►
From YouTube: What's Next for Databases & OpenShift Virtualization Peter Lauterbach Red Hat OpenShift Commons 2022
Description
What's Next for Databases & OpenShift Virtualization
Peter Lauterbach (Red Hat)
OpenShift Commons Gathering on Databases held on 02/23/2022
Slides: https://bit.ly/3M2nRvf
Join OpenShift Commons: https://commons.openshift.org/index.html#join
Full Agenda here:
https://commons.openshift.org/gatherings/OpenShift_Commons_Gathering_on_Databases.html
A
Actually,
if
you
know
openshift,
there's
probably
about
45
of
my
closest
colleagues
that
manage
different
parts
of
openshift,
I'm
responsible
for
a
couple
of
different
things
at
red
hat,
one
is
rev
or
the
red
hat
virtualization,
which
is
our
traditional
virtualization
platform
openshift
on
rev,
which
is
the
ability
to
run
the
brand
new
spanking
container
platform
on
the
old
infrastructure
or
a
traditional
infrastructure,
and
then
openshift
virtualization,
which
is
the
ability
to
run
vms
inside
of
openshift,
and
I'm
sure
everybody's
wondering
really.
What's
the
what's
the
point
of
running
vms
inside
of
openshift.
A
A
It's
an
operator
that
you
download
load
it
up
into
your
openshift
cluster,
so
you
need
bare
metal,
though
I
will
talk
about
some
cool
cloud
stuff
coming
and
what
you
want
to
be
able
to
do
is
run
virtual
machines
and
containers
together
right
in
what
would
you
would
call
a
composite
application,
so
a
couple
of
things
that
are
available
in
openshift
virtualization
in
some
of
the
futures
that
we're
delivering
now
are
being
able
to
back
up
and
restore
your
virtual
machines
in
that
environment
as
you
back
up
your
openshift
cluster,
the
other
thing
is
that
openshift
itself
has
many
different
deployment
options
right,
so
you
can
actually
deploy
in
a
large
bare
metal
cluster.
A
The
other
thing
that
we're
now
getting
into-
and
this
is
kind
of
a
new
future
for
us
and
a
very
interesting
one,
which
is
running
virtual
machines
in
a
kubernetes
cluster
in
a
public
cloud
right.
So
we
already
have
openshift.
You
know
vms
and
openshift
in
aws
bare
metal
instances,
so
you
can
actually
have
a
and
an
openshift
cluster
in
aws
without
any
bare
metal
at
all
and
add
bare
metal
worker
nodes
that
you
can
actually
then
deploy
virtual
machine
workloads
on
we've.
Just
with
4.10
we
will.
A
We
will
be
also
previewing
tech,
previewing,
ibm,
bare
metal,
bare
metal
in
the
public
cloud
there
and
you'll
actually
see
as
other
customers
and
and
partners
demand
other
footprints.
We
have
other
cloud
providers.
Some
are
the
well-known
global
ones
like
azure,
who
were
in
conversations
with
and
then
much
smaller
ones.
That
tend
to
be
regional.
A
Pardon
me.
So
the
idea
of
running
virtual
machines
in
kubernetes,
as
I
said,
is
maybe
not
a
a
common
one
or
a
well-known
one,
but
the
idea
is
actually
helps
you
modernize
your
applications,
but
at
the
same
time
you
want
to
make
sure
that,
yes,
I
get
all
the
benefits
of
the
virtual
of
the
container
platform,
but
I
still
have
as
a
virtual
infrastructure
admin
right,
an
idea
like
hey.
I
need
to
create
vms.
I
need
to
be
able
to
manage
them.
I
need
to
be
able
to
update
them.
A
The
other
thing
that
we
can
also
do
with
these
virtual
machines
is,
as
you
know,
openshift
has
a
great
story
for
aiml
right
just
by
itself
with
containers
and
that
lends
itself
to
a
lot
of
interesting
use
cases
with
openshift
data
hub
and
a
lot
of
other
cool
things,
but
you
can.
A
So
over
on
the
right
here,
we've
got
a
quote:
I'm
not
going
to
read
it
to
you,
but
it
really
is,
and
I'm
going
to
show
an
example
in
a
second
here,
our
friends
over
at
sahih
benden,
who
you
heard
from
really
have
embraced
sort
of
virtual
machines
and
openshift
as
part
of
their
modernization
strategy,
and
it
really
has
a
couple
of
benefits
here.
A
A
Advantage
down
at
the
bottom,
you
I'll
not
go
into
the
different
use
cases,
but
we've
got
folks
looking
at
using.
You
know
virtual
machines
and
openshift
in
every
one
of
these
environments
and
what
would
that
look
like
right?
So
I've
got
your
traditional
three-tier
app.
