►
From YouTube: Customer Keynote: 3 Years Later at ING - Thijs Ebbers & Sandeep Kaul OpenShift Commons 2022 Detroit
Description
Customer Keynote: 3 years later at ING
Red Hat OpenShift Commons 2022 @ Kubecon/NA
Detroit, Michigan
October 25, 2022
Speakers:
Thijs Ebbers ING
Sandeep Kaul ING
https://commons.openshift.org/gatherings/kubecon-22-oct-25/
A
A
Hi
I'm
Sandeep,
basically
one
of
the
managers
at
ING.
So
what
do
I
do
as
a
manager?
My
work
is
usually
about
abstracting,
the
mundane
tasks
that
an
Enterprise
can
throw
at
our
Engineers
so
that
my
team
can
have
time
for
awesome
engineering
and
that's
been
a
privilege
to
be
in
ing
so
or
something
about
Ing
and
a
bit
of
banking,
so
banking
in
its
current
form.
If
you
look
at
it,
it
has
been
same
since
since
the
14th
century.
A
A
That's
pretty
long
back
time
right,
but
we
have
a
bit
of
digital
as
well.
So
we
started
our
first
digital
Endeavor
in
1997
as
ING
Bank
Direct.
It
was
basically
a
bank
without
branches
offering
they're,
really
nice,
attractive
savings
accounts
and
some
retail
products,
so
that
so
so
with
that
you
can
say
ing
had
Innovation
at
its
core
right.
A
That's
how
we
do
it
basically
and
to
look
at
it
from
how
do
we
really
do
it
seriously?
We're
also
part
of
a
strategic
partner
for
our
Center
for
climate,
financial
and
finance.
Basically,
what
we
do
there
is
develop
and
help
building.
You
know
sustainable
ways
of
decarbonizing
the
economy
so
and
our
own
internal
targets
are
typically
targeted
towards
2050,
which
is
20
years
before
the
actual
Target
that
that
most
of
the
other
Enterprise
are
doing
so
I'll
give
to
ties
for
the
next
one.
B
A
B
Why
am
I
on
stage
today?
Is
it
because
it's
good
for
my
employer's
corporate
image?
Is
it
because
you
want
to
invest
in
a
relationship
with
redhead?
Is
it
because
we
want
to
contribute
to
this
community?
All
of
those
are
true,
but
what
gives
me
the
energy
to
be
here
on
stage
in
that
movie
made
me
realize?
Why
did
I
jump
all
those
corporate
Hoops
to
get
here?
B
B
Almost
daily.
We
see
in
the
media
data
Lake
which
went
somewhere,
etc,
etc.
Each
data
language
is
a
failure
of
us
to
society
and,
looking
at
that
movie,
I
realized
why?
Why
is
that
happening
in
that
movie?
The
main
character
took
a
manual
and
suits
in
the
garbage
bit,
stating
your
opponents
and
veggies
as
well
and
are
using
that
knowledge
against
you,
that's
not
fiction.
B
B
B
B
B
So
being
on
stage
on
openshift
Commons,
you
might
assume
that
ing
container
hosting
is
always
using
openshift.
That's
not
the
case.
We
are
using
openshift
for
five
clouds,
but
we
are
also
using
other
Cloud
vendors
and
using
their
native
offerings
the
the
theme
of
the
movie.
We
have
an
F35
and
we
added
an
F14
and
an
F-18
to
our
lineup.
All
three
can
take
off
and
land
on
an
aircraft
carrier
and
all
three
can
basically
take
the
same
set
of
Ordnance.
B
To
be
quite
honest,
I've
been
struggling
with
this
slides
how
to
convey
this
message
and
seeing
the
movie
I
realized
this.
This
is
what
basically
the
essence
of
what
we
are
doing.
This
is
your
mission
plan
and
this
Mission
plan
for
us
wash
to
build
the
best
possible
container
hosting
platform
for
immutable,
modern
workloads.
B
Regretfully,
there's
lots
of
people
in
my
cooperation
in
this
industry
who
think
it's
a
good
idea
to
put
something
externally
to
that
F-35
and
I.
We
have
to
the
people
from
lockage
in
the
room.
They
can
probably
explain
even
better
than
I.
Why
that's
a
very
bad
idea,
so
we
have
an
intended
purpose
and
we're
sticking
to
that.
We
are
very
opinionated.
B
So
no
is
a
valid
answer
and
that's
what
happens
with
our
analytics
colleagues.
