►
Description
There’s 40x more bytes of data on planet Earth than there are stars in the observable universe… and this data is dispersed over public cloud, private cloud, SaaS, on-premise data centers, smart devices, and more! In this vast universe of tools, environments, and networks, our data is flying all over the place all of the time as these distributed services are connected together. With so much more data in motion than we’ve ever seen before, how can we make sure every single byte of it is visible, governed, and managed?
Put your spacesuit on and explore with us why traditional API management is not built for the new world and discover how Kong, a Service Connectivity Platform totally compatible with OpenShift, is built for these new challenges. Adventure through the worlds of API gateway, Kubernetes Ingress Controllers, Service Mesh, and APIOps to bring back home a new approach to connectivity.
A
I'm
going
to
one
two
three:
okay,
great!
Thank
you
chris
for
getting
us
rolling,
I'm
michael
waiting.
This
is
another
edition
of
the
openshift
commons
briefings
our
operator
hours.
Tv
show
which
is
streaming
live.
I'm
now
going
to
do
what
I
like
to
call
share
my
whole
screen
and
I'm
going
to
start
from
the
first
slide
here.
It's
not
going
to
be
slide
heavy,
but
I
did
want
to
put
up
the
the
intro
slide
here
today.
We
are
lucky
enough
to
have
our
good
friends
from
kong
with
us.
A
The
cloud
connectivity,
company
and
they're
going
to
be
talking
about
why
api
management
is
dead
somewhat,
like
some
people
might
think.
That's
somewhat
of
a
controversial
title,
which
I
think
is
a
is
a
perfect
segue
for
for
this
conversation
here
today
and
they're
going
to
talk
about
why
we
need
a
new
approach.
So
we
have
the
field
cto
melissa,
van
der
hek,
joining
us
from
overseas
in
the
uk,
and
we
have
claudio
aqua
viva
one
of
their
lead
solutions.
A
A
A
Snow,
I'm
tired
getting
a
little
tired
of
shoveling
snow,
so
kong,
king
kong.
What
what's
in
a
name
melissa!
You
know
when
I
when
I
was
looking
up
interesting
facts
about
kong.
The
other
day
I
you
know
I
typed
in
kong
and
you
can
imagine
there
were
all
kinds
of
references
to
things
that
were
you
know
something
other
than
your
company
tell
us
a
little
bit
about
the
company.
A
B
B
We
are
on
across
kong.
We
are
on
a
mission
to
our
connections
to
build
a
reliable
digital
world.
What
does
that
mean?
I'm
sure
you're
asking
mike
everybody
watching.
I
want
you
to
close
your
eyes
picture
the
night
sky
if
you've
ever
seen
the
line
of
the
milky
way.
Imagine
that
beautiful,
beautiful
image,
what
you're
seeing
in
front
of
you
is
one
billion
trillion.
B
A
B
So
we
have
more
40
times
more
bytes
of
data
on
planet
earth
than
there
are
stars
in
the
observable
universe.
It's
quite
a
challenge
that
we
have
within
it
at
the
moment,
and
it's
it's
not
just
that
we
have
more
data,
but
our
data
is
more
and
more
distributed.
There
are
an
estimated
50
billion
connected
smart
devices.
Today,
the
average
enterprise
uses
over
900
different
applications
from
on-premise
sas
different
cloud
environments.
B
B
That
means
we
and
cardio
took
a
second
about
story.
How
we
came
to
be
being
the
world's
most
widely
used
open
source
api
gateway,
but
we
also
power
microservices
service,
mesh
kubernetes
everything
coming
next
trends
that
we
might
not
even
have
dreamt
to
imagine
about.
Yet
we
are
powering
the
service
connectivity
platform
that
allows
you
to
have
this
reliable
connectivity
layer.
A
B
Absolutely
absolutely,
in
fact,
a
customer
that
I
work
very
closely
with
one
of
the
largest
retailers
in
the
world.
They
are
a
fashion
retailer
and
they
sell
to
202
markets
through
their
online
platforms
and
they're
running
kong
in
open
shift
to
power
all
of
this
e-commerce
and
power.
All
of
that
online
shopping
experience.
A
Okay,
great
and
do
you
guys
have
an
operator?
I
know
I
know
red
hat
spent
quite
a
bit
of
investment
in
working
with
our
partners
to
to
build
operators
on
openshift.
You
know
not
being
the
most
technically,
you
know
savvy
person
in
the
world,
I'm
pretty
sure
that
the
operators
are
pretty
important.
For
you
know
improving
day.