So
let
me
orient
you
on
this
slide
time
flows
down
the
page
here
right.
So
in
the
first
row
I've
got
a
traditional
three-tier,
app
that's
on
a
legacy
application.
A
Maybe
it's
your
favorite
hypervisor
like
rev,
maybe
zen
server
or
say
vsphere.
I
can
actually
bring
those
virtual
machines
right
and-
and
this
is
your
standard-
hey
database-
you
know
in
linux-
it
would
be
mysql,
jboss
and
apache
right,
and
I
can
bring
that
over
in
windows.
A
It
could
be,
you
know,
sql
server,
dot
net,
you
know
framework
and
then
iis
is
the
front
end
and
you
can
bring
those
vms
directly
into
openshift
and
run
them
as
vms,
they're,
kvm,
vms
they're,
the
same
technology
that
you
would
use
on
other
red
hat
platforms.
A
Now
that
you
have
them
there,
it's
actually
pretty
cool
because
you
have
the
same
in-guest
properties
right.
The
operating
system
that
you're
using
are
the
same,
so
you
manage
them
internally
the
same
way,
but
now
you
have
all
the
benefits
of
running
on
the
container
platform
right.
So
the
idea
of
infrastructure
as
code
and
being
able
to
roll
out
new
parts
of
your
application
and
update
that
using
thing
like
get
ups
is
a
very
powerful
message,
even
without
making
any
code
changes
at
all.
A
So
now,
as
time
flows
down
the
page,
you
can
go
ahead
and,
let's
take
the
easy
part,
hey
the
web
front
end
we've
got
some
new
cool
cloud
stuff.
We
want
to
put
on
the
front
of
this,
maybe
a
mobile
app.
You
can
actually
containerize
that
and
still
leave
your
middleware
in
your
database
in
a
virtual
machine
now.
The
third
slide
here
is
actually
the
most
interesting
right,
which
is
there's
probably
a
couple
of
people
listening
to
the
sound
of
my
voice.
A
Right
now
going
like
you
know
what
databases
and
containers,
I'm
not
convinced,
I'm
going
to
run
it
on
my
bare
metal,
I'm
going
to
run
into
my
virtual
machine,
and
I
like
it
that
way
and
that
works
best
for
me.
So
you
can
actually
stop
at
this
point
and
you
know
essentially
containerize
and
modernize
everything
the
front
end
and
the
middleware
and
leave
the
database
in
a
virtual
machine.
A
If
that
makes
sense
for
you
right.
So
this
isn't
really
any
sort
of
forcing
of
a
technology
choice.
You
can
move
down
this
down
this
path
at
your
own
pace
right
and
once
you
get
comfortable
and
the
dba
decides.
You
know
what
I've
seen
enough
like
sql
server
runs
great
inside
a
container
right
and
we've
got
lots
of
people
doing
it.
A
You
can
actually
migrate.
Your
sql
server
database
from
your
virtual
machine
into
a
container
and
migrate
that
over
and
the
very
cool
thing
is,
if
you
set
it
up
right
and
you're
using
things
like
services
on
the
you
know,
in
the
cont
on
the
container
side
that
actually
that
transition
is
actually
seamless
to
the
application.
There's
no,
if
you're,
using
something
like
a
service
mesh
to
connect
everything
together
and
services
inside
the
cluster
that
will
be
completely
transparent
to
the
application.
A
A
They
had
a
lot
of
automation
that
deployed
into
their
old
infrastructure
and
what
they
were
able
to
do
is
actually
easily
you
know
as
they
built
this
open
shift
bare
metal
cluster.
Let
me
orient
you
here
to
this
picture,
so
every
server
here
is.
This
is
all
a
single
openshift
cluster
running
on
bare
metal
and
it's
a
combination
of
two
different
vendors
hardware
right
and
as
long
as
it's
certified
for
relay
you're
good
to
go.
A
A
One
other
thing
I
want
to
point
out,
which
is
actually
very
cool,
is
this
little
green
box
over
here
is
actually
a
large
legacy
storage
array
and
what
this
means
is
they
were,
I
don't
want
to
say,
skeptical,
but
they
wanted
to
sort
of
hedge
their
risk
of
like
well.
What
if
this
whole
bare
metal
cloud
thing?
Does
you
know
on-prem
cloud
thing
doesn't
work
for
us.
A
We
want
to
have
a
backup
plan
and
it
turns
out
that
I'm
not
going
to
name
the
storage
vendor,
but
like
there's
no
parts
of
the
application
that
are
stored
on
that
storage
and
when
they
actually
deploy
the
next.
You
know,
there's
actually
going
to
be
an
h,
a
site
for
this
when
they
deploy
that
that
part's
not
going
to
be
there.
