►
From YouTube: OpenStack Austin Meetup
Description
No description was provided for this meeting.
If this is YOUR meeting, an easy way to fix this is to add a description to your video, wherever mtngs.io found it (probably YouTube).
A
B
A
A
C
B
Kick
off,
I
was
planning
to
kick
off
at
seven,
we
talk
is
jim,
is
that
jim
mark
curry
is
mark,
is
mark
curry,
going
to
be.
D
A
B
Here
in
austin
and
we
I
keep
getting
asked
by
people
about
what
freaking
cloud
strategies,
so
anybody
got.
You
know
answers
on
the
postcard
and
take
them
over
here.
It's
funny
on
the
way
back
when
the.
B
A
E
B
B
Cloud
infrastructure
into
a
data
center
since
there's
not
a
lot
of
systems:
engineering
skills
there,
we
basically
do
all
the
engineering
integration
help
them
plug
it
in
and
support
it.
We
were
eucalyptus
and
we're
in
production
with
eucalyptus,
and
we
got
the
hell
out.
F
Two
three:
we
are
on
a
fork
of
one
and
we
were
unable.
B
To
get
our
code
merged
upstream
and
ended
up
on
ford,
which
is
a
really
bad
place
to
be
as
a
business.
So
we
made
the
decision,
we
had
to
either
go
to
eukary
or
something
else,
and
we
chose
something
else
and
we're.
B
C
F
F
A
B
Relations,
this
is,
I'm
rob
hirschfeld.
I've
worked
for
dell,
I've
been
organizing
the
meetups
along
with
the
marketing
team.
Usually
we
have
a
couple
of
guys
from
our
marketing
team.
Also,
I'm
an
engineer
to
help
kick
things
off
so
I'll.
Do
that
I
think
I'll
be
the
user.
So
my
I've
been
from
my
background
with
openstack.
I've
actually
been
involved
with
openstack
and
before
the
announcements
were
made,
and
things
like
that.
So
I've
watched
this
project
get
incubated
for
a
long
time,
and
I
was
a
part
of
the
team.
B
That's
built
the
crowbar
open
source
project,
which
is
was
originally
targeted
as
an
openstack
deployer.
So
one
of
the
things
that's
been
interesting
in
the
dynamics
to
watch
the
market
is
that
openstack
actually
has
more
emphasis
on
deployment
from
a
cloud
infrastructure
than
even
much
more
established
alternatives
in
the
market.
I'm
really
excited
to
see
that
more
deployable,
I
think,
than
any
of
any
of
the
competitive
products
out
there.
B
So
yeah
I'll
do
the
marking
that
I'll
head
off
so
dell
and
puppet
are
the
coast.
Co
sponsors
tonight
dell
rents
this
space.
It's
part
of
our
normal
way
that
we
contribute
to
make
the
meetups
happen,
and
then
we
ask
other
people
in
the
community
there's
a
lot
of
feathers.
B
Interesting,
so
that's
my
introduction.
There
you
go
and-
and
so
you
know
you're,
please
come
in
the
the
couple
things
that
we
ask.
You
know
one
is
you're
going
to
wi-fi
here.
Is
it
says
tech
wretch
guest?
We
have
a
hashtag
that
we
use
for
the
meetup
and
we
try
to
actively
maintain
a
back
channel
to
the
hashtag
osatx.
B
Please
you
know
tweet
and
make
comments
we'll
try
to
monitor
that
as
we
go
and
throw
in
the
hashtag.
So
we
can
track
it
because
it's
the
same
hashtag,
that's
sometimes
used
by
occupy
san
antonio.
B
B
Create
some
really
exciting
cross-pollination
going
on.
So
that's
that's
important,
but
I
want
to
thank
you
for
this.
For
coming
and
being
part
of
the
community.
Please
stay
involved.
We
always
we
try
to
keep
these
interactive
sessions.
