
will
have
the
reminder
of
the
shell
schedule
in
the
in
the
while.
I
think
thanks
kevin
for
this
great
live
demo
and
kevin.
Looking
at
your
demo,
I
was
thinking
what
about
if
any
new
app
is
a
serverless
app?
What
do
you
think
about.
D
So
every
app
is
a
serverless
app.
It's
a
that's
a
good
question.
So,
basically,
are
there
downsides
to
deploying
it
as
a
serverless,
but
I
think
taro
had
a
had
a
good
point
in
in
december
when
we
were
talking
about
serval.
Is
that
you
know,
even
if
it
scales
down
to
zero,
there's
still
resources
in
the
in
the
background
right,
there's
still
an
operator
that
needs
to
run
and
and
and
it.
D
But
yeah
I'm
trying
to
think
of
like
a
real
reason
why
you
wouldn't
run
as
a
serverless
component,
I'm
not
entirely
sure,
because
you
can
configure
k
native
right,
that's
behind
openshift
serverless
to
keep
you
know,
let's
say
two
three
pods
running
at
all
times
and
then
scale
up
and
down
based
on
based
on
that.
So
you
know
the
it's
not
that
you
have
to
scale
to
zero.
So
you
know
I'd
be
interested
to
to
know
a
use
case.
A
Yeah
we
see
in
the
chat
a
great
shout
out
to
you
kevin
sebby
gina.
Thank
you
all
for
joining
their
super,
the
so
the
say
that
was
a
really
cool
demo.
Also
on
twitter.
I've
seen
good
feedback,
a
quick
reminder
for
everyone,
the
the
say
the
the
recording
of
this
show
it's
on
the
same
youtube
link
or,
if
you
are
attending
from
twitch,
is
gonna,
be
on
also
available
on
twitch
and
fabio
andrea.
We
go
into
our
reminders
for
today
we
have
our.
You
know
usual
schedule
here
at
openshift
tv.
A
C
A
A
Everything
we
like
we
love
coffee
diversity
are
also,
you
know,
hot
beverage,
diversity.
You
know
30
coffee,
whatever
it's
fine
for
us,
so
thanks
kevin
kevin
for
joining
us
for
having
joining
us
today.
If
you
would
like
to
follow
kevin
on
twitter,
the
twitter
handle
is
in
its
picture.
You
can
see
kevin
dubois
if
I'm
not
wrong
right
kevin.
So
please
follow
very.