Two
supportability
of
you
know
open
shift
and
business
apps
are
you
you
guys
have
an
operator
in
our
in
our
registry
and
catalog
yeah.
C
Yeah,
I
can
take
this
one
melissa,
google
yeah.
Absolutely
we
provide
a
co-operator
as
a
matter
of
fact,
it's
part
of
the
cong
supporting
kubernetes,
not
just
having
kong
running
in
a
kubernetes
pod.
It
is
a
matter
of
fact
to
support
sporting
kong
is
a
much
broader
way.
You
know
supporting
operators
supporting
hemp,
shorts,
crts,
k,
native
certification
managers
and
so
forth.
C
C
Yeah
absolutely
yeah.
I've
been
I've,
been
working
at
kong
for
almost
two
years
and
a
half
now
as
a
solution
architect
and
as
such,
I'm
responsible
for
you
know,
helping
our
customers
and
partners
and
users
with
you
know,
designing
reference
architectures,
taking
care
of
these
modernization
process,
like
you
know,
workload
from
transfer
from,
let's
say
one
premise:
to
moody
cloud
deployment.
This
kind
of
thing:
it's
a
fun
job
as
a
matter
of
fact,.
A
D
A
A
B
B
I
think
everyone
who's
familiar
with
apis,
I'm
sure
we
can
all
go
back
a
few
years
and
remember
the
time
that
we
spent
educating
people
on
what
apis
were
and
then
kind
of.
Finally,
reaching
that
milestone
where
business
people
would
start
understanding
that
maybe
not
understanding
what
apis
are,
but
understanding
that
joining
the
api
economy
was
good
for
business,
and
there
are
very
many
reasons
why
joining
the
api
economy
is
great
for
business
all
types
of
enterprises.
B
They
are
joined
together,
everything
we
do
digitally
or
even
manually,
there's
some
process
somewhere
that
is
enabled
through
apis
and
through
microservices,
one
of
the
one
of
my
favorite
examples.
Actually,
I'm
sure
we
all
remember
kind
of
at
the
beginning
of
the
various
lockdowns
that
we
had
globally
last
year
in
the
uk.
I
know
various
parts
of
the
us
and
also,
I
wonder
which
other
countries
this
happened
in.
B
Do
you
remember
going
to
the
supermarket
and
seeing
all
these
empty
shelves
everywhere
and
then
just
people
with
their
phones
out
taking
pictures
of
essentially
the
apocalypse
in
the
supermarket
they're
being
the
epic
toilet
roll
shortages
in
the
uk.
What
we
saw
was
that
there
was
this
increased
demand
because
of
covid
to
deliver
ppe
and
medical
equipment
to
hospitals,
to
all
the
staff
who
needed
it,
and
there
was
the
need
to
continue
to
deliver
all
these
supplies
to
the
supermarkets.
A
We're
just
as
a
reminder
for
everyone:
we're
live
right
now
on
facebook
and
twitch
and
youtube,
and
if
anyone
has
questions
while
we're
here
with
our
with
our
friends
from
kong,
please
feel
free
to
put
them
in
chat
and
our
and
our
production
team
will
make
sure
that
they
get
them
in
here
and
they'll
be
addressed,
I
just
it.
It
did
strike
a
chord
when
you're
talking
about
what
it
was
like
at
the
beginning
of
the
pandemic.
A
D
C
A
D
C
That
shave
exactly
so
yeah
you
as
americans,
are
quite
an
interesting
story
like
you
know,
it
begins
when
that
these
two
italian
friends,
augusto
marietti,
our
ceo
and
marco
palladino,
our
cto.
He
came
to
united
states
and
decided
to
found
this
company
called
nation
that
was
around
10
10
years
ago.
My
shape
was
a
focus
on
building
an
api
marketplace
application
at
that
time.
But
then
they
realized
that
you
know
two
things
number
one.
They
realized
that
there
wasn't
at
that
time.
C
C
So
because
of
all
this,
they
decided
to
sell
the
marketplace
and
then
keep
on
and
move
on
with
the
epidemic,
and
that's
how
kong
was
born,
that
the
name
kong,
as
you
can
imagine
you
know,
is-
is
related
to
machine
learning.
You
know,
there's
a
jungle
thing
related
to
it,
and
then
they
decided
to
move
on
with
this
team.
Like
you
know,
kong,
the
gorilla
as
a
matter
of
fact
that
will
be
the
strongest
ape
of
these
species.
That's
the
story
behind
it.
B
It's
a
it's
a
great
name,
stay
with
me
here.