We
have
some
new
people
here,
so
we
sometimes
we
take
time
and
talk
about
what
openstack
is.
Sometimes
we
don't
we'll
play
that
by
ear
with
that
I'll
head
it
over
to
dan
is
our
big
sponsor
tonight,
so
he
I
think
he
actually
has
that
referee.
B
B
All
right
so
moving
on
and
it's
too
bad,
I
actually
tried
I
kind
of
got
my
typical
longer
presentation.
What
should
I
shoot
for
in
terms
of
time?
You
know
what
we
we
were
going
to
have
mark
collier
here
to
do
a
foundation
presentation
he's
not
so
5-10
minutes,
especially
if
you're
presenting
so
here's
here,
usually
for
a
sponsor.
What
we
do
is
it's
about
five
minutes.
Past
sponsors
has
been
csa
hp.
B
Dell
chef
we've
had
a
whole
string
of
sponsors,
there's
actually
waiting
lists,
and
I
actually
my
marketing
team
would
be
upset
if
I
didn't
say,
if
you're
interested
in
sponsoring
one
of
these
meetups
you're,
basically
buying
the
pizza
or,
if
you
want
to
spring
for
fancier
food
you're,
welcome
to
do
that.
We,
we
are
always
soliciting
people
who
want
to
sponsor
the
meetups.
We
also
coordinate
them
in
boston.
B
So
if
you
have
an
office
in
boston,
we
encourage
that,
but
usually
it's
about
five
minutes
for
this
sponsor
to
stand
up
and
talk
about
what
their
product
is
and
how
their
pieces
work.
And
then
we
move
on
to
other
content.
B
So
I
just
had
a
few
slides
and
I
was
going
to
talk
a
little
bit
about
kind
of
our
integration
efforts,
why
we
think
we
bring
value
to
the
openstack
community
and
a
little
bit
about
the
project
that
I've
been
working
on.
So,
as
I
said
before,
my
name
is
dan
billy.
I'm
going
to
do
a
real.
A
B
Puppet
is
but
puppet's
basically
a
configuration,
and
I
guess,
for
the
sake
of
time,
we'll
just
go
right
into
kind
of
the
graphical
representation
of
what
is
public.
So
this
is
basically
from
its
model
that
you
want
something
to
be
in
a
certain
state.
That
thing
is
in
a
real
estate,
so
what
puppet
does?
Is
we
describe
state
using
what
we
call
resources?
B
B
If
it's
a
debian
based
system,
we
might
run
defective
squared,
then
puppet
basically
says:
are
they
the
same?
If
they're
not
the
same,
we
create
an
event
in
order
to
remediate
which
would
cause
it
to
synchronize.
So
in
this
example,
maybe
a
young
install,
maybe
an
app
get
updated,
and
the
last
thing
is
that
we
actually
describe
everything
as
state
transitions.
So
with
puppet
you
get
this
historical
record
of
configuration.
Management
has
state
transitions.
B
You
know
these
things
have
transitioned
from
this
state
to
this
state
at
these
times
kind
of
historical
record,
for
your
your
system,
the
last
cool
thing
about
puppet.
This
is
our
model.
We
support
something
called
no
off.
This
is
noah.
You
don't
have
to
remediate.
You
can
just
get
these
the
state
transitions
that
would
occur
without
actually
having
to
affect
your
systems.
B
B
B
The
same
every
single
time
second
thing
is
deployment.
Expertise
needs
to
be
consolidated,
there's
so
many
companies
deploying
in
so
many
different
ways.
If
you
codify
those
deployment
practices,
people
can
share
them.
People
can
collaborate
by
the
way
everything
I'm
doing
is
open
source.
I
have
a
long
list
of
large
companies
that
are
collaborating
with
who
are
helping
to
codify
their
best
practices
in
these
public
modules,
so
that
people
can
share
and
learn
from
each
other
and
grow.
B
B
Openstack
is
hard.
