►
Description
OpenShift Commons Gathering
Milan Italy 2019
Title: OpenShift Managed su Azure
Speaker: Marco D’Angelo (Microsoft)
A
I
introduce
myself
I'm
marco
d'angelo
I
work
for
microsoft,
italia,
more
or
less
in
the
sense
that
I
work
for
the
last
international
who
is
in
charge
of
working
with
the
developers.
I
have
two
or
three
different
roles,
and
for
this
reason
the
presentation
plastered
not
exactly
with
a
white
shirt
but
then
you'll.
Forgive
me
maybe,
and
we
want
the
speech
towards
the
slightly
more
technological
sphere.
A
Basically,
the
presentation
today
on
what
it
is
today
is
on
a
technology
that
is
called
raw
and
jules
drawn
up
for
choices
that
what
is
rare
practically
represents
the
terminal
path
of
a
partnership
that
has
been
going
on
for
some
time,
but
that
in
recent
years,
has
become
stronger
and
stronger.
Let's
say
that
in
the
last
open
source
days,
when
we
sponsored
the
event,
people
wondered
what
let
's
go
on.
They
rightly
said.
I
never
expected
to
have
a
community
developer
conference
with
a
guy
from
microsoft
talking.
A
Yet
we
got
there
because
this,
because,
regardless
of
partnership,
regardless
of
everything
we
are
two
realities,
then
call
us
companies
walk
c
vendor.
What
you
want,
within
your
area
of
expertise,
address
us
95
percent
of
the
top
five
hundred
drafted
address,
100
percent
force,
a
global
500,
microsoft
and
president
to
be
clear
across
even
smaller
companies,
but
hey
jude.
That's
the
focus
of
the
chat
of
today
still
the
penetration,
let's
say
at
95,
exactly
100
percent.
A
We
are
therefore
here
to
discuss
what
of
the
partnership
that
date,
how
to
say
back
a
long
time
date
back
of
when
we
supported
edited
as
a
virtual
machine
on
hyper
v
data
of
when
we
brought
the
images
from
the
on
premises
world.
We
have
set
up
hazor
to
render
enterprise
linux
data
of
the
fact
that
we
have
in
Italy
all
in
all,
extremely
important
examples
of
sap-type
application,
loads
that
run
on
rella
up
and
down
sapri
Italian
industrial
production
realities.
A
A
I
tend
to
not
have
to
explain
to
you
what
the
value
of
openshift
is,
because
you
are
here
because
you
have
seen
what
are
all
the
considerations
made
on
the
evolution
of
the
platform
and
should
data
be
correlated
with
the
consideration
that
I
am
giving
you
a
slightly
more
general
consideration
with
respect
to
koper
net
and
sarko
bernet
done
well,
it
is
complicated
we
have
as
microsoft
and
the
our
implementation
by
aku
bernacer
swayze,
which
is
called
k
s
not
that
many
customers
are
in
production
right
now
with
that
stuff
there.
Why?
A
Why
the
complexity
behind
it?
All
that
needs
to
be
taken
into
account
when
doing
an
enterprise
grade
implementation
with
applications
that
if
they
stop
they
stop
the
company,
it
is
not
trivial.
The
cluster
must
be
installed
must
be
maintained,
must
be
secure
data.
Then
there
must
be
that
of
the
tool
operation,
I.e.
easy
to
install
set
it
up.
Then
it
must
be
sure
that,
as
your
developers,
step
by
de
then
by
various
types
are
able
to
take
advantage
of
the
platform.
A
For
this
reason,
what
comes
out
comes
out
that
many
of
the
challenges
behind
using
a
platform
cover
net
smail,
where
everything
is
beautiful.
What
you
want
the
world,
the
pops
everything
is
beautiful.
Our
process
already
has
the
with
system
developers
who
deal
with
network
security
in
databases
who
perhaps
really
make
use
of
methodologies
and
tools
in
the
direction
of
what
they
do,
but
the
security
part
the
mind
part
is
not.
The
delivery.
A
Part
are
things
with
which
openshift
from
this
point
of
view,
really
helps,
since
it
gives
the
possibility,,
regardless
of
the
technique
of
the
underlying
technology,,
to
abstract
a
lot
and
to
make
a
lot
predictable.
If
you
want
everything
that
is
the
process
that
pertains
to
these
three
areas
now,
why
you
choose
cooper
nate,
and
if
you
choose
open
yes,
we
said
they
have
already
explained
it,
and
it
is
almost
if
you
want
to
leave
the
task
of
the
colleagues
drafted
to
tell
you
well
as
for
me.