This
is
going
to
be
very
cheesy
the
moment
that
I
knew
that
joining
kong
was
the
right
thing
to
do
for
me
was
when
I
discovered
the
puns
channel
in
slack.
It
is
full
of
everything
that
is
exactly
how
you
imagine,
but
we
do
things
like
our
internal
internet
is
not
confluence.
It
is
kong
fluent.
When
somebody
achieves
some
kind
of
victory
or
joins
the
company.
We
say:
congratulations.
B
A
C
Oh
yeah,
absolutely
kind
of
you
know
so
then
the
the
kong
project,
the
open
source
project
kind
of
you,
know
a
massive
adoption
all
over
the
world.
Recently
we
reached
the
historical
number
of
200
million
downloads,
that's
for
the
open
source
project
and
then
you
know
in
terms
of
github.
For
instance,
we
have
reached
the
number
of
40k
stores.
B
D
B
As
congress,
we
represent
24
different
countries,
which
is
an
incredible
feeling
pretty
much.
Every
single
call
I
get
on
internally
or
externally:
there'll
be
people
joining
from
different
countries
or
even
different
continents.
Like
even
my
claudia
between
the
three
of
us,
we
are
representing
three
continents.
Right
now
means
you've,
got
people
who
speak
different
languages.
People
who've
got
different
experiences
and
backgrounds
and
different
schools
of
thought,
and
this
is
such
an
an
enriching
and
creative
environment
to
sit
in
the
other
values
are
real.
B
You
know
we're
we're
a
bit
of
an
overused
word,
but
we
are
authentic.
We
are
ourselves,
we
are,
you
know,
not
afraid
to
be
humans,
we're
unstoppable.
We
are
customer
obsessed.
We
want
to
achieve
all
of
the
outcomes
by
partnering,
with
our
customers
partnering
with
the
open
source
community
that
they
can
achieve
the
c
that
is
for
champions.
You
know
we
we
strive
to
be
the
best
we
set
a
very
high
bar,
but
an
achievable
bar
for
ourselves,
and
the
e
at
the
end
is
for
explorers
we're.
B
You
know
everybody's
very
curious
about
the
world.
Intellectually
curious.
We
we're
innovating.
We
are,
you
know,
we're
not
just
doing
api
management,
we're
saying
apis
are
dead,
we're
doing
something
different,
because
that
is
what
connectivity
needs.
We
are
really
exploring
and
peeling
back
the
layers
of
the
onion
and
questioning
the
assumptions
to
make
sure
that,
actually
what
what
we
do,
what
we
focus
on
what
we
say
are
all
the
right
things,
not
just
the
things
that
people
are
used
to
hearing.
A
B
I
think
pretty
much
all
business
leaders,
regardless
of
which
domain
they're
in,
are
aware
that
joining
the
api
economy
being
api
first
thinking
about
apis
is
not
just
a
nice
to
have
but
a
must-have.
This
is
because
you
know
we're
all
familiar
with
the
you
know:
the
nice
shiny
term
of
digital
transformation.
B
We
are
trying
to
power
digital
experiences
for
our
customers.
For
our
employees.
We
are
businesses
that
are
composed
of
digital
assets.
All
of
this
is
underpinned
by
that
connectivity,
underpinned
by
apis
and
microservices,
that
are
the
that
bridge
between
those
different
applications,
and
by
powering
this
connectivity
through
apis.
You
can
accelerate
your
time
to
market.
B
You
can
improve
your
operational
efficiency.
This
is
because
the
you
need
to
build
for
consumption
right.
You
need
to
make
sure
that
these
services,
that
internal
developers
or
external
developers
build,
are
well
designed
that
they
are
reliable,
that
they
are
secured
and
also
that
they
are
then
discoverable
that
you
can
access
them
and
therefore
this
contributes
to
a
reuse
rather
than
rebuild
mindset.
B
Obviously
there's
a
bit
of
culture
change
that
goes
along
with
that.
But
the
end
result,
if
you
do
it
right,
is
that
you
end
up
already
having
built
a
huge
amount
of
work
on
your
projects
before
you've
even
begun,
because
there
are
these
apis
in
place
that
you
can
easily
find
and
access
and
then
reuse,
you
don't
spend
time
rebuilding
stuff
and
then
on
the
other
half
of
it.
B
There's
there's
a
very
valid
focus
right
now
on
data
privacy
and
security,
and
if
we
have
no
visibility,
no
control
over
how
our
data
is
moving
between
different
systems
between
different
locations,
then
each
one
of
these
endpoints
is
a
security
risk.
So
by
powering
apis
by
securing
them
with
a
platform
like
com,
you
can
ensure
that
every
single
one
of
these
connection
points
is
managed
that
it's
well
secured
that
you
actually
are
compliant
with.