We
can
capture
this,
we
can
describe
them,
so
we
have
a
bunch
of
openstack
related
content
for
people
that
look
the
same
experience.
This
stuff
might
look
familiar,
we
have
things
for
horizon
and
we
call
these
modules
and
keystone
glance
and
swift
and
nova
even
a
bunch
of
peripherals
things
for
managing
rapid.
B
Which
is
like.d
directory,
implementations
for
single
files
and
ndp,
and
our
sync
and
seriously
have
been
you
know
going
on
to
include
things
like
articulation
for
centralized
logging
and
various
things
for
monitoring
the
collection,
basic
ideas.
What
I've
implemented
is
a
bunch
of
building
blocks,
and
this
is
an
example
of
just
the
nova
building
blocks.
So
if
you
want
to
build
a
node
api
server,
no
compute.
B
B
So
so,
essentially
each
one
of
these
is
actually
a
configuration
interface
that
represents
this
logical
component,
so
these
components
present
configuration
interfaces
on
top
of
the
openstack
components.
As
an
example,
keystone
might
look
like
this
from
the
user
perspective.
What
is
it
it's
a
keystone?
It
has
a
blind
host,
a
public
port,
an
admin
port,
an
admin
token
and
a
compute
port,
but
the
reality
is
it's
actually
composed
of
packages
in
groups
and
files
that
are
configured
based
off
those
interfaces.
B
B
So
one
way
that
people
are
using
this
is
that
this
is
kind
of
the
initial
way
is
you
know
you
have
expertise,
you
do
it
your
way,
pick
our
building
blocks
and
use
that
to
compose
the
way
that
you
want
to
do
it
and
here's
an
example.
You
know.
Let's
say
I
want
to
compose
things
like
this.
Maybe
I'm
going
to
build
something
called
a
controller
okay.
What
does
that
consist
of
I'll
need
a
memcache?
I
need
a
keystone
where
each
of
these.
B
B
Or
if
you
would
rather
get
more
constrained
ways
of
deploying
things,
I
have
interfaces
called
openstack,
compute,
openstack
controller
and
openstack
all
so
this
is
not
all
of
the
configuration
that's
available.
For
example,
we
have
something
called
an
openstack
controller.
It
supports
a
public
address,
the
public
interface,
the
private,
address,
private
interface,
all
kinds
of
passwords
for
all
the
things
you
would
expect
an
ipv4
range
floating
range
and
maybe
what
network
manager
you
want.
So
now
you
can
associate
that
with
nodes
that
are
going
to
be
controllers.
B
Maybe
some
of
those
are
going
to
compute
nodes
and
you
would
have
configuration
interface,
so
you
can
think
of
compute
nodes
in
terms
of
what's
the
private
interface.
What's
the
private
address?
Well,
how
do
we
connect
the
sql?
How
do
we
affect
the
rabbit?
What's
the
libert
type
and
is
it
running
multi-house.
D
B
B
B
B
B
Should
the
node
be
authoritative
versus,
should
you
have
a
central
data
school
and
I
guess
we
support
both
if
this
is
one
way
to
go
and
here's
some
references
that
here,
this
is
again
repository
everything
I've
talked
about
is
open
source,
so
we
have
a
git
repository
which
actually
has
a
rate
test,
it'll
clone
in
the
other
git
repositories,
and
this
contains
openstack
all
openstack
controller
openstack
compute.
B
B
Is
you
know
some
of
the
some
of
the
times
we
do
with
meetups,
where
we
talk
down
and
down
details
and
real
specific
deployment
things
and,
and
sometimes
we'll
we'll
we'll
step
back,
and
I
would
say
too,
that
kind
of
my
audience
that
I'm
most
interested
in
targeting
is
people
who
have
openstack
knowledge,
yeah
people
who
are
struggling
to
deploy
openstack
with
things
like
dev
stack,
which
isn't
really
the
fundamentals
and
and
we're
going
to
talk
about
that
stack
actually
as
part
of
the
topic
for
this
meeting.