A
So
it's
a
fairly
large
number
of
vms
there's
a
fair
number
of
management
challenges
here.
To
be
quite
honest
with
you,
because
you're
doing
both
things
right,
you're
moving
to
a
new
data
center,
but
you're
also,
you
know,
some
of
the
things
that
you
do
need
to
learn
is
you're,
actually
managing
a
kubernetes
cluster
right.
So
this
isn't
your
traditional
rev
or
vsphere
or
even
zen
server,
but
the
very
cool
thing
that
they
did
was.
A
The
team
was
very
focused
on
automation
and
they
were
quite
good
at
that
and
what
they
were
able
to
do
is
point.
You
know
the
build
pipeline
that
they
had
for
you
know
deploying
on
their
old
infrastructure.
A
And
when
that
went
excuse
me
when
that
those
vms
got
put
in
there,
then
they
actually
set
up
database
and
application
replication
to
pull
the
data
off
of
the
old
servers
into
the
new
ones.
Right
and
now
now
I've
got
a
form
of
disaster
recovery
here
and
then
I
can
actually
unplug
that
old
data
center
and
de
decommission
it
and
what
they're
going
to
use
that
same
exact,
that
same
exact
methodology
for
building
a
new
data
center
in
another
city.
A
A
A
It's
probably
it's
over
100
physical
servers
and
what
they
wanted
to
do
is
say,
look
we're
going
to
build
this
new
container-based
application,
but
parts
of
it
are
not
ready
for
containers,
and
the
two
pieces
specifically
were
a
sql
cluster,
a
carrier
grade
which
is
a
version
of
my
sql,
that's
very
high
performant
and
highly
available
and
the
load
balancer
that
they
had
chosen
right.
Containers,
weren't
ready.
A
So
at
that
point
you
either
don't
start
the
project
or
they
have
the
option
to
run
those
particular
pieces
of
the
application
as
vms
in
the
same
openshift
cluster
that
all
the
new
cloud
native
stuff
was
built,
and
the
very
cool
thing
is
is
when
I
checked
in
on
this
lately.
Somebody
said,
oh
by
the
way
they're
not
using
your
product
anymore.
A
By
the
way
this
went
in
probably
maybe
over
a
year
ago.
They
said
oh
they're,
not
running
that
stuff
in
vms
anymore,
and
I
was
like
that
is
super
exciting,
because
it's
that
wasn't
the
point
to
run
vms
long
term
over
there.
It's
to
let
them
get
their
architecture
set
up
and
prove
that
the
architecture
works,
and
they
can
do
the
scale
that
they
want
to
and
build
the
application
as
they
wish.
A
So
this
is
actually
a
hey.
I
can
actually
do
stuff
much
quicker
instead
of
waiting
for
a
perfect
world.
When
I
have
all
of
my
stuff
in
containers,
you
know
that
might
be
years
down
the
road.
So
this
really
lets
you
accelerate
and
take
the
benefit
of
the
things
that
you
already
know
how
to
do
like
run
a
database
in
a
virtual
machine
and
get
performance
out
of
it.
A
So
the
other
interesting
part
I
want
to
point
out
here
is
that
part
of
our
release,
criteria
for
openshift,
virtualization
or
vm's
and
openshift
is
a
not
only
a
regression
baseline
against
ourselves,
but
against
our
other
products
right.
So
this
is
the
way
to
read.
This
is
this:
is
a
relative,
a
relative
performance
chart
of
database
workloads,
sql
server,
postgres
and
mariadb,
and
obviously
they're
they're
not
running
them?
A
At
the
same
time,
this
is
a
set
of
benchmarks
that
we
run
internally
and
what
you
want
is
sort
of
parity
here
where,
like,
if
everything's,
equivalent
it'll
be
at
zero
and
in
this
particular
case
running
the
database
in
a
vm
on
openshift
was
a
couple
of
percentage
points
faster
and
in
the
case
of
sql
server,
it
was
like
one
and
a
half
percent
slower.
A
Now
I
don't
want
you
to
draw
any
conclusions
that
oh
my
gosh,
it's
better
to
run
it
in
containers.
That's
actually
within
the
except,
especially
for
a
big
database
benchmark
like
this,
you
know
five
or
six
percent
is
within
the
margin
of
error
and,
like
I
said,
this
is
a
release
criteria.
If
we
aren't
within
you
know,
performance
parity
of
what
other
customer,
what
customers
run
on
other
platforms,
the
product
doesn't
ship.
A
So
in
conclusion
here
you
can
do
a
lot
of
very
cool
things.
Openshift
4.10
is
going
to
drop
in
a
couple
of
weeks
and
has
all
this
new
cool
stuff.