B
B
I
think
a
good,
a
good
segue
into
this
is
obviously
there
is
a
valid
needs
deal
for
the
traditional
api
management
that
we
became
familiar
with
five
years
ago.
You
know
maybe
10
years
ago,
but
we,
like
you,
said
at
the
beginning.
It's
a
controversial
statement.
We
are
saying
that
api
management
is
dead.
In
fact,
our
ceo
ivy
he
stood
up
at
our
summit
a
couple
of
years
ago
and
opened
up
the
event
with
this
statement.
B
The
the
rationale
here
is
that
we
used
to
think
of
apis
as
restful
services
that
sat
in
our
dmz
or
next
to
our
esb
and
exposed
data
in
a
nice,
pretty
json
format
to
the
web.
It's
all
about
that
edge
gateway
management,
but
this
is
a
subset
of
the
use
cases
that
we
see
today.
Probably
that's
that's
a
good
place
to
start
off
claudio
there's.
This
is
still
a
valid
use
case.
The
platforms
in
which
this
needs
to
run,
though
that
has
changed,
maybe
go
into
that
a
bit
more.
C
C
Other
platforms
were
were
in
a
single
run
time,
but
this
time,
because
of
several
reasons,
you
know
cloud
computing
micro
service
phenomenon,
customers
are
going
to
other
places.
They
want
to
run
their
applications
in
as
a
container,
for
instance,
in
other
words,
to
run
docker
and
then
going
a
little
bit
further
to
run.
Kubernetes
cloud
number
one
cloud
number
two,
in
other
words,
to
have
these
moody
cloud
deployments.
C
C
This
time
you
know
in
that
time
you
know
very,
very
easy
to
go
like
you
know,
the
customers
wanted
to
implement
a
new
protocol,
a
new
platform,
a
new
architecture
pattern
and
their
deployment,
and
so
on.
Usually
the
the
answer
for
these
were.
Like
you
know,
I
am
so
sorry
this
product
over
here
doesn't
support
this
protocol
this
time,
maybe
next
year,
because
you
know
that
is
supposed
to
be
available
in
next
version.
We
are
about
to
launch
next
year,
this
kind
of
thing
not
anymore.
C
C
At
the
same
time,
they
are
building
very
strong
devops
teams,
very
strong,
effective
tunes
and
then-
and
then
these
teams
are
looking
for
best
of
breed
in
terms
of
technology
options
offered
for
by
the
market,
diversity,
so
kind
of,
and
then
again
we're
not
talking
about.
You
know
your
logical
application,
your
application,
being
you
know,
being
deployed
in
a
highly
distributed
environment.
C
You
know
spread
it
out
in
multiple
runtime
the
same
time,
but
then
again
the
moody
vendor
deployment,
and
you
know
again,
customers
are
looking
for
the
best
of
breed
options.
That
is
exactly
where
kong
is
inserted.
Crude,
like
you
know,
in
these
new
ecosystems,
so
and
then
that's
why
we
think
the
api
management,
as
we
know
it,
it's
of
course
still
there.
The
hiv
is
still
there
playing
the
very,
very
critical,
important,
important
role
for
the
edge
computing
layer
to
protect
your
diesel
asset.
But
then
again
not
enough.
C
You
should
have
other
points
in
your
architecture
to
take
care
of
the
micro
service
communication
and
so
so
forth.
So
in
a
sense
that
you
know
regardless,
where
you
are
regardless,
where
you
are,
you
should
have
a
service
control
point
in
your
enterprise
architecture
to
take
care
of
the
service
conception,
take
care
of
the
micro
service
to
microsoft
or
service
to
service
communication.
C
If
you
will
that's
why
we,
we
paid
the
api
in
engineering
again
it's
dead
and
you
know
we
have
to
provide
the
next
duration
for
these
new
requirements
and
platforms
and
so
forth.
As
a
matter
of
fact,
we
position
we
position
ourselves,
not
just
the
the
epidemic
vendor
anymore,
but
instead
he
has
service
control
platform,
vendors.
A
You
know,
containers
have
been
around.
Containers
have
been
around
for
a
very
very
long
time.
I
mean,
I
think,
mainframe
yep,
you
know
solaris,
but
people
weren't
really
using
them
because
it
became.
It
was
just
that
really
the
orchestration
tooling
around
that
wasn't
mature
enough,
and
I
think
docker
did
a
lot
to.
D
A
You
know
cncf
ran
a
study.