A
The
idea
was
to
tell
you
for
a
moment
how
you
put
these
two
things
together.
So
sorry,
it's
hot
a
bit
for
that
a
bit
because
we
won
with
the
slides
with
strength
and
garner.
Let's
take
off
the
jacket
then
and
joan
how
many
people
know
what
it
is
and
swear
in
this
audience.
Do
me
the
courtesy
of
huge
hand,
how
many
people
in
this
audience
use,
and
your
also
for
the
sitarello
as
if
to
say
institutional,
but
they
know
something
in
production
its
days.
A
Ample
room
for
improvement,
as
they
say
it's
fine.
Joking
aside
for
I,
can't
explain
what
the
cloud
is
for
microsoft,
scalables
that
plays
on
a
par
with
vs,
gp
and
behind,
and
you
know
those
things
over
there.
You
read
them
how
to
say
on
the
internettaro
gossip
every
day
the
idea
was
to
tell
what
shore
is
not
so
much
as
how
made
a
question
of
services
97
types
of
class.
I
machine
services
up
to
64
tel
annoyed,
knowing
how
to
find
it
easily
hanjour.
Is
this
hey
jude?
A
There
are
57
regions
scattered
around
the
world,
the
last
two
that
we
have,
as
it
were,
opened
and
baptized
last
month,
are
in
switzerland
and
norway
over
the
course
of
2020.
We
will
open
two
more
regions
in
europe,
I
won't
say
even
country.
It
looks
ugly
what
makes
the
difference
compared
to
kooper
neither
tense
complicated.
It
takes
openshift.
Sometimes
the
infrastructure
is
complicated
and
when
we
say
that
the
infrastructure
is
complicated,
we
need
backups.
We
need
resilience,
we
need
business
continuity.
A
We
spend
treviglio,
they
don't
alter
3
billion
per
rb
Lyonna
semester
to
open
new
data
centers
having
to
deal
with
a
microsoft
data
center
or
putting
an
open
class
about
microsoft,
and
jules
means
that
somewhere,
where
we
choose
to
put
it,
we
have
at
least
at
least
two
data,
centers,
1
and
2,
basically
3,
because
we
are
finishing
the
rollout
of
the
zonin
concept.
So
you
will
find
three
resilience
zones
for
each
of
the
57
areas
in
which
we
have
always
given.
A
Ok,
this
I
needed
to
pass
a
message
related
to
the
fact
that
we
don't
have
to
learn
how
to
build
infrastructure.
Hardly
anyone
you
will
have
seen
the
latest
things.
The
latest
types
of
data
centers
that
we
make
available.
It
takes
three
months
from
design
to
depressing
a
data
center
because
it
is
not
present.
They
look
like
this
stuff
here
we
don't
have
one
in
real
production.
A
Yet
we
are
like
them
to
say
literally
sinking
these
days,
with
the
approval
of
all
that
the
control
for
environmental
impact,
the
fish
by
hand
that
extra
degree
that
a
data
center
gives
them
in
the
waters
of
northern
Europe.
But
basically
this
is
the
same
as
saying
why,
generally
yes,
he
chooses
or
works
with.
A
He
has
with
him,
because,
if
I
put
a
short
production
suation,
what
you
risk
is
that
the
man
claude,
is
actually
not
only
branded
with
an
availability
of
59
of
ability,
but
because
actually
this
thing,
then
there
is
one
of
the
advantages
of
the
clouds
for
scalable
and
also
as
regards
let's
say,
openshift
and
the
fact
that
these
things
can
be
tested.
We
can
verify
it
without
implementing
it
without
investing
a
kidney
and
then
decide
if
we
see
that
the
premises
and
promises
are
kept
eventually
going
into
production.
A
A
What
does
microsoft
have
we
do
and
we
are
at
the
moment
more
or
less
the
largest
company
in
terms
of
capitalization
in
the
world,
simply
because
there
is
a
big
return
from
this
investment.
Ok
and
that's
the
end
of
the
chat
about
now
I
get
angry.
Sir
I
am
I
would
have
made
a
what's
more.
The
king
pulled
this
stuff
out
here,
which
is
why
microsoft
works
with
redacted.
A
Why
microsoft
for
five
years
now
has
gone
back
to
working
not
only
with
companies,
office
exchange
companies,
for
goodness
sake,
but
has
gone
back
to
talking
to
developers,
and
when
we
talk
to
developers
we
realized
that
we
can
no
longer
be
just
software
vendor
loan.
It
makes
sense
to
be
the
ones
selling
dot
net
sequels
and
windows
server.