We
were
not
in
a
position
to
service
them
because
we
didn't
prototype
within
the
engineer,
a
service
for
high
performance
computer,
so
they
built
their
own
and
maybe
that
will
change
our
future.
But
today
it's
not
an
ing
container
I
was
thinking.
B
Second
plan:
if
you
build
a
jet
fighter
fighter,
you
start
out
with
requirements
and
that's
what
we
did.
We
builds
or
contain
a
security
framework.
Okay,
10
security
standards
and
basically
that
boils
out,
goes
down
to
two
basic
models,
so
the
Top
Model,
what
it
depicts
is.
It
takes
a
holistic
approach.
You
cannot
do
this
just
from
container
hosting.
You
need
to
help.
You
need
a
team.
Basically,
the
team
consists
out
of
the
container
hosting
platform
your
pipeline
platform
and
your
consumers.
They
all
need
to
work
together.
B
B
B
B
B
A
I'll
hand
over
to
Sandeep
yeah,
so
so
they
have
been
some
learnings
for
us
as
well,
but
like,
like
you
say,
the
environment
keeps
changing
and
therefore
we
need
to
keep
an
uptake.
And
if
you
look
at
that
has
been
a
lesson
that
you
can
see.
If
you
have
seen
the
movie,
then
you
see
that
that's
the
lesson
that
that
is
there
as
well,
and
it's
it's
very
much
explicit
for
us
in
our
real
lives.
A
So
we
no
longer
have
a
room
for
flying
the
planes
at
5000
feet
and
have
time
for
manure.
We
really
need
to
know
how
to
do
it
at
a
mission
stage,
which
is
typically
at
3
300
feet
right.
So
there's
very
less
time
for
error
and
if
you're
running,
Mr
critical,
you
know
infrastructure,
then
the
only
way
to
address
it
is
to
automate.
So
that's
what
we
did
basically
looking
at
automation
all
the
way.
A
So
there
have
been
some
learnings
as
well
in
this
in
this
journey
that
we
had
for
last
three
years,
I
mean
more
or
less
like
ties
explained.
We
have
been
doing
well
and
we
have
seen
a
lot
of
adoption
and
we
did
see
some
challenges
coming
out
so
surprises
coming
our
way
when
we
saw
a
huge
Among,
Us
capacity
growth
happening,
and
it
was
good
that
our
Engineers
were
up
for
up
to
it
and
they
could,
you
know,
address
these
challenges
pretty
quickly.
What.
A
A
Some
more
a
bit
of
sometimes
you
get
surprises
and
you
get
what
I
mentioned
is
Roadblocks,
and
this
is
one
of
those.
So
what
we
had
is
what
we
call
as
a
dark
star
moment
for
ourselves.
The
the
platform
itself
was
really
accelerating
at
Mach,
5
and
Mark
10.,
and
before
it
could
hit
it,
we
saw
a
lot
of
red
lights
going
up
what
that
meant
was
we
had
to
take
action
and
luckily
our
team
was
up
to
it
and
they
they
managed
to
get
the
situation
in
control.
A
A
So
there
is
a
learning
here
as
well.
So
if
you
look
at
so
there's
the
learning
here
as
well,
so
so
there's
one
do
a
stress
test
for
your
system
make
sure
that
you're
monitoring
the
platform
limits,
because
that's
that
can
really
help
you
to
find
these
limits
much
earlier
and
you
don't
run
into
issues.
A
Okay,
some
more
surprises
coming
our
way.
What
we've
found
is
that
typically
like
in
the
movie
we
were
talking
in
that
they
are
in
real
life.
Also,
you
always
have
these
characters
who
say
who
think
then
they
know
everything
and,
and
they
they
understand
everything
and
then
till
then
something
goes
wrong
and
they
and
all
the
you
know,
Bells
start
to
ring
and
that's
what
we
also
had.
What
we
found
is
that
our
initial
consuming
devops
teams
we're
very
good
at
adapting
the
best
practices.
A
They
were
good
at
collaboration
with
their
platform
team,
but
as
as
we
saw
more
growth
coming
our
way,
our
Focus
was
more
turning
towards.
How
do
we
do
more
self-service?
How
do
we
provide
better
features
and
capabilities?
Of
course
we
did
keep
updating
our
documentation
and
thinking
that
consumers
were
still
looking
at
the
best
practices.
They
were
surprisingly
in
store
for
us
and
we
found
that
a
lot
of
them
were
not
using
the
best
deployment
patterns
and
and
we
suffered
on
the
resiliency
front.