I
think
it
was
was
it
this
year
that
showed
that
84
of
the
people
who
responded
to
their
study
were
now
running
containers
in
production.
So
how
does
that?
How
does
that
affect
or
impact
the
need
for
for
kong
api
gateways,
or
does
that
complicate
things
or
does
it
make
it
easier?.
C
A
C
Yeah,
as
a
matter
of
fact,
we
we
brought
some
be
more
precise.
We
brought
four
videos
not
that
long
around
two
minute,
long,
each
and
then
yeah.
We
wanted
to
do
to
show
them
just
to
illustrate
what
we're
talking
about
here.
What
is
this
new
platform?
What
is
the
service
control
platform
we're
talking
about
so
yeah?
C
A
A
A
D
A
C
Yeah
yeah
and
I
get
yeah
kind
of
fun,
yeah,
so
yeah.
Here
again,
my
beautiful
again,
my
beautiful
ship
cluster
deployed
in
in
gcp,
and
then
I
got
already
some
kong
run
time
to
redeploy.
As
you
can
see,
you
were
asking
about
an
operator.
The
operator,
as
a
matter
of
fact,
is,
is
deployed
redeployed
as
well
as
the
kung
fu
kubernetes.
C
So
you
can
see
there's
an
english-con
deployment
in
there.
That's
my
english
controller
and
then,
as
you
can
see,
there's
only
one
pod
running
this
time.
Yeah,
here's
the
part,
but
then
again
I
got
in
background.
I
got
a
a
loop,
injecting
request
to
my
cluster
and
then
on.
Your
right.
You
can
see
punk
vitals
on
vitals
is
one
component
we
provide
in
our
kong
enterprise
edition.
C
Convert
virals
is
responsible
for
keeping
track
of
the
main
kpi's
of
the
gateway.
Like
you
know,
the
number
of
parts
that
had
been
processed
proxy
latency
data
upstream
latency-
they
they
start
caching
so
on
and
then,
as
you
can
see
over
here,
is
my
pod
ingress
english
dash
con
card
running
and
then,
as
I
said
before,
I'm
already
injecting
some
requests.
C
C
Similarly,
we
can
reduce
the
number
of
quads
along
the
way
so
kind
of
a
very,
very,
very
nice
scaling
capability
provided
by
kinematics.
I'm
doing
as
you
can
see,
I'm
doing
it
manually
so
kubernetes
brings
another
ability,
which
is
what
we
call
the
hpa
horizontal
product
scaler.
So
this
time
these
we're
achieving
elasticity
the
number
of
parts
this
time
is
going
up
or
down
depending
on
the
traffic.
C
B
I
think
it's
it's
funny,
because
a
lot
of
the
companies
that
we
see
and
that
we
work
with
are
still
using
and
having
to
work
out
how
to
support
apis
and
services
deployed
in
legacy.
Environments
quote
unquote
legacy
or
on-premise
environments,
their
own
data
centers,
but
so
many
of
those
companies
are
also
running
kubernetes,
running
containers
and
pushing
kind
of
at
that
other
end
of
adoption.
B
So
we
see
a
lot
of
use
cases
where
it's
not
just
focusing
on
the
kind
of
the
stuff
that
we
saw
a
few
years
ago
or
not
just
focusing
on
you
know
the
really
nice
shiny
stuff,
but
actually
use
cases
that
have
to
stand
both
both
ends
of
that
spectrum,
and
I
think
this
is
where
going
back
to
what
you
said
a
minute
ago.
Cardio.
This
is
one
of
the
reasons
why
I
absolutely
love
our
technology.
It
is,
it
is
the
same.
It
is
consistent.
B
It
is
equally
capable
and
performant
in
all
of
these
different
deployment
environments.
It's
not
kubernetes
specific.
It
is
not
bm
specific.
It
is
not.
You
know,
cloud
provider
specific.
It
is
exactly
the
same
thing
that
you
can
deploy
in
each
one
of
these
distributed
environments.
So
ultimately,
you
end
up
with
these
kind
of
localized
or
federated
like
governance
and
security
enforcement
points,
but
then
all
of
this
visibility
bubbles
up
to
what
is
to
use
the
phrase.
You
know
that
single
pane
of
glass
that
gives
you
visibility
across
that
whole
distributor.
D
A
No,
I
just
when
we
were
talking.
I
think
it
was
last
week
you
were
saying
about
you,
know
how
everything's
constantly
evolving
and
increasingly
distributed
landscapes,
and
that
that
you
know
you
need
to
be
able
to
apply
security
and
governance
consistently
everywhere.
How
does
how
does
kong
help
with
that.
