►
From YouTube: OpenShift Coffee Break: Red Hat Hackfest Office Hours
Description
Get your espresso ready for the EMEA OpenShift Coffee Break as we kick-off the Red Hat Hackfest Office Hours! Join the monthly meeting with Red Hat and Red Hat Partners experts around Edge, RHEL and OpenShift as part of the Red Hat Hackfest.
Twitch: https://red.ht/twitch
A
My
name
is
andrea
battaglia
and
I'm
I'm
gonna
be
your
host
for
this
monthly
session,
and
today
today
with
me,
I
have
tania
ripo
who's,
the
producer
for
the
event
and
he's
one
of
these
team
members
of
the
heartfest
community.
A
We
are
really
looking
forward
to
hear
from
our
guests.
Today
we
have
very,
very
special
guests
that
will
will
be
able
to
give
us
several
several
overviews
on
the
business
and
the
technical
side
of
the
red
attack.
Fest,
so
get
your
espresso
ready
and
enjoy
the
show.
A
The
agenda
for
today
we'll
have
michael
telling
us
what's
next,
with
the
hack
fest
in
apac,
and
then
we
will
have
a
two
champions
red
champions
with
us.
Andreas.
A
Giving
us
an
overview
of
microchip
feel
free
to
put
your
questions
in
the
chat
anytime.
We
would
like
this
to
be
a
kind
of
a
a
live
show
where
people
interact
with
the
with
the
presenters.
We
are
more
than
happy
to
reply.
Your
questions
live
and
for
the
people
watching
the
episode
offline
feel
free
to
reach
out
to
us,
anytime
and,
and
we
will
be
happy
to
start
conversation
on
the
topic.
A
So,
let's
start
with
with
michael
michael
and
his
team
have
worked
hard
to
help.
The
hackers
team
scaling
the
red
hatfish
program
program
in
the
asia
and
pacific
region.
So
today
it's
my
privilege
to
introduce
him
michael
geisler,
head
of
partnering,
consistent
development
in
asia
and
pacific
and
michael
feel,
free
to
say
a
few
words
about
yourself
and
and
tell
us
what
the
hack
fest
means
for
for
the
abac
region.
C
Yeah
thanks
andre,
so
I've
been
at
red
hat
for
six
years
and
well
over,
like
two
decades,
probably
closer
to
three
decades
in
the
industry,
which
is
starting
to
fill
my
age
that
across
a
range
of
roles,
I
started
in
pre-sales
and
technical
roles
and
I've
been
in
sales
and
partner
ecosystem
roles,
probably
for
most
of
the
last
15
years.
C
So,
as
andrea
said,
we're
we're
kicking
off
the
first
red
hat
apac
hack
fest
at
the
end
of
july,
with
the
great
support
of
andre
aaron,
tanya
and
team
in
emea,
we're
taking
all
of
the
best
practices
and
experience
that
they've
they've
learned
over
running
hack
tests
over
several
times
in
emea
and
and
trying
to
replicate
that
success
in
apac.
C
The
great
news
is:
we've
also
got
some
fantastic
sponsors
in
the
intel
and
aws.
C
So
we're
looking
forward
to
working
with
them
and
pulling
together
solutions
with
those
key
partners
in
apac
we've
sort
of
significantly
increased
our
investment
in
our
partner
ecosystem
this
year.
So
we've
built
actually
built
out
a
brand
new
team
to
focus
on
partner,
enablement
and
drive
more
of
these
sort
of
proactive
activities
like
hackfest,
historically
with
really
been
a
programmatic
approach
to
enablement.
I'm
sure
many
of
you
are
familiar
with
our
partner
training
portal
and
largely
just
most
of
our
enablement
activities
have
been
focused
on
that.
C
C
We've
really
again
increased
our
focus
on
working
with
partners
to
develop
and
deliver
solutions
that
solve
problems
for
our
customers,
rather
than
just
focusing
on
our
red
hat
products
and
product
training.
We
really
want
to
get
solutions
that
solve
problems
for
our
customers
that
really
deliver
those
tangible
business
outcomes
and
the
other
area
of
focus
for
us
is
we've
doubled
down.
C
On
our
recruitment
of
new
isp
partners
in
the
region-
and
I
think
we've
got
about
150
new
edge
and
hybrid
isb
workloads
on
openshift
now
and
we're
hoping
that
with
hackfest
and
the
partners
in
hackfest
that
will
find
innovative
ways
to
incorporate
one
or
more
of
those
isvs
into
their
solutions,
as
hackfest
is
something
that
really
supports
one
of
our
goals
and
ambitions
to
build
more
collaboration
across
our
ecosystem.
C
C
We
all
know
customers
have
problems
to
solve
efficiencies
to
gain,
and
our
partners
understand
these
issues
better
than
we
do.
So.
I
really
see
hackfest
as
this
catalyst
to
foster
the
collaboration
to
build
the
solutions
that
deliver
these
outcomes
for
customers
and
drive
our
respective
business,
and
given
that
the
apac
region
has
sort
of
become
the
factory
of
the
world
and
in
apac
red
hat
is
increasing
our
investment
quite
significantly
into
the
manufacturing
sector.
We've
decided
to
use
the
same
manufacturing
use
case
that
you've
been
using
in
emea.
C
So
I
guess
in
summary,
what
what?
What
our
expectations
for
pack,
fest
in
apac
and
to
us
it's
a
really
key
enablement
activity
for
our
partners,
and
you
know
bringing
in
partners
like
intel
and
aws
and
some
of
our
isv
partners
and
and
driving
enablement
around
those
products
and
technologies.
C
Collaboration
across
our
ecosystem
is
a
fundamental
goal
within
the
wider
ecosystem
team
in
apac,
and
then
we
really
want
to
deliver
the
basis
of
solutions
that
are
going
to
solve
those
real
customer
problems,
not
only
in
manufacturing
but
a
solution
that
can
be
applied
into
other
verticals,
like
retail
oil
and
gas
and
mining
and
so
forth
so
andrea
I'll
hand
back
to
you.