If
developers
use
a
flea.
Nikon
is
based
on
microsoft.
A
If
linux
hesitate
for
translation,
if
ever
sequel,
posner,
kewell
and
reds
are
each
for
its
areas,
the
preferred
database
areas
for
caching,
rather
than
making
transactions
rather
than
being
scaled
wherever
it's
perfectly
fine
that
this
is
the
case
because
it
no
longer
exists.
I
told
you
microsoft
that
it
only
has
to
sell
sequel
servers
that,
for
heaven's
sake,
a
great
product
like
windows
with
aias,
etc,
etc,
but
it's
not
just
our
business
anymore.
So
what
is
it
and
you're
drafted
for
shift
or
aro,
as
we
say
in
Italian?
A
Basically,
it's
a
fully
manager
implementation
of
openshift
engineered
with
drafted,
therefore,
earth
earth,
the
sells
microsoft.
You
are
the
few
users
who
use
island
in
production.
You
could
even
turn
it
on
in
your
subscriptions.
It
is
not
the
only
way
that
exists
to
run
openshift
up
and
down.
There
is
still
the
possibility
of
the
farm
action
temple
that
puts
the
master
in
od
and
everything
we
are
used
to
seeing
in
terms
of
openshift
farm
management,
but
it
is
an
implementation
that
carries
with
it
a
90,
99
sla.
A
The
99
means
that
if
viridia
goes
down,
we
do
it
with
was
the
credit
note
of
the
of
the
cost.
Let's
say
of
the
disbursement
already
made,
which
is
responsible
for
being
sufficiently
open,
then
image
sufficiently
open.
We
will
be
present
at
the
open
spaces
with
renate.
We
will
tell
cases
of
customers
who
are
using
other
cases
of
customers
who
said
I,
see
it
as
a
principle
of
do
things
like,
say
a
certain
type.
I
have
to
change
so
much
of
a
structure
on
which,
in
rome
green
a
match
in
90
percent
of
the
time.
A
If
you
imagine
what
is
a
basic
implementation
of
a
rovin
shift,
suasion
class,
what
it
would
look
like,
it
would
look
like
a
series
of
virtual
machines
that
have
the
role
of
master,
basically
30
blueprint
machines
of
infrastructure
that
make
das
that
the
set
and
application,
therefore
always
sorry,
but
that
virtual
of
which,
as
on
any
other
cloud
provider,
you
manage
and
choose
the
part
of
premium.
Non-Premium,
storage,
expensive
or
less
the
component
of
machines
with
the
need
to
manage
them,
because
it
is
true
which
isolates
it
turns
on
logging.
It
does
the
ealing
part.
A
It
passes
it
from
one
data
center
to
another
from
one
region
to
another,
but
in
any
case
they
are
virtual
machines
that
exist
and
you
need
to
know
that
they
exist.
There
is
the
balancing
component
there.
It
is
the
need
to
expose
all
the
external.
There
is
the
need
to
protect
what
we
spend
on
the
outside.
A
There
is
the
possibility
or
the
need
to
manage
the
security
components
related
to
authentication
and
isolated
directories
rather
than
circles,
within
the
store
of
all
the
access
credentials
to
what
can
be
database
application
components
and
anything
else
that
are
above
and
below
the
infrastructures.
This
stuff
here
is
implemented
in
the
air
and
has
been
available
for
the
past
two
years
with
what
is
called
joint
reda
support.
A
If
a
customer
hazor,
something
doesn't
add
up,
call
and
ur
support
and
ur,
then
called
0
and
dat,
and
vice
versa,
what
this
stuff
entails
and
how
does
it
change
the
rules
of
the
game?
Openshift
suasion,
this
thing
entails.
We
were
saying
that
there
are
clusters
to
manage.
There
is
the
security
part,
how
to
say
about
application,
infrastructure
and
logic
having
to
manage?
There
is
a
need
to
also
verify
how
to
monitor
and
operate
the
vm,
if
only
to
make
efficiency.
A
One
of
the
advantages
of
the
scalable
cloud
that
I
turn
on
and
scale,
then,
if
I
don't
need
it,
I
spend
I,
make
sure
to
scale
down
resources
when
I
don't
have
a
load
so
as
to
save
all
this
stuff.
Here,
if
a
sky
cluster
in
production
using
the
vm
paradigm,
the
management
is
loaded.
When
we
move
on
to
give
what
happens.
A
Sorry
a
little
too
much
that
all
how
much
this
infrastructure
here
we
absorb
it,
we
make
it
disappear
and
what
remains
simply
the
management
for
me
is
administrator
of
the
components
of
must
integrate
with
the
tv
directory
of
the
company
if
I
want
it.