B
I
mean
I
know:
cardio
has
prepared
something
to
actually
show.
I
think
that
that
will
be
a
really
great
way
of
bringing
that
to
life,
but
this
is
something
I
did.
I
presented
this
at
an
event
we
did
last
week,
I'm
speaking
to
customers
almost
daily.
This
is
the
biggest
challenge
that
I
see
companies
struggling
with
today.
B
But
the
trouble
with
this
approach
is
that
this
puts
all
of
the
owners
straight
onto
the
developer,
to
make
sure
that
they
manually
change
how
they
work.
They
stay
up
to
date
with
what
best
practices
are,
and
that
typically
means
reading
some
out
of
date,
page
in
confluence.
That
says
this
is
how
you
should
follow
api
best
practices
within
our
organization.
B
So
what
we
enable
this
api
ops
approach?
This
is
going
from
that
manual
workflow
to
a
self-serve
automated.
This
is
using
the
fact
that
kong
supports
declarative
configuration
in
insomnia
from
the
very
beginning
of
the
design
of
your
apis
through
to
the
deployment
through
to
the
configuration
of
the
different
runtimes.
B
C
C
A
C
So
here's
the
the
exactly
sync
yeah,
let
me
start
a
video,
so
exactly
the
same
api
clustering
there.
So
then
again,
my
yeah
first
of
all,
just
another
quick
introduction,
so
the
my
api
ops
process
is
gonna,
show
you
end-to-end
api
provisioning
process,
so
kind
of
starting
with
the
api
spec
edition
and
then
I'm
using
it.
As
a
matter
of
fact,
mike
I
got
inspired
by
the
latest
blog
posts.
C
A
C
All
right,
okay,
good!
So
again,
you
know,
beginning
with
the
api
spec
editor
we're
going
to
to
put
the
cicd
pipeline
in
action
and
then
the
ec
icd,
of
course,
will
be
responsible
for
publishing,
not
just
the
api
spec
to
our
to
our
api
database
runtimes,
but
also
to
publish
the
api
documentation,
not
that
portal,
not
in
our
in
our
kung
enterprise,
that
work
and
then,
of
course,
they're
totally
responsible,
for
you
know
exposing
the
documentation
to
the
developers
and
so
forth.
C
So
yeah
me
use
my
github
repository
over
here.
This
is
my
on
my
oops,
and
this
is
in
sony
designer,
as
I
said
before,
beginning
with
the
sunday
designer,
which
is
the
api
spec,
editor
kind
of
you
know
and
that's
the
tool
the
api
designer
is
supposed
to
use
to
craft,
to
define
to
to
enhance
the
perspective.
We've
been
working
on
and
then,
first
of
all,
I'm
going
to
send
some
requests
to
the
api
identification,
which
I
know
you
should
there's
nothing
there.
The
ipad
is
totally
there's.
C
There's
no
api
deployed,
there's
nothing
there,
yeah
so
and
then
api
is
the
funny
design.
You
can
expect.
There's
a
linkedin
process
running
in
the
background,
so
every
time
I
got
a
syntax
here,
I'm
going
to
get
a
warning
saying
you
know
alarming.
I'm
learning
about
these
syntax
here
and
then,
of
course,
I'll
have
two
opportunities
to
get
it
agreed.
C
At
the
same
time,
on
my
my
right,
you
can
see
the
visual
presentation
of
my
api
spec
until
then,
again
in
my
left,
you
can
see
the
opening
the
op
api,
3.0
spec
and
then
on
my
right,
the
user
representation
of
it,
not
just
the
visual
presentation,
but
actually
you
can
try
the
api
out,
send
your
request
to
it
and
so
on.
Another
important
capability
provided
by
the
the
editor
is
the
the
capability
to
inject
some
economic
extensions
in
order
to
define
policies.
C
As
a
matter
of
fact,
my
are
going
to
I'm
going
to
deploy
this
api
over
here,
along
with
two
policies.
At
the
same
time,
the
first
policy,
as
you
can
see
up
here,
the
rate
limiting
policy
gonna,
define
a
three-minute
rate
limiting
policy,
meaning
that
I'm
supposed
to
send
only
three
requests
at
a
given
minute
to
the
api
gateway
and
then
the
second
policy.
That's
the
another
integration
point
with
the
openshift
more
precisely
keep
rook.
So
in
order
to
consume
this
api,
I
am
supposed
to
send
key
closed
credentials.