A
Yeah
thanks,
michael
that's,
an
interesting
perspective,
so
asian
pacific,
as
you
taught
me,
is,
is
a
big
and
diverse
market,
and
I
have
a
memory
of
when
of
the
time
we
kicked
off
this.
This
collaboration
between
emea
and
apac
to
roll
out
the
hackfest,
also
in
your
region,
and
I
was
impressed
from
the
diversity
in
terms
also
of
language,
barriers
or
or
marketing
marketing
needs.
So
considering
this
diversity,
I
would
like
to
hear
from
you
what
do
you
think
we
can
do
together
for
partners
in
terms
of
enablement?
A
So
what
are
your
plans
or
your
expectation
and
also
what's
your
vision
from
the
customer
standpoint?
So
what's
in
the
hot
fest
in.
B
C
So
for
the
customers,
we're
really
seeing
the
the
opportunity
to
like,
I
said
earlier,
to
really
find
problems
that
we
can
solve
for
them,
using
our
ecosystem,
bringing
together
our
our
partners
together
to
deliver
that
outcome
or
solve
that
problem
for
a
customer
as
individual
partners.
You
know,
as
I
said,
openshift
doesn't
solve
a
problem
for
a
customer.
C
That's
it
forms
the
platform
and
the
basis
to
build
a
solution
for
a
customer
and
si
doesn't
deliver
a
solution
unless
they're
pulling
together
a
number
of
different
products
and
services
and
so
forth,
intel
doesn't
solve
a
problem.
Aws
doesn't
solve
a
problem,
it's
bringing
all
of
these
partners
together
and
delivering
an
outcome
for
a
customer
and
that's
what
we
really
want
to
focus
on.
We
want
to
move
away
from.
B
A
Believe
in
in
this
methodology,
so
having
a
less
product,
centric
and
more
customer
challenged,
focused
and
activity
now,
productiveness
is
crucial
in
this
and,
let
me
also
say
hi
to
maverick
who's
joining
from
bangalore,
so
we
have
a
broad
audience
today,
thanks
also
michael
for
sharing
and
advertising
on
our
social
media.
I
heard
you
mentioning
aws,
and
I
know
also
another
co-sponsor
is
intel.
A
This
synergy
is
fantastic
right,
so,
as
I
had
already
the
chance
to
share
in
the
during
previous
openshift
coffee
break
sessions,
hosted
by
andrea
natale
and
the
team,
our
vision
for
the
hack
from
the
harvest,
standpoint
of
a
complex
solution
is
always
the
synergy,
and
collaboration
and
commitment
of
several
entities
could
be
hyperscaler
in
your
case,
aws
could
be
hardware,
vendors
more
in
general,
like
intel
and
other
counterparts
and
the
crucial
point.
Of
course,
the
central
pillar
is
the
partner
ecosystem.
C
Yeah
so
let's,
let's
start
with
intel-
and
I
actually
forgot
to
mention-
intel
actually
awarded
us-
the
global
isv
partner
of
the
year
today
as
red
hat
and
apac,
so
we've
been
working
particularly
closely
with
them
for
many
years,
but
certainly
over
the
last
12
months,
we've
spent
a
lot
more
time
on
proactively
working
with
them,
so
intel
and
red
hat
collaborate
on
customer
pocs.
Today
we
intel
co-funds
a
number
of
different
pocs
in
the
telco
and
fsi
space
in
particular,
and
that's
been
a
very
successful
partnership
in
that
space.
C
This
year,
we've
really
put
a
lot
more
focus
on
working
with
intel
to
tap
into
their
ecosystem,
so
let's
go
and
find
partners
that
intel
have
that
want
to
partner
with
red
hat
as
well
or
may
not
be
aware
of
what
red
hat
has
to
offer
them.
So
at
the
moment
we
have
a
pipeline
of
about
90
partners
of
intel's
in
apac
that
we're
bringing
on
board
and
intel's.
Obviously,
I
think
in
in
the
case
of
hackfest,
has
a
long-term
interest
in
this.
C
They
want
to
see
this
grow
and
grow
and
grow
and
continue
over
the
coming
years
because
they
want
to
bring
these
part
their
partners
in
into
this
environment
as
well,
and
yet
we've
probably
picked
up
about
10
new
partners
of
intel's,
where
intel
has
actually
proactively
worked
with
us
to
bring
them
into
the
red
hat
ecosystem.
So
that's
intel
aws
we've
had
a
long-standing
relationship
with
we
have
rosa
as
our
sort
of
aws
native
version
of
openshift
for
those
who
aren't
familiar
with
that.
C
We've
been
working
very
closely
with
them
recently,
where
we've
been
actually
many
of
the
new
isvs
that
we're
recruiting,
subscribing
to
rosa
to
to
deliver
their
solutions
through
aws,
but
on
openshift.
So
that's
been
a
really
strong
collaboration
as
well.
So
we'd
find
multiple
angles
with
each
of
these
partners
intel
and
aws
to
work
together
and
collaborate,
so
hack,
fest
is
sort
of
a
natural.
I
guess
progression
of
bringing
those
partners
together
with
us
and
bringing
in
those
wider
partners
of
the
ecosystem
to
work
together.
A
That's
great
to
hear
thank
you
michael
and
one
last
question
for
you
which,
which
jumped
into
my
mind,
while
you
were
talking.
How
do
you
see
the
collaboration
we?
We
have
several
partners
right.
We
want
to
identify.
We
have
a
big
system,
integrators
local
system,
integrators
also
isvs,
which
are
completely
different
because
they
they
create
their
own
product,
starting
or
from
from
our
or
from
a
certain
technology
stuck
in
their
resell
their
product
right.
A
How
do
you
see
different
type
of
partners
collaborating
in
such
an
initiative
so
and
also,
what's
your
view
in
terms
of
a
framework
after
the
hackfest
that
puts
together
different
partners
to
build
and
solve
customer
challenges.