If
I
want
access.
Also,
the
cause
of
administration
of
the
principles
is
done
with
adi
credentials.
I
have
to
manage
secrets
using
a
secure
store.
A
I
will
certainly
not
go
to
put
the
configuration,
disturbances
somewhere
else,
so
I
will
use
another
service,
which
is
to
look
for
volts
if
I
want
otherwise
francesco
data
somewhere
else.
If
you
want,
with
light
rights,
step,
has
the
right
to
use
an
hsm
module
to
allow
us
credential
certificates,
and
basically,
at
that
point
it
is
only
the
administration
console
and
app
management
going
forward.
So
what
comes
out
is
that
the
focus
component
as
a
person
iti
as
a
developer
as
another
sky
up
the
application
components.
A
Those
do
not
see
anyone,
because
he
is
my
father,
therefore
create
contain
containers,
manage
the
automation
component
of
the
builds
choose,
the
one
that
is
mine
and
create
my
part.
I
said
to
yes,
they
say:
I
want
to
use
and
you're
the
rocks
rather
than
guitar,
rather
that
jean
king
from
isili,
that
is
in
variant.
A
As
far
as
let
me
say
it
touches
us
regarding
the
focus
of
red
hat
and
microsoft
is
all
the
lowest
level
orchestration
part
networking,
storage
registry,
telemetry
security
up
to
the
infrastructure
component,
then
the
cable
calimera
that
by
mike
and
it
doesn't
work
in
this
specific
case.
It
doesn't
exist,
but
in
the
interaction
component,,
let's
say,
integration
is
stuff
that
we
take
care
of,
and
it
is
contractualized
between
the
customer
component
and
the
microsoft
drafted
part.
A
So
we
basically
said
what
In
my
opinion,
I
take
home:
the
things
that
give
a
minimum
of
information
the
more
they
are,
the
high
reliability
component
and
not
scaling
given
data
as
a
feature
of
the
platform.
The
product
scales
without
having
to
make
investments
its
where
we
are
is
the
first
party
age
of
service
component,
be
a
3
sparti
urging
means
it's
not
a
blueprint.
A
Show
you
in
between
the
second
is
what
then
allows
me
to
manage
the
security
if
I
have
made
an
investment
entirely
on
the
cloud
component?
If
I
have
a
hybrid
investment
for
which
I
have
an
openshift
cluster
with
applications
in
the
cloud?
Is
a
series
of
application
work
that
are
still
on
premises,
and
these
must
and
I
have
to
tell
you.
The
networking
component
below
is
now
meant
to
help
and
not
make
it
complicated.
A
Once
this
is
done,
once
it's
set
up
it's
another
of
the
things
that
you
mentioned,
and
today
we
won't
show
you
how
to
say
everything
natively,
because
then
I'll
explain
how
to
do
it
to
see
it
together.
Another
of
the
things
the
integration
with
active
directory,
for
which
I
can
login
to
my
board
or
for
in
some
specifi
container
platform,
you
saw
earlier
by
logging
in
using
credentials
active
directory.
Does
this
know
what
can
be
useful?
A
What
can
be
useful
can
be
useful
if
my
company
e
implements
or
implements
giarty
directory
for
the
security
of
those
who
are
the
environments
a
gis
prize,
but
it
can
also
be
useful
because
the
tv
directors
of
and
swears
then
binds
with
those
who
are
some
quite
important
services,
which
are
the
multi
factor.
Authentication
services,
therefore,
either
free
or
already
paid
for
in
the
service
possibility
of
doing
mfa
with
mobile
identification
token
up
with
the
control
of
those
who
are
access
protection.
A
Therefore,
the
control
of
who
accesses
my
cluster
from
where,
in
real
time,
ip
geography,
time
in
area
etc.
therefore
another
example.
If
you
want
of
those
security
issues
which
I
can
not
but
take
into
consideration,
but
that
if
the
work
as
a
cloud
platform
as
the
days
can
come
back
already
available,
another
aspect
how
they
were
before
the
virtual
network
integration
component,
the
virtual
network
integration
server
component.
We
said
why
not
because
microsoft
has,
and
now
it
is
become
the
hammer
for
all
the
nails.
A
We
are
perfectly
aware,
and
we
are
the
ones
who
generated
at
the
time
the
problem
of
servers
under
the
desks
in
75.
We
know
that
workloads
remain
for
a
thousand
reasons.
