►
Description
Learn how to utilize the Kubernetes Container Orchestration software in conjunction with an open source stateful application deployment solution called KubeDirector. In addition you will see ways to manage multi-cloud and hybrid-cloud based deployments of Kubernetes clusters using these tools.
A
Hi
welcome
to
our
demo
I'm
don
wake,
I'm
a
technical
marketing
engineer
with
hewlett
packard
enterprise
and
today
we're
going
to
talk
about
kubernetes
and
how
you
can
utilize
that
and
also
manage
a
multi-cloud
environment
with
me.
Today
is
terry
chang,
a
data
scientist,
hello
and
also
sandeep
deshmoop
platform,
product
manager
he's
going
to
talk
to
us
about
kubernetes
and
our
coop
director
application,
hello,
all
right.
So,
let's
jump
right
in
here.
Let's
set
the
stage
you
know
what
are
we
talking
about
with
regard
to
containers
and
kubernetes
with
containers?
A
It's
really
the
key
to
the
hybrid
cloud
and
we're
going
to
be
talking
about
that
concept
throughout
this
presentation.
A
That
container
infrastructure
is
really
fueling
the
software
development
of
the
next
decade
and
put
that
into
the
context
of
on-premises
and
also
public
cloud.
You
can
see
from
the
90s.
You
know:
we've
all
experienced
the
clout,
the
client
server,
everything
over
the
network,
lots
of
high
bandwidth
communications.
A
We
then
virtualize
servers
to
take
take
more
advantage
of
the
hardware
we
were
using
the
cloud
started
to
emerge
in
the
2010s,
so
you
had
a
lot
of
public
cloud
vendors
coming
out
and
making
it
easier
to
go
in
and
sort
of
rent
to
own
or
just
rent,
your
own
hardware
over
the
cloud.
A
But
then
we
started
to
see
just
a
growth
in
management
complexity,
so
virtual
machines
are
fairly
heavy
in
terms
of
a
management.
You
may
have
one
physical
server,
and
now
you
virtualize
it.
You
have
hundreds
of
servers
that
you
still
have
to
manage
as
if
they're
real
servers,
so
in
the
2015
to
now
time
frame
containers
are
a
way
of
a
much
lighter
weight
way
to
focus
on
the
application
itself.
A
As
a
virtual
entity
and
reuse,
a
lot
of
the
operating
system
pieces,
you
can
have
multi-tenant
multiple
different
applications,
containerized
in
the
same
physical
resource
or
virtual
resource.
You
can
see
that
has
grown
substantially
and
by
2021
1.8
billion.
Different
enterprises
will
be
using
containers
in
some
way
and
the
by
2022
just
right
around
the
corner.
75
of
enterprises
are
going
to
be
using
and
deploying
containers,
and
just
over
half
of
those
enterprises
will
still
be
using
on-prem
data
centers.
A
B
Yeah
sure
don,
so
this
is
a
very
high
level
overview
of
the
hpe
esmerald
container
platform.
As
you
can
see,
this
satisfies
a
lot
of
different
personas
different
team
members.
Your
it
can
actually
pull
together,
resources
from
on-prem,
bare
metal
or
even
public
cloud,
or
even
vms,
and
pull
them
into
a
single
pane
of
glass.
B
As
of
late,
we
were
just
certified
as
a
gold
member
of
the
cncf
certified
kubernetes,
so
we
have
that
which
pretty
much
shows
that
hpe
is
pretty
active
in
providing
open
source,
kubernetes
solutions
and
the
entire
platform
is
actually
sitting
on
top
of
the
hpe
esmo
data
fabric,
which
is
formerly
known
as
map
r.
A
Okay,
fantastic,
so
there's
really
a
lot
to
dig
into
about
how
you
can
utilize
kubernetes
and
in
this
open
source
environment.
So
I'm
going
to
go
ahead
and
jump
into
a
demo
environment
we
have-
and
in
fact
this
is
one
that
you
can
check
out
for
yourself,
I'm
going
to
show
you
a
link
to
a
demo
portal,
so
you
can
go
in
here
and
experience
the
kubernetes
interface
that
we've
created
in
our
container
platform
yourself.
A
So
you
also
have
access
to
a
video
that
you
can
play
to
get
a
little
more
information
about
that,
but
let's
dive
into
creating
a
kubernetes
cluster
in
a
multi-cloud
environment.
So
we
we
have
in
this
demo
the
possibility
to
jump
into
an
interactive
demo.
So
terry,
I
guess
this
would
now
be
the
container
platform.
So,
in
order
to
create
my
cluster,
I
guess
the
first
thing
I
got
to
do
is
log
in.
B
Yeah,
so
we're
going
to
log
into
the
platform
as
our
site
administrator
role
and
here
you'll
see
the
global
settings
and
today
we're
going
to
focus
mainly
on
the
kubernetes
section.
We
do
have
our
a
legacy-
epic
section,
but
here
we're
going
to
see
our
creation
of
a
kubernetes
cluster
within
the
container
platform.
So.
A
It
looks
like
just
to
interrupt
you
for
a
second
that
I've
seen
a
few
other
clusters
as
well.
So
it
looks
like
you
know:
it's
not
just
one
cluster
you're
managing
you
can
manage
multiple
clusters
within
this
in
this
environment.
B
Correct
so
you
can
create
your
own
cluster
here
and
you
could
specify
the
name
whatever
you
name.
You
want
to
give
it
to
your
cluster
and,
most
importantly,
you
can
specify
what
compute
resources
you
want
to
allocate
for
either
the
master
or
worker.