C
Yeah,
absolutely
so,
for
this
first
hack
first
we're
we
are
keeping
it
to
a
smaller
subset
of
our
partners,
primarily
so
in
apac
we
have
a
partner
program
called
openshift
practice
builder,
where
we've
brought
in
particularly
local
systems
integrators,
but
not
so
much
the
gsis
in
this
case,
but
the
local
systems
integrators
or
regional
systems
integrators
and
provided
them
incentives
and
other
program
benefits
and
so
forth
to
encourage
them
to
build
practices
around
openshift.
Hence
the
name.
C
So
that's
been
a
very
successful
program
that
we've
run
for
many
years
and
we
see
hackfest
as
a
means
to
bring
another
level
of
engagement
with
those
partners
and
we'll
focus
on
those
partners
for
for
this
version
of
heck
first,
but
where
the
isps
come
into
play
is
that
we've
started
seeing
and
and
brokering
relationships
between
those
isvs
and
our
strongest
openshift
partners,
because
those
isvs
who
are
building
their
solutions
on
openshift,
don't
necessarily
have
the
reach
or
capability
to
go
and
deliver
an
openshift-based
solution
in
another
country
or
another
region.
C
So
they,
those
isp
partners
which
might
be
based
in
australia,
for
example,
are
looking
for
our
strongest
open
shift
partners.
In
other
countries-
and
you
know
emir
eventually
as
well
so
the
partners
are
on
this
call.
We,
as
I
say,
we
have
many
isvs
who
are
looking
for
all
sorts
of
channels
to
market
and
particularly
looking
for
strong
partners
in
openshift
and
rosa
and
arrow
and
those
sorts
of
technologies.
So
that's
that's
where
we
see
the
really
strong
link
going
forward
as
well.
C
A
Well,
fantastic,
michael
thanks,
a
lot
thanks.
A
lot
really-
and
let
me
also
remind
and
and
stress
this
topic
with
our
audience.
The
goal
of
our
program,
initiative
and
and
now
support
supported
by
michael
and
supported
by
mia
and
hopefully
to
be
rolled
out
also
in
in
in
the
other
region
as
soon
as
possible
is,
is
to
put
partners
together.
So
we
don't
really
want
to
keep
boundaries
there.
Okay
technology
is
not
really
a
boundary.
A
There
are
tools
and
methodologies
to
enable
and
strengthen
skills
of
our
partner
ecosystem,
but
the
boundary
could
be
geographical
or
could
be
language
barrier
that
we
want
to
address
to
make
sure
that
we
create
a
strong,
diverse
and
global
pattern,
ecosystem
that
can
support
technology,
vendors
and
customers
all
together.
Now
the
the
main
goal
here
for
michael
myself
and
the
rest
of
the
hackfest
contributors
and
sponsors
is
really
to
have
a
strong
support
to
the
parthenon
ecosystem.
A
For
for
our
for
our
end,
customers
always
focusing
on
technology
and
technology
stuck
and
specifically,
emerging
technologies,
and
by
emerging
technologies
I
mean
a
microchip.
So
thanks
a
lot
michael
microchip
is
the
main
topic
of
of
this.
Second
part
of
the
hackfest
office
hour-
I
can
see
here
two
gentlemen,
frank
and
andreas,
so
just
to
quickly
introduce
the
topic
guys,
probably
most
of
our
live
and
offline
attendees
here
have
heard
all.
A
Is
a
project
sponsored
by
the
red
office
of
gto
today,
frank,
who
is
the
microsoft
microchip
engineering
team
lead
will
give
us
an
overview
of
microshift
and
now
it's
the
time
for
technical
people
to
start
asking
questions,
because
frank's
is
even
though
he's
not
paid
he's
here
to
to
answer
all
your
questions,
and
also
we
have
andreas
andreas
is
one
of
the
most
esteemed
member
of
the
hacques
technical
team,
and
he
is
also
a
champion
when
it's
about
impra,
rail
and
and
container
technology,
so
he's
our
specialist.
A
B
So,
hey
everyone.
First
of
all,
thanks
a
lot
for
having
me
on
this
stream
really
glad
to
be
here
and
talk
about
microshift.
My
name
is
frank:
starsky,
I'm
a
senior
principal
software
engineer
at
inves
office
of
the
cto,
and
I've
joined
red
hat
in
2014
at
the
time
when
the
telco
industry
was
embarking
on
this
journey,
to
move
away
from
bespoke
hardware
software
into
cloud
and
put
all
the
infrastructure
and
operations
and
processes
onto
the
cloud,
and
my
first
task
was
actually
to
evangelize.
B
If
you
want
open
source
or
upstream
first
open
source
in
those
communities
and
eventually
build
a
team
that
would
extend
first,
open,
spec
and
open
shift,
then
openshift
with
all
the
features
that
they
would
need
to
run.
Telco
5g
workloads.
We
were
working
closely
with
our
partners,
our
ecosystem,
because
you
know
there's
like
just
a
ton
of
experience.
B
We
can
gain
from
from
doing
hack,
first
from
working
with
partners,
etc,
and
also
also
our
product
engineering,
and
the
thing
we
did
is
we
always
worked
on
a
horizontal
platform,
because
it
turns
out
a
lot
of
the
features
that
you
need
in
telco.
You
also
need
in
other
industries
and
then
in
2018.
B
B
B
D
Well,
thanks
frank,
so
my
name
is
andre.
Stolzenberger
I've
been
with
redhead
for
almost
10
years.
Now,
I'm
a
principal
solution
architect
and
I
started
with
working
with
large-scale
customers
in
in
germany
and
moved
over
to
partners
and
partner
enablement,
and
I
joined
the
hackfest
team
because
I'm
the
only
guy
that
doesn't
code,
I'm
the
one
responsible
for
the
infrastructure
so
that
the
guys
that
do
code
have
something
to
execute
on
and
we
came
up
with.
D
The
andrea
came
up
with
the
idea
of
the
hackfest
architecture,
with
a
data
center
with
an
edge
and
we
needed
to
find
something
where
we
can
actually
execute
on
the
edge
outside.
D
And
when
we
started
that
we
built
something
with.