They
can
be
linked
to
industry
4.0
issues
with
production
lines
that
are
the
action
in
the
factory.
How
can
you
bring
it
up
or
to
scenarios
where
data
collection
or
the
next
data
must
be
done
in
the
field?
A
One
of
the
most
important
functions
that
we
tell
you
is
the
networking
cartoon
component
behind
the
shore
that
the
one
that
I
allows,
for
example,
with
services
like
express
round
to
open
a
direct
peering
between
what
is
my
network
and
the
networking
component
of
hazor
by
matching
what
my
network
mpls,
rather
than
talking
to
my
connectivity
provider
and
saying
I,
don't
want
pass
on
the
internet
av
plan,
wolves
n.
I
want
to
speak
directly
to
cut
latency
rather
than
to
guarantee
security
in
communications
with
it.
Also.
A
This
is
a
set
and
mines,
a
problem
that
historically
applies
ground
to
air,
but
which
has
been
expressly
having
are
services
such
as
we
are
at
provision
for
hoped,
and
the
last
aspect
was
the
unified
support
issue,
not
because
microsoft
is
above,
we
claim
to
be
the
ones
who
support
everything
by
then
dropping
the
tickets.
But
the
issue
is
all
that
the
infrastructure
of
that
part
below
I
was
saying
is
managed
by
us.
A
It
is
maximum
responsibility,
it
will
not
be
making
it
work
as
we
do
it
for
the
approximately
970
enterprise
customers
that
we
have
only
in
northern
Italy
and
which
count
an
enterprise,
consumption
of
and
games.
Let's
say
that
enterprise
I
mean
from
about
200
km,
but
it
is
also
for
the
part
if
you
want,
if
you
want
reality
so,
regardless
of
how
I
introduce
myself,
if
clientele
given
microsoft,
customer
support
will
be
consulted.
These
basically
are
all
things
that
serve
to
pass
the
message
of.
A
We
have
the
awareness
we
said
we
started
the
discussion
saying
the
data
openshift
is
a
response
to
a
complicated
problem:
how
to
do
cooper,
nate
es
in
a
serious
and
functional
way.
It
is
again
if
we
implement
it
up
and
down
doing
what
pointing
out
that
also,
the
security
part
tends
to
be
linked
to
authentication
issues
we
mentioned
in
the
part
you
had
tls
sanders,
the
component
of
certificates
that
can
lead
to
using
mine
directly
the
part
of
networking,
but
also
the
part
of
hardening.
A
Let
me
say,
basically,
one
of
the
complexities
that
is
sometimes
noticed
is
in
terms
of
security.
How
long
it
takes
a
vendor
to
implement
what
are
the
fixes
of
security
that
are
made
available
to
the
community
in
this
case
on
the
gubernatis
source
project
compared
to
open
shirt?
What
you
see
here
tends
to
be
when
openshift
brings
home,
what
are
the
fixes?
A
As
we
said
before,
concluding
the
service
for
sparti,
all
the
other
pieces
of
jor
are
placed
in
front
of
or
in
parallel
to
complementary,
harper
or
increase.
What
are
the
characteristics
now
in
terms
of
experience
once
you
have
implemented
it
once
we
have
done
az,
open
shirt,
client.
We
went
to
the
regional
portal,
we
created
the
cluster
aro
from
there
on
the
management.
Experience
is
exactly
the
same,
so
what
I
would
have
done
with
the
classic
tours
of
openshift
or
yes,
login
or
yes,
new
project
os
in
you
up
are
one
hundred
percent.
A
A
I
need
it
to
get
things
done
well
by
definition,
but
quickly,
so
that
I
can
quickly
provision
test
environments
and
open
ones
to
make
tea,
probably
production
environments,
and
if
something
goes
wrong,
if
the
project
is
delayed,
are
considerations
to
skate
dow
olos
that
the
app
of
the
application
is
something
literally
built
into
the
platform.
I
have
to
buy
hardware
I
don't
have
to
over
audition.
I
don't
have
to
do
things
kind,
ok,
I
close
with
two
considerations.
A
If
you
have
any
to
find
out
what
are
the
dates
of
these
workshops,
we
said
milan,
rome.
If
we
see
what
is
there
a
request,
we
will
also
organize
others.
We
will
then
be
present
at
the
re
dat
open
source
days
in
milan
and
rome
at
the
end
of
november,
beginning
of
december
to
make
interactive
demos
in
a
boot
with
the
component.
Let's
say
anything
if
you
have
other
information,
requested,
etc.
other
than
a
term
which
are
around.
You
can
find
me
on
linkedin
and
on
twitter.