Here
we
only
have
vms,
but
you
can
add
cloud
host
bare
metal
host
and
you
can
mix
and
match
in
a
hybrid
deployment
of
your
kubernetes
cluster
and
once
you're
ready.
A
A
All
right,
so
that's
that's
fantastic,
so
we've
got
a
new
kubernetes
cluster,
we're
going
to
go
ahead
and
return
home,
and
then
I
so
now
tenants
so
so
sandeep,
I'm
gonna
bring
you
in
here.
I'm
gonna
go
into
the
interactive
demo
on
tenants.
So
what
is?
Can
you
explain
to
me
what
it
is
I'm
doing
here
as
a
tenant
within
that
kubernetes
cluster,
that
I
just
created.
B
C
C
You're
going
to
give
and
and
at
the
bottom
of
the
screen,
you'll
see
some
of
the
quotas
and
stuff
like
that,
and
that's
one
way
that
admins
would
restrict
a
tenant
from
taking
over
an
entire
cluster,
so
kubernetes
cluster
would
have
tens
or
hundreds
of
tenants,
but
we
don't
want
one
or
two
users
or
one
or
two
tenants
to
take
over
the
entire
resources
of
the
kubernetes
cluster.
So
this
is
one
way
of
allocating
the
right
set
of
quotas
based
on
priorities
and
the
work
the
users
have
to
do.
A
Okay,
so
we're
leveraging
really
a
kubernetes,
auto
scaling
approach.
I
guess
the
the
kubelet
would
then
enforce
those
those
limits
that
we
put
on
there,
then
okay,
so
that
so
we
created
a
tenant.
We've
we've
created
a
cluster
we've
divided
up
the
resources,
we've
set
some
quotas
and
then
we'll
go
ahead
and
talk
about
creating
and
deploying
kubernetes
apps.
So
we
now
have
this
potentially
multi-cloud
hybrid
cloud
kubernetes
cluster
with
a
tenant
we've
created
we're
going
to
log
back
in
and
now
going
to
that
tenant.
A
So
now
we've
say:
we've
hired
some
developers
or
a
new
team
and
they're
going
to
use
our
new
tenant,
and
now
we
need
to
give
them
applications.
So
this
is
where
another
product
that's
also
open,
source
comes
in.
So
why
don't
you
walk
me
through
this
sandeep?
What
am
I
doing
here
in
the
tenant
looks
like
I've
got
a
dashboard
of
all
of
the
the
multi-cloud
resources
I'm
using
for
this
particular
tenant,
and
then
I
can
connect
actually
here
to
the
terminal.
So
I'm
in
now
for
the
tenant.
C
Yeah
this
is
this
is
a
very
important
feature
from
the
container
platform,
which
is
called
the
web
terminal,
and
the
web
terminal
is
essentially
a
pod,
that's
running
in
kubernetes
and
for
a
user.
It's
very
convenient
because
it
sets
all
the
context
so
you're
already
within
your
namespace
and
all
of
the
kubernetes
context
is
set
up
ahead
of
time.
When
you
open
this
web
terminal-
and
now
you
can
here,
you
can
do
your
cube,
cuddle
commands
or
you
can
do
you
know,
helm
or
anything
else
to
develop
and
deploy
apps.
A
C
This
is
something
maybe
I
should
spend
just
a
minute
on
cube
director,
so
this
application
that
you're
trying
to
do
is
being
you
know,
deployed
as
a
cube
director,
app
and
cube
director
is
part
of
as
mineral
container
platform
is
essentially
a
custom
controller
of
kubernetes,
which
provides
an
easy
way
to
create
register
and
deploy
stateful
applications
as
a
web
service
and
when
authoring
these
apps
using
cube
director
any
tenant
user.
A
Okay,
so
that's
great
so
we're
using
really
standard
processes
within
within
our
environment.
Here
we
we've
defined
an
application
that
we're
going
to
launch
now
we're
in
our
cube
director
running
app
and
we're
going
to
launch
that
it
looks
like
I
now
have
a
yaml
file
to
further
configure
the
launch
of
this
app
that
we've
employed
imported
with
coop
director,
so
I
can
adjust.
The
memory
looks
like
set
limits,
so
I'm
saying
the
memory
I
can
reduce
to
five.
My
cpu,
I
can
say
well
this
particular
one
needs
four
and
cpu
limits.
A
Okay,
so
one
stop
shop,
really,
okay,
so
exit
full
screen
here
on
our
demo
and
so
coming
back
to
really
just
reiterate
what
we
just
talked
about.
So
the
container
platform
is
a
way
that
you
can
use
kubernetes
and
create
clusters,
and,
as
terry
showed,
when
we
were
allocating
the
resources,
we
were
choosing
virtual
machines.
We
could
choose
bare
metal
within
our
on-prem
data
center.
A
So
I
hope
that
this
information
was
interesting
to
you
and
helpful,
as
always,
there's
more
that
you
can
find
out
if
you
just
want
to
go
to
our
hp.com
esmo
platform,
the
hp.com
container
platform
and
learn
more
about
coop
director
download
it
and
use
it
yourself
in
your
own
kubernetes
environment.
It
gave
you
the
link
here
to
the
github
account
so
fully
open
source,
container
platform,
email
and
also
the.
A
All
on
the
web,
you
also
can
follow
us
on
twitter
at
hpe,
underscore
esmeral
or
hpei.
That
will
get
you
lots
of
great
feeds
of
information
that
relevant
to
kubernetes
and
container
management.
So,
thank
you
again
for
your
time
and
it
was
really
appreciated
and
hope.
You
learned
something
and
come
back
for
more.