You
know
installing
an
edge
device
like
an
intel
fitlet
with
rel
and
then
taking
the
containerized
applications
that
we
have
and
use
them
with
part
men
until
we
stumble
across
frank's
microshift
project
that
we've
embraced
ever
since,
because
for
us,
that's
the
logic
extension
from
a
large
infrastructure
where
the
customers
have
the
data
center
running
openshift
clusters
to
an
edge
with
a
maybe
single
node
openshift
and
then
going
directly
on
the
device
on
the
iot
device,
and
they
want
to
have
their
applications.
D
Go
all
the
way
with
the
same
container
image
with
the
same
settings
around
it.
So
microshift,
as
the
miniature
version
of
openshift
to
extend
to
the
edge
device
is
the
absolute
logic
decision
to
go
and
that's
why
we
embrace
that
in
our
architecture
and
we're
going
to
use
it
from
now
on
forwards
to
actually
run
the
code
there
and-
and
that's
that's
my
part
in
the
team
to
get
that
up
and
running.
D
And
that's
why
I'm
talking
with
you
here
so
and
yeah
so
back
to
frank,
to
show
us
how
it
actually
works.
B
Excellent,
thank
you
address.
Let
me
share
slides
and
I
promise
you
I
will
keep
it
short
just
use
them
as
a
as
a
background.
Just
to
give
you
a
little
bit
of
context
on
where
we
want
to
go
with
with
microshift
and
the.
C
B
Is
because
it's
reason
for
existence
really,
because
there
is
a
world
of
use
cases,
that's
very
different
from
the
cloud
and
data
center
use
cases
that
openshift
has
been
built
for
so
of
course,
I'm
not
paid
at
all
to
say
that
openshift
is
is
awesome
if
you're
running
on
the
cloud
and
data
centers
and
then
ever
smaller
and
smaller
server
infrastructures
running
them
in
in
pretty
controlled
environments.
B
But
then
there
is
a
whole
different
set
of
edge
competing
use
cases
which
involve
what
I
call
field
deployed
edge
devices
and
in
in
this
world
deployed
edge
devices.
You
know
what
devices
are
if
you
want
single
board
computers,
systems
on
chip
system
or
modules
that,
unlike
servers,
have
extremely
limited
resources
and
are
not
typically
not
extensible.
B
So
if
you
look
at
the
left,
I'm
just
giving
you
some
examples
of
these
platforms
and
what's
interesting
in
that
edge
device
space
is
that
a
family
of
platforms
typically
has
a
preferred
use
case
or
the
other
way
around
right.
So
if,
for
example,
if
you
have
aiml
use
cases,
then
the
india
jetson
platform
is
super
popular
just
because
they
give
you
a
really
really
good
performance
per
dollar
or
a
performance
per
watt
of
energy
that
you
invest
likewise,
the
nxp
imax
platform
is
super
popular.
B
If
you
have
like
media
kiosks
like
applications,
science
ultra
scale
is
a
platform
that
people
who
are
into
signals,
processing,
fpgas,
etc,
prefer
and
then,
of
course,
all
of
us
makers.
You
know
they
like
here's,
the
raspberry
pi
4.
and,
apart
from
resource
limitations,
a
lot
of
things
change
when
you
deploy
or
use
these
devices,
for
example,
the
whole
procurement
provisioning
process
right.
B
So
it's
typical
to
have
those
devices
imaged
at
the
factory
shipped
directly
on
site,
where
a
non-technical
person,
just
you
know,
mounts
them
to
a
head
rail
or
to
the
wall
or
wherever
plugs
in
the
power
of
our
ethernet,
cable
and
walks
away
right.
So
that's
kind
of
the
kind
of
provisioning
you
have
there's.
No,
you
know
iso
based
installation
or
pixie
build,
or
none
of
that,
and
one
of
the
reasons
also
that
these
devices
do
not
have
any
out
of
band
management
right.
B
So
there
is
no
ipm-I,
no
bmc,
and
the
consequence
of
that
is
that
these
devices
are
not
remotely
recoverable
over
the
network.
In
case
you
push
a
bad
software
or
configuration
update,
and
the
other
part
is
that
these
devices
are
field
deployed.
So
the
deployment
and
maintenance
of
these
devices
is
super
expensive.
You
need
to
do
truck
rolls
in
some
case.
In
case
of
some
of
our
customers.
Helicopter
runs
there's
low
physical
access
control.
B
There
are
severe
networking
constraints
so
think
about
devices
that
are
only
connected,
maybe
once
a
week
connected
over
a
network,
that's
like
on
the
order
of
megabits
per
second
or
or
super
expensive,
because
you
pay
by
the
megabyte
so
a
lot
of
challenges
related
to
to
this
network
environment,
and
when
you
look
at
what
reddit
then
did
is
to
say,
hey,
we
need
an
operating
system
that
is
tuned,
especially
for
this
site
set
of
use
cases.
B
Of
course,
it's
not
going
to
be
a
new
operating
system,
it's
basically
a
deployment
configuration
of
our
red
enterprise,
linux,
right
and
so
red
has
been
working
over
the
last
year,
plus
to
add
features
to
red
enterprise
leadings.
Exactly
for
that
use
case
and
what
those
features
are
is
basically
the
ability
to
create
rpm,
os
3
based
system
images.
Why
do
you
do
system
images
because,
like
on
an
on
your
cell
phone
right,
you
want
to
have
the
whole
operating
system.
B
Its
configuration
and
the
default
set
of
applications
may
be
into
a
single
image
that
you
can
develop
as
a
system
image
you
test
as
you
create
it
as
a
as
an
image
and
you
deploy
it
as
an
image
out
to
the
devices
in
the
field
and
thereby
you
ensure
that
all
devices
look
almost
exactly
the
same,
and
that
helps
you
really
really
with
on
the
operations
side,
because
you
know
the
less
variation
you
have
between
devices
the
easier
to
troubleshoot
in
case
there
are
problems,
and
then
there
is
so
there's
a
tool
called
image
builder
that
takes
a
blueprint,
so
a
text
description
of
what
is
the
operating
system?
B
What
do
you
want
to
embed
into
that
system?
Image,
builds
it
and
allows
you
to
publish
this
os3
repo
os3
repos
are
os.