A
Some
of
it
went
to
the
Press,
but
that's
that's
a
different
story,
so
it
did.
Force
us
to
you
know,
introduce
some
mandatory.
You
know
controls
on
the
system
and
some
of
them
were,
you
know,
Canary
releases.
You
know
pod
resiliency
deployment,
defining
the
correct
liveness
probes,
so
these
were
kind
of
made
mandatory
for
the
for
the
teams.
We
also
introduced
a
lot
of
documentation
around
resiliency
how
to
do
proper
resiliency.
A
B
Thank
you.
So
this
is
about
Partnerships
and
like
in
the
movie.
You
need
an
ecosystem.
You
can't
complete
a
mission
with
a
couple
of
jets.
You
need
an
aircraft
carrier,
you
need
an
ec2
archive,
providing
OverWatch,
you
need
a
satellite,
providing
Communications
and
so
on,
and
so
on.
The
same
goes
for
cloud
native
Journey.
You
cannot
do
it
alone.
You
need
partners
and
that's
what
we
did
as
ing
so
starting
in
the
right
bottom
corner.
We've
taken
hosted,
notes
bare
metal
machines,
sometimes
virtual
machines,
former
colleagues,
in
ing
on
those
notes.
B
We
deploy
software
from
our
partners,
redhead
and
boardworks.
They
gave
us
the
software
and
services
we
need
to
base
basically
build
the
essential
platform.
That's
what
we
call
our
engineering
phase,
the
platform
on
its
own
yeah.
It's
not
an
operating
platform.
In
order
to
run
that
platform
of
operate
it.
We
need
a
lot
more
services
and
we
take
those
internally
from
our
private
Cloud
Partners
monitoring,
observability
Etc
focused
on
the
platform
itself.
A
A
I
am
related
to
link
templating,
so
we
use
them
as
our
partners
to
make
the
namespace
consumable
that
can
generate
value
and
if
you,
if
you
notice
on
the
right
corner
top
Corner
you'll,
see
how
consuming
teams
the
consuming
Partners
as
well,
because
they
are
the
one
who
actually
generate
the
business
value
that
we
intend
to
have
right
so,
but
there
are
also
our
partners
and
they
provide
a
pretty
robust
feedback
to
us
so
that
we
can
continuously
keep
improving.
That's
that's
how
our
platform
seems
to
be
evolving
itself.
That's
our
way
overall.
A
B
A
Have
a
strategy
around
how
the
platform
would
evolve
and
how
does
it
meet
our
organization
goals
so
and
ing
in
that
sense
has
a
strategy
about
building
platforms
that
can
sustain
that
can
scale
providing
growth
and
also
at
the
marginal
cost,
so
ing
so
ichp
actually
fitted
in
that?
What
it
did
is
that
it
allowed
us
to
offer
self-service
capabilities
with
tooling
around
it
and
services
around
and
as
well
provide
managed
management,
Knowledge
Management
around
and
had
has
good
support
channels
to
support
our
consumers.
A
The
platform
team
itself
is
is
basically
operating
as
small
stream
aligned
teams,
and
why
do
you
speak
of
it?
Is
it's
because
we
see
this
getting
validated
even
in
Concepts,
like
team
topologies,
where
they
also
talk
about
similar
things,
and
how
do
we
really
optimize
flow
and
that's
what
we
have
been
operating
since
the
last
three
years,
where
the
platform
has
really
evolved
itself.
B
So
poacho,
what's
next
today
we
are
working
as
a
mutable
platform
and
that's
our
contribution
to
showing
the
yachty
industry.
It
can
be
done.
It
is
really
hard
to
penetrate.
We
have
some
pen
tests
and
the
results
are
excellent,
but
we're
not
done
there's
still
additional
work
to
be
done.
For
instance,
we
can
look
at
a
multi-cluster.
We
can
improve
from
that
plus
as
a
scatter.
We
can
improve
on
that.
We
can
improve
even
on
expanding
the
scope
and
even
enabling
more
use
cases
to
be
hosted
in
the
most
secure
manner.
B
B
Make
professional
connections
prepare
for
your
own
story,
and
I
would
like
to
extend
an
invite
to
all
of
you
to
meet
ing
in
our
hometown
of
Amsterdam
for
qcom
Europe
2023
next
spring
means
to
travel
will
welcome
you
in
person
for
everybody
who
will
join
online.
We'll
gladly
welcome
you
online,
and
that
brings
us
to
the
end
of
this
presentation
and
we're
open
for
questions.