C
C
C
So
right
after
right,
after
having
the
ci
cd
pipeline
done,
I
can
see
the
new
services
new
route
new
plugins,
the
plugin
is
considered
a
plugin
as
an
extension
for
the
api
gateway,
so
each
one
of
these
plugins
again
is
responsible
for
an
event
policy,
as
I
showed
you
before,
like
you
know,
two
policies,
the
rate
thinking
policy
and
the
opid
connect
policy
with
the
key
clock.
Integration
is
already
there
deployed
and
configured
for
me.
C
Likewise,
here's
my
dev
developer
portal
and
here's.
My
exactly
the
same
api
spec
deployed
documentation
this
time
deploy
the
label
for
my
developers,
so
I'm
going
back
to
the
editor
and
then
this
time
I'll
be
able
to
send
requests
to
my
api.
Finally,
but
since
I
got
the
the
key
cloak
integration,
I
am
supposed
to
send
keep
cloak
credentials.
Otherwise
again
it
wouldn't
be
possible
to
consume
the
api.
So
let's
say
I'm
going
to
send
wrong
credentials.
C
C
C
A
Now
that
everyone
knows
exactly
what
your
gmail
email
address
is
yeah,
I
I
wrote
it
down
when
you,
I
think
it
was
when
you're
logging
in
how
do
you?
How
can
people
get
in
touch
with
kong?
I
mean
I'm
guessing.
They
don't
want
to
send
email
to
your
gmail.
Do
you
guys
have
a
you
know
claudio
at
or
is
there
like
a
you
know?
It
usually
would
do
this
at
the
closing,
but
I
figured
I'd
sneak
it
in
right
here,
seeing
as
your
gmail
address
was
just
all
over
all
over
the
screen.
There.
C
Yeah,
so
again
you
know
you
can
get
in
contact
using
my
email,
my
copy
to
me
you
if
you
need
to
use
by
my
personal
one,
my
calling
basic
exactly
my
name
claudio
at
konghpu.com
and
then,
if
you
go
to
our
official
website
konghq.com,
you
have.
You
know
multiple
resources,
for
instance,
one
nice
resource
available
is
that
you
know
our
calendar.
For
you
know,
virtual
events,
like
you
know,
we
just
had
a
our
destination
event,
are
totally
focused
on
security.
C
A
Okay
or,
and
if
anyone
didn't
get
any
of
that
they
can,
they
can
send
me
an
email
address,
it's
just.
It's
just
wait,
redhat.com
and
I
can
get
you
connected
with
them,
so
we're
we're
coming
up
on
the
top
of
the
hour.
We
still
have
some
more
time
here.
I
think
you
had
another.
Didn't
you
have
a
service
mesh
demo
that
you
wanted
to
run.
B
Oh
so
glad
you
brought
that
up
mike,
I
was
like
right.
You've
got
one
more
thing
that
I
really
want
to
make
sure
that
we've
got
time
to
talk
about
the
from
I
keep
kind
of,
I
guess,
speaking
about
the
differences
right
between
api
management
a
few
years
ago,
and
api
management
or
service
management,
as
it
is
today.
B
Api
management
a
few
years
ago,
was
only
really
focused
on
one
connectivity,
and
that
was
api
management
at
the
edge.
So
typically
at
the
edge
of
our
network,
we
host
it
ourselves
in
our
dmz
or
we
use
some
kind
of
cloud
solution.
That
is
a
centralized
deployment
in
the
cloud
that
we
then
push
all
of
our
api
requests.
B
This
is
one
of
three
key
connectivity
patterns.
The
second
one
is
powering
application
to
application
connectivity,
that
is,
internal
api
and
service
management.
For
example,
you
have
a
product
domain
within
your
business
that
releases
apis
as
services
that
are
consumable
by
other
teams
across
the
same
business.
So
people
working
within
the
customer
domain
can
reuse
apis
that
the
product
team
have
shared
with
them,
but
they
are
not
externally
consumable.
B
The
third
connectivity
pattern
is
the
hot
topic
really
across
our
industry
of
the
service
mesh
right.
This
is
the
we're
breaking
down
our
monolith
into
microservices,
and
we
need
for
that
overall,
large
application
to
be
comprised
of
all
those
tiny
little
chunks
of
code
that
talk
to
each
other
and
the
service
mesh
provides
that
security
provides
that
traffic
management
provides
the
load
balancing
to
ensure
that
all
of
these
little
nodes
that
need
to
talk
to
each
other
can
do
so
reliably
and
effectively
claudio.
I
really
should
stop
talking
and
let
you
show
it.
C
Yeah,
you
know
the
subs
have
smashed
a
tactical
pattern.