Trees
are
interesting
because
you
update
and
roll
back
transactionally
the
operating
system
right.
That
means
there's
no
more.
You
know
dnf
update
on
each
individual
device,
you're,
basically
updating
the
golden
file
system.
B
If
you
want
and
then
there
are
very
efficient
mechanisms
to
delta
download
from
whatever
version
the
devices
are
on
to
the
target
version
in
this
rpm
os
3
repo,
and
because
these
file
systems
are
actually
a
b
installation,
so
they're
both
present
on
the
system.
At
the
same
time,
the
currently
running
one
and
the
one
for
the
update,
there's
the
possibility
to
basically
install
or
stage
the
new
file
system,
and
only
when
it's
complete
you
can,
when
it's
complete,
you
can
completely
lose
network
connectivity.
B
If
you
want
right,
the
update
is
actually
no
longer
requiring
a
network,
so
you
basically
just
reboot
and
10
seconds
later,
and
it's
really
going
quickly
like
that
you're
in
the
new
file
system.
B
And
then
you
have
the
ability
to
add
health
checks
and
say:
is
the
file
system
coming
up
correctly
and
if
not
automatically
you,
the
system
rolls
back
to
its
previously
known
good
state
and
the
reason
I'm
giving
you
this
context
is
because
when
we
talk
about
microshift,
it
is
actually
an
open
shift
exactly
for
this
device
edge,
and
so
we
are
leveraging
all
those
features
that
welfare
edge,
gives
customers
and
then
built
on
top
of
it.
So
what
what?
What
metric
is?
B
Basically
an
essential
subset
of
urban
shift,
kubernetes
in
a
small
form
factor
and
optimize
for
these
field
deployed
edge
devices.
A
lot
of
our
customers
actually
want
to
embed
microshift
as
a
kind
of
tiny
cubist
runtime
into
their
products.
So
we
also,
of
course,
want
to
make
it
easy
to
embed
microsoft
itself
as
well
as
kubernetes
workloads
and,
as
I
said,
because
we
are
inheriting
refresh
features.
B
The
whole
installation
process
of
microshift
will
eventually
just
be
you
know,
procure
a
device
or
fight
a
device
on
board
plug
it
into
the
network,
and
the
final
device
onboard
in
rel
will
ensure
that
these
devices
automatically
discover
their
owner
and
their
owner's
management
system
and
onboard
into
that
management
system,
and
they
will
then
already
have
if
a
customer
chooses
to
embed
microshift.
This
might
run
runtime
embedded
into
the
system
image
right
and
then
you
can
deploy
kubernetes
applications
in
one
of
two
ways.
B
You
can
use
this
combination
of
rel
for
edge
and
microshift
as
a
classical
say
in
in
a
in
a
host
hosted
model.
So
you
can,
just
you
know,
install
let's
say
a
management
agent
on
that
device
that
automatically
pulls
from
git
or
via
via
acm
and
the
advanced
customer
management.
B
The
target
workloads
that
you
want
to
run
dynamically
right,
add
and
remove,
but
we
also
have
customers
that
are
super
interested
in
actually
embedding
kubernetes
applications
into
the
system
image.
So,
if
you
think
about,
I
don't
know
and
your
your
phone
again,
there
are
some
applications
that
come
pre-installed
with
your
operating
system
right.
B
There
are
some
that
you
download
at
runtime
through
the
app
store,
and
we
want
to
support
both
of
these
models
actually
and
the
nice
thing
about
doing
that
with
rpm
os
trees
is
that
you
actually
have
the
possibility
to
download
updates
of
the
rpm
tree
over
the
network,
or
you
can
do
a
sneaker
network
right.
You
can
basically
take
a
usb
stick
with
the
new
rpm
os
3
version,
plug
it
into
the
device.
The
usb
stick.
B
I
mean
the
device
is
protected
using
usb
guard,
so
you're
basically
validating
that
this
usb
stick
is
in
fact
signed
by
the
device
owner
and
the
content
is
signed,
and
only
then
you
basically
copy
the
new
version
onto
the
device
and
then
automatically
update
into
the
new
version,
and
so
we
like
to
keep
this
just
as
two
different
ways
of
doing
transport.
B
B
So
there's
a
lot
of
stuff
we
cut
out,
that's
mainly
using
supporting
the
you
know
the
built
host
or
build
cluster
use
case
that
we
don't
need
at
the
edge,
and
the
other
thing
we're
changing
is
that
with
openshift
you
know,
openshift
takes
care
of
the
operating
system,
the
infrastructure
and
all
of
those
things
for
you
right
it
abstracts
from
it.
It
controls
it
like
top
down.
If
you
want,
and
because
of
the
use
case
of
microsoft
being
embedded
into
somebody
else's
system,
we
actually
do
the
opposite.
B
So
much.
If
it's
just
there
for
running
workloads,
it
is
not
meant
to
control
the
operating
system,
its
configuration,
etc.
It's
the
job
of
the
user.
You
know
to
tune
rel
for
their
specific
use
case,
and
then
microchip
is
just
there
to
if
you
want
to
expose
those
resources
from
the
operating
system
into
the
cluster.
But
once
the
cluster
is
up,
my
cluster
is
up.
B
D
Sure-
and
I
could
illustrate
what
he
was
just
presenting
on
slides
and
show
you
how
you
simply
really
is
to
build
the
image
and
get
it
running.
A
And
that's
fantastic
thanks
thanks
andreas
and
frank
that
your
presentation
was
interesting
and
quite
insightful
to
be
honest,
so
that
actually
made
made
a
question
jumped
into
my
mind
as
you
are.
As
you
are,
the
the
the
team
leader
of
the
engineering
team
from
microsoft.
A
B
Yes,
absolutely
so
so
far
or
to
date,
microshift
is
a
community
project.
It's
a
project,
not
a
product
right
out
of
my
team
and
the
the
it
was
basically
used
to
test
the
warts.
If
you
want
with
customers
and
see
where
do
we
hit
the
nail
on
some
of
their
use
cases,
what
do
customers
actually
want,
and
we
have
had
collected
this
feedback
to
now
better
understand
what
the
priorities
of
our
customers
are.
B
Recently,
there's
been
this
decision,
we
are
going
to
practice
microshift.
So,
by
the
end
of
this
year
there
is
going
to
be
first,
you
know
tech
preview,
one
of
microshift
basically
released
at
or
around
the
openshift
4.12
release,
and
still,
if
you
want
limited
availability
in
the
sense
that
we
need
to
basically
want
to
limit
the
production
rate,
support
to
a
smaller
set
of
customers.
B
First
right,
but
always
microsoft
will
be
available
in
downloadable
loadable
for
everyone,
of
course,
and
then,
hopefully,
with
4
13,
we
will
be
able
to
move
to
general
availability.
What
we're
doing
right
now
is
on
again
focusing
on
integrating
microshift
into
our
production
chain.
So,
as
you
can
imagine,
you
know
the
integration
into
ci,
the
producing
all
of
our
artifacts
as
part
of
the
reddit
production
chain,
signing
the
content
having
the
documentation
in
place
having
the
qa
in
place.
B
That's
a
lot
of
work,
that's
needed
to
actually
productize
something,
and
so
we're
going
through
that
mainly
and
the
other
part,
is
that
we
basically
want
to
get
mic
shift
into
its
final
shape
for
the
first
product
release.
So
we're
basically
looking
into
can
we
provide
like
a
different
or
better
networking
or
storage
provider
we've?
B
As
I
said
before,
we
we
are
pruning
the
apis
to
really
laser
focus
on
those
production
deployments
as
runtime
and,
unfortunately,
with
the
side
effects
that
a
lot
of
our
community
stuff
currently
is
a
bit
understaffed
right,
but
we
are
of
we'd,
be
super
thrilled
to
have
you
know,
contributions
on
the
on
the
community
side,
because
our
focus
is
on
microsoft.
Rpm.
The
reason
is,
as
I
said,
we
want
to
be
able
to
embed
those
instances
into
rpm
os
trees
and
unless
the
containerized
version
as
a
consequence.
B
So
after
the
first
release
is
out
the
door,
then
we
can
basically
breathe
hard
again
and
catch
breath
and
and
then
focus
on
on
where
we're
going
to
go
from
there
and
there's
like
a
ton
of
stuff
around.
You
know
integration
with
red
for
edge,
so
ensuring
that
you
know
you
have
like
fullest
encryption
and
with
key
stored
in
the
tpm
that
you
have
this
update
process
that
I
mentioned
in
place.
So
a
lot
of
that
stuff
already
exists
in
well.
B
It's
just
you
know
a
matter
of
formalizing
and
qa
in
it.
Multi-Node
is
not
our
initial
priority
and,
to
be
honest,
just
because
customers
feedback
was,
you
know,
hey,
we
want
to
start
with
single
note.
First
probably
need
to
go
and
multi-node
in
the
future,
but
right
now
all
we
want
is
you
know
one
way
of
developing
software,
one
way
of
operating
software,
no
matter
whether
it's
cloud
or
near
edge
or
on
this
device
edge.
So
that's
going
to
be
the
focus
for
us.
A
A
Are
we
ready
because
actually
frank
gave
you
also
an
interesting
point
to
discuss
to
discuss
microchip
usability
today?
So
stage
is
yours.
D
So,
just
you
know,
frank
was
showcasing
the
the
edge
installer
and
how
you
would
create
an
edge
image
and
that's
what
we
were
using
for
the
hackathon
for
f
for
the
hack
fest,
and
if
you
bring
up
my
screen
I
can.
I
can
show
you
what
we
were
using
in
the
past
and
what
we're
going
to
do
in
the
future.
This
is
just
a
the
cockpit
session
of
a
rel
image
and
that
well
virtual
machine.
Has
the
image
builder
installed?
D
That's
the
one
that
frank
was
referring
to,
because
what
you
can
simply
do
here.
Is
you
create
an
image
you
give
it
a
name
and
then
you
add
packages
and
it's
merely
drag
and
drop.
So
this
was
the
image
that
we
actually
built
for
the
first
run
of
the
hackathon,
so
we
put
the
the
the
cockpit
and
portman
stuff
in
it
and
from
there
you
literally
go
and
say,
create
image,
and
then
you
say
rel
for
edge
commit
and
what
you
get
off.
D
That
is
a
simple
structure
with
packets
that
you
just
put
on
a
web
server
and
then
you
can
install
it
from
there.
So
the
next
step
in
our
progress
and
as
frank
already
mentioned,
will
be
to
rip
out
the
department
and
apartment
docker
packets
and
instead
come
in
with
microshift.
D
So
I
already
built
that,
but
sadly
enough,
I
built
it
on
the
fedora
system.
Then
I
deleted,
so
the
the
image
builder
with
actually
microsoft
in
it
was
on
a
virtual
machine.
But
anyway,
if
you
are
interested
in
actually
using
and
trying
microshift,
the
documentation
is
right
here
on
on
microshift.io.
D
It
is
pretty
simple
and
straightforward,
and
actually
for
a
test
environment
that
I'm
using.
This
is
all
you
need
to
do
so
I
got
a
fedora
35
system
and
it
is
literally
is
this
four
lines
of
infrastructure
that
you
have
to
do
that?
That's
all
there
is,
and
then
you
will
install
microshift
on
on
a
fedora
virtual
machine,
physical
machine,
I'm
using
one
of
the
old
notebooks
that
I
no
longer
use
in
in
production,
and
I
run
that
as
a
microshift
server.
D
D
So
we
don't
need
when
our
docker
something
as
fat
as
that,
and
so
that's
what
you
install
the
cryo
tools
and
then
frank's
team
created
an
rpm
where
you
can
install
the
microshift
off
in
in
a
copper
environment
in
a
copper
repository
and
that's
all
there
is
so
there
is
a
description
saying
that
you
can
run
microshift
as
a
containerized
version,
but,
as
frank
may
mentioned,
they're
focusing
on
the
ipm
build,
because
if
you
run
it
as
containerized
itself,
you
would
have
apartment
container,
running
microsoft.