You
know
kind
of
very,
very
important,
like
you
know
again
we're
not
talking
about
the
you
know
the
external
traffic
coming
to
your
microservices
architecture,
but
instead
you
know
you
know
different
traffic
internal
traffic.
We've
got
internally,
you
know
it's
mesh
or
microservice
kind
of
you
know
it's
very,
very
important
to
keep
in
mind,
and
you
know
talking
about
microservices
talking
about
multiple
instances.
At
the
same
time,
multiple
runtime.
C
At
the
same
time,
so
kind
of
you
know
this
micro
service
to
microsoft,
communication.
It's
not
a
straightforward
topic
to
address,
and
then
you
have
to
deal
with
the
secret
breakers
load.
Balancing
health
checks,
neutral
gls.
You
have
it
to
encrypt
the
tunnel
with
these
microsoft,
microservice
communication
as
well.
So
again,
we're
not
you
know
it's
not
a
an
easy
test
to
to
implement
and
then
set.
Some
mesh
is
exactly
addressing
this
kind
of
requirements,
technical
requirements,
cool.
A
We
got
you
know
I
mentioned
we're
live
on
youtube
and
facebook
and
twitch.
We
we've
got
some
questions
that
just
came
in
from
youtube.
I
want
to
bring
this
up
here,
it's
if
I
may
so
I
don't
know
if
this
is
from
melissa
or
claudia,
but
in
context
are
we
talking
about
kong
on
os
deployment
or
kong
integrated
with
the
operating
system
in
the
latter
case?
Does
that
in
the
sorry
don't
have
my
readers
on
in
the
latter
case?
Does
that
do
for
the
model
ingress
service
routes.
C
Yeah,
I
can
take
this
one.
I
think
os
in
this
context
is
open.
That's
my
understanding.
I
might
be
wrong,
but
yeah.
If
it
is
yes
totally
right,
I
mean
we.
What
is
kong
running
open
ships?
The
song
is
at
the
po
is
a
open
ship
deployment.
Implementing
the
english
controller
is
spectacular.
Kubernetes
english
control
is
back.
In
this
sense.
We
totally
we're
totally
following
the
ingress
controller
specification.
C
So,
yes,
we
provide
crds
in
order
to
define
an
ingress,
to
modify
it
in
ingress
to
apply
policies
to
the
ingress
using
exactly
the
same
plugins.
I
showed
you
before,
like
in
oil
authentication
policy
and
rating
team
policy,
and
so
on
so
forth,
yeah.
Absolutely
we
follow
this
service
route
ingress
model
as
we
as
we.
You
know
again
in
this
openshift
deployment,
as
you
as
you
are
referring
to.
A
C
Yeah,
as
a
matter
of
fact,
we
provide
the
the
kong
specific
sexy
mesh
implementation.
It's
called
kung
mesh,
it's
totally
based
on
kuma
the
second
open
source
project
that
is
started
as
a
congo
resource
project
a
couple
months
ago
we
donated
due
to
cncf
in
in
yes.
Currently
it
is
a
cncf
project
and
then
again,
kung
mesh
is
sas
mesh
accommodation,
some
differences
when
comparing
to
guys
like
istio,
for
instance,
much
easier
to
manage,
deploy
to
to
configure
and
so
on
so
forth,
other
different,
very,
very
important
difference.
C
We
we're
totally
we're
totally
fine
implementing
your
sexy
mesh
in
a
again
in
a
hybrid
environment,
not
just
using
kubernetes
but
again
self-smashing,
defining
yourself
smash
them
in
a
multi
platform
deployment.
Like
you
know,
services,
running
kubernetes,
sensing
running
and
non-combination
questions
and
still
being
part
of
exactly
the
same
sex
and
measurement
so
yeah.
That
will
be
it.
A
A
So
thank
you
for
coming
today.
If
anyone
wants
to
know
more
about
kong,
you
can
see
the
link
here
on
the
screen.
I
was
going
to
say
click
on
it,
but
I
don't
think
that'll
work
and
then,
of
course,
you've
got
bizdev
at
konghq.com
love
to
have
you
back
claudio.
If
you
want
to
be
part
of
our
of
our
blog
postings
and
so
forth,
let
me
know:
you've
got
my
contact
information.
B
A
You
we're
very
conservative,
but
I
got
my
fingers
crossed
so
anyways.
You
folks
can
stay
on
the
bridge
if
you
want,
but
we're
gonna
wrap
it
up
for
the
day.
Michael
waite
for
the
openshift
commons
briefings
our
operator
hours,
and
we
do
this
every
wednesday
one
o'clock
eastern.
So
please
join
us.
Thank.