D
That
then
controls
cryo,
which
doesn't
make
a
lot
of
sense.
So
you
just
install
it
as
is
pure
microshift.
So
that's
basically
it-
and
this
is
this-
is
how
you
can
install
it
and
and
this
as
a
practical
thing,
it's
not
only
for
for
edge
things.
That's
the
wrong
window!
That's
the
right
window!
What
I.
B
D
Here,
for
example,
is
a
microshift
environment
that
I
use
for
testing
for
development
and
these
things.
So
in
this
case,
as
you
can
see,
I've
been
playing
around
with
a
couple
of
databases
running
them.
D
Containerized
built
the
stateful
sets
build
the
yaml
codes
to
create
these
applications,
and
you
know:
here's
even
a
microsoft,
sql
server
running
on
top
of
fedora
inside
microshift,
which
is
a
pretty
simple
setup
and
a
cassandra
cluster,
and,
as
you
can
see
from
my
t-shirt,
if
you
mentioned
ea,
so
I'm
doing
a
lot
with
with
ansible
and
for
my
tests
with
ansible.
D
That's
basically
what
I
do-
and
this
is
what
also
is,
is
working
brilliantly
on
on
microshift
is
I
have
the
awx
operator,
so
all
the
kubernetes
operators,
or
most
of
them
work
on
top
of
microshift
the
same
way
as
they
would
on
openshift.
So
in
this
case,
I
rolled
out
the
awx
operator,
which
is
the
operator
behind
the
upstream
ansible
controller,
and
then
that
one
rolled
out
the
postgres
database
and
the
awx
setup,
and
then
you
can
use
it
for
for
test
and
development
purposes.
D
Likewise
here
so
you
have
your
awx
setup
and,
as
you
can
see,
you
can
also
use
concepts
like
roots.
So
you
see
the
c
name
for
the
system
that
points
out
so
the
x,
230,
f,
fedora
31,
that's
the
name
of
the
the
server
and
then
the
cname.
On
top
of
it
points
to
that
application,
which
was
built
as.
D
So
you
see
that
one
as
forwarding
to
the
service
that
I
created
for
for
awx
and
then
you
have
the
service
here,
which
is
that
one-
and
this
is
where
you
can
also
use
the
the
micros,
the
development
and
the
testing
and
then
whatever
you
do
here,
take
it
forward
to
your
openshift
deployments.
D
That's
just
as
a
small
glimpse
and
again,
if
you're
not
too
happy
with
the
command
line,
only
there's
an
open
source
project
called
k9s,
which
gives
you
this
wonderful
text,
ui
in
front
of
kubernetes,
and
that
just
for
a
simple
and
small
demo
of
how
things
work
and
how
you
can
roll
out
the
stuff
easily.
So
with
that
do
we
have
questions
on
the
matter
and
options.
A
D
A
That
that's
fantastic
and
that
also
helps
with
understanding
how
what
we
can
do
and
what
our
audience
can
do
to
support
frank.
A
I
can
see
no
question
in
the
chat
from
from
our
audience,
but
I
actually
have
a
question
for
you,
andreas
if
I
may
so,
you
are
in
the
partner
development
for
the
ecosystem
for
a
long
time
right.
So
you
have
a
quite
extended
and
large
experience.
D
D
Community,
what
we
also
are
doing
is
go
to
everything
is
code,
so
even
the
configuration
of
the
ansible
controller
itself
comes
as
code,
so
you
know
if
you
have
played
with
ansible,
if
you
have
used
tower,
if
you
have
used
controller,
you
see
all
the
configurations
and
templates
and
workflow
and
all
that
kind
of
stuff
that
you
have
defined
in
your
automation
tool
that
is
actually
yaml
code.
D
You
put
that
in
a
github
and
then
you
have
an
ansible
playbook
to
actually
deploy
the
ansible
controller
configuration
and
the
way
you
test
that
is
set
up
a
microshift
environment.
So
you
set
up
a
microshift
environment.
You
get
the
awx
operator
on
it.
You
just
fire
up
an
awx
instance
which
takes
literally
seconds
to
get
up
and
running.
Then
you
take
your
prepared
automation
to
create
your
configured
ansible
controller
out
of
an
empty
aws
setup.
D
You
run
the
playbook,
you
test
it
and
if
everything's
not
the
way
you
expected
it,
you
just
delete
the
instance
and
bring
up
a
new
one
correct
your
code
do
the
same.
So
this
is
a
great
platform
for
testing
and
developing,
and
then
whatever
you
create
here,
you
put
it
on
on
open
shift.
So
as
an
example,
I
I
just
did
some
educational
stuff
on
how
to
run
a
database
inside
an
openshift
or
kubernetes
environment
and
actually
the
code
how
to
create
the
stateful
sets.
D
I
created
on
microsoft
because
it's
easy,
it's
simple,
it's
small.
It
runs
inside
a
virtual
machine
and
once
you
got
all
your
code
ready,
you
can
take
it
then
and
go
to
the
open
shift
environment
and
and
run
it
there
and
for
many
customers.
Probably
that
will
also
be
the
case
because
their
developers
have
to
work
with
something
which
is
as
close
to
the
real
thing
as
possible,
but,
on
the
other
hand,
be
simpler
and
cheaper.
D
So
you
can
sit
on
a
plane
fire
up
your
micro
shift
inside
the
virtual
machine
on
your
notebook
and
then
continue
to
work
being
offline,
and
that's
one
of
the
major
aspects
from
my
perspective
and
make
it
easy
for
especially
people
that
you
know
especially
people
that
come
from
monolithic
applications
and
need
to
learn.
The
kubernetes
learn
the
scale
out
way
and
start
playing
with
applications
ripping
them
into
pieces
in
containers.
D
D
Frank
is
doing
with
with
industry
customers,
especially
now
think
about
the
future
of
that
you
might
have
a
car
and
inside
the
car.
The
only
thing
you
have
is
an
armed
device
with
micro
shift
on
it
and
and
that's
actually
controlling
your
stuff
and
and
and
that's
interfacing
with
your
front
end,
and
you
can
do
a
simple
over-the-air
update
of
a
container
with
your
navigational
system.
Without
you
know,
you
know
what
it
is
you.
C
D
Today
you
have
to
go
to
the
garage
and
then
they
have
to
plug
in
your
25
control
systems
to
25
special
systems.
You
know
I
had
my
car
at
bmw
for
two
days
because
it
took
them
all
night
to
update
my
software.
You
know,
and
that's
thing
of
the
past
simple
thing:
a
simple
system
microshift
on
it
drop
a
new
container
in
it.
Go.
A
Thanks
andreas,
that's
definitely
awesome
yeah,
it
is
definitely
awesome,
so
we
have
a
few
minutes
left.
I
would
like
to
ask
a
couple
of
quick
questions.
If
we
can
use
the
last
few
minutes
we
have
so.
First
of
all
frank,
let
me
ask
you,
is
andrea,
mentioned
an
automotive
use
case.
A
B
Yeah,
so
I
mean
automotive
is,
of
course,
a
changing
space
because
of
all
the
safety
criticality
right.
So
as
long
as
it's
safety
critical
workload,
you
know
you
want
to
create
the
complexity
and
of
the
code
as
small
as
possible,
which
is
why
we
are
actually
looking
into
smaller
form
factors
than
rel.
Is
this
well
in
vehicle
operating
system
right?
B
But,
as
andrea
said,
there
are
maybe
some
other
workloads
that
are
not
safety
critical,
where
maybe
you
might
see
a
microscope
one
day
right,
the
other
part
is
we
have
a
customer
from
the
industrial
space
looking
into
embedding
microshift
into
their
portfolio
of
industrial
iot
gateways.
So
that's
that's
a
pretty
good
use
case.
B
We
have
customers
wanting
to
use
it
on
autonomous
vehicles
right
on
robots,
on
drones
everywhere,
where
you
know,
network
connectivity
is
probably
scars
and-
and
they
want
to
run
microsoft,
for
example,
on
rail,
on
the
india
jetson
family
of
boards.
That
I
mentioned
at
the
very
beginning-
and
maybe
from
my
perspective
the
most
fun
use
case
is
that
microsoft
is
actually
currently
running
in
space.
So
just
end
of
may
there
was
a
a
falcon
9
block.
5
rocket
that
launched
a
cubesat
satellites
like
a
micro
satellite.
B
If
you
want
of
ibm
space
tech
into
orbit
and
ibm's
base
tax
use
case,
is
they
want
to,
as
they
say,
democratize
space
right,
so
they
want
to
give
school
to
age,
children
the
possibility
to
to
basically
develop
their
own
little
application
locally
up,
link
it
into
the
satellite
and
then
access
the
satellites,
cameras
and
and
sensor
feeds
to
do
some
processing
and
send
the
results
down.
So
that's
really
fun
and
ibm's
basic
approached
my
team
because
they
said
well.
We
actually
plan
to
launch
on
a
raspberry
pi
on
a
satellite.
B
Running
a
raspberry
pi,
I
don't
know
why
why?
But
so,
obviously
they
are
extremely
like
not
only
space
constraint
but
also
energy
constraint,
because
those
boards
create
heat
and
heat
is
always
a
problem
in
space,
as
you
can
imagine,
and
so
they
actually
asked
us
whether
they
could
use
microshift
on
that
satellite
to
run
those
students,
workloads
and
and
have
consistency
with.
B
You
know
their
kind
of
the
development
environment
they
make
available
on
openshift
in
the
cloud
right
and
they're
actually
running
makeshift
in
a
very
special
mode,
which
we
call
like
the
all-in-one
container
image.
That
and
the
reason
is
because
they
want
the
ability
to
quickly
suspend
and
so
freeze,
and
also
for
the
mic
shift
instance,
including
their
respective
workloads
to
conserve
energy
or,
basically,
you
know,
just
kill
the
container
start.
A
new
one
and
very
rapidly
have
a
fresh
instance
running
so
that's
kind
of
their
their
use
case
and.
A
Yeah-
this
is
definitely
impressive,
frank,
so
wow
thanks
for
this.
Unfortunately,
we
are
almost
at
the
top
of
our
time,
so
I
would
say
to
close
here
the
session
and
let
me
first
of
all
thank
our
our
producer
tanya
repo
thanks
a
lot
for
the
amazing
job
and
putting
together
all
of
us
thanks
to
our
special
guests
today,
so
michael
frank
and
andrea
andreas,
sorry,
so
I
I
I
really
enjoyed
the
session
for
our
audience
thanks
a
lot
for
joining.
A
If
you
have
any
questions,
please
feel
free
to
reach
out
to
us
anytime,
our
our
also
special
guest
from
natalie's
team,
andre
obiali,
who
supported
us
with
with
the
production
and
and
the
show
overall
set
up
thanks
a
lot.
A
A
quick
overview
of
what's
going
to
happen
in
the
next
episode,
so,
first
of
all,
we
will
launch
the
third
session
of
the
hot
fest
office
hour
on
the
last
wednesday
of
july.
As
usual
as
usual,
we
we
go
live
every
last
wednesday
of
every
month
same
time,
slot
we
will
have
already
some
some.
We
have
already
some
presenters,
so
we
will
have
gunther
herald
the
the
head
of
emea
party
development
team,
and
we
will
give
you
an
overview
of
the
next
use
case.
A
The
hackfest
technical
team
is
is
actually
implementing.
Last
but
not
least,
we
we
are
trying
to
collect
some
some
additional
topics
and
we
would
love
also
to
considering
we
spoke
about
microchip.
Today.
Our
goal
is
to
talk
about
some
control
and
automation,
to
be
adopted
by
the
hackfast
community
team,
and
it's
also
advanced
cluster
management,
aka
acm
and
that's
basically,
the
rough.
B
A
For
the
next
for
the
next
meeting
with
that,
I
want
again
to
say
thank
you,
everyone
guys
any
any
last
comment
from
you
also
michael.