B
B
There
are
people
who
are
targeting
clusters
of
over
a
hundred
thousand
dollars
in
this
utility,
but
I
mean
in
the
long
term-
that's
not
today,
but
there
definitely
are
people
who
are
talking
to
you
about
over.
D
B
B
So
I
think
a
lot
of
people
are
are
still
in
the
state
of
trying
to
figure
out
what
tools
to
invest
in
so
great.
So
one
thing
that
jim
suggested
and
and
I'll
repeat
this,
the
group's
already
interrupted
in,
but
what
we're
going
to
start
we'll
start
doing
is
we'll
do
a
30
minute
intro
to
openstack
as
part
of
the
the
meeting
prep
and
we'll
just
build
that
into
the
schedule,
so
6
30
to
7
we'll
do
this
is
the
background
of
openstack.
B
I
guess
this
is
kind
of
a
newer
audience
than
the
general
crowd.
It
is
always
true.
B
Physical
blade
or
some
physical
set
of
hardware
that
needs
to
be
used,
and
in
openstack
terms,
you
would
generally
have
some
nodes
that
are
dedicated
to
being
controller
nodes
and
they're,
basically,
managing
the
process
of
managing
virtual
machines,
but
most
of
the
nodes
in
openstack
are
going
to
be
actual
compute
nodes
or
nodes,
they're,
actually
running
hypervisors.
That
need
to
be
managed
by
the
controller.
B
Number
of
services
to
be
able
to
coordinate
so
they
can
broadcast
their
capabilities.
So
so
you'll
have
the
main
controller
and
they
will
have
the
individual
nodes.
A
B
G
B
And
in
this
example,
everything's
kind
of
combined
and
that's
initially
what
we've
been
targeting,
but
each
of
these
are
essentially
things
that
could
be
a
lot
of
these
higher
level.
Ones
are
things
that
could
be
associated
with
english,
and
even
things
like
compute,
even
compute,
can
be
split
into
in
number
of
nodes,
and
even
compute
has
an
api
service,
a
schedule
or
a
network
management
service
where
those
things
can
reside
on
their
own
notes,
but
really
the
main
components
here
is.
In
this
controller,
you
have
clients,
which
is
the
image
database,
which
manages
the
images.
B
Have
keystone,
which
is
managing
all
the
authentication
and
authorization
for
the
system,
and
often
swift,
which
is
the
remote
object,
store
and
horizon,
is
the
front
end.
B
And
in
this
case,
it's
showing
the
the
simplest
case
so
put
all
the
control
stuff
on
onenote,
in
this
case,
just
split
up
the
database
and
rather
than
q
server.
And
then
what
we
really
care
about
scaling
in
general
are
the
compute
nodes,
which
are
the
actual
hypervisor
nodes
that
are
being
managed
by
the
controller.
B
Like
like
redundant
configurations,
like
you
mentioned,
you
know
explaining
on
your
nice
cluster
or
out
of
cube
if.
D
B
Now
you
could,
you
could
assign
the
the
you
know
you
could
it's
the
same
configuration
element
which
you
can
assign
to
number
of
nodes,
but
the
code
for
setting
up.
However,
the
aha
is
done,
isn't
it
so,
I
think
you're,
basically
going
from
infrastructure
to
platform
and
openstack's
infrastructure
right,
so
yeah
so
like
setting
up.
B
Each
one
of
people-
and
I
think
that
as
well-
that's
a
lot
of
you
know
this.
This
is
me
kind
of
leading
the
effort,
but
it's
also
a
lot
of
companies
who
are
really
dedicated
openstack
that
are
implementing
a
lot
of
these
things
and
so
everyone
who's
working
on
on
this
project.
Even
though
it's
it's
pretty.
I
basically
started
in
january
on
this
stuff,
even
though
it's
pretty
new
everyone's
invested
that
whenever
these
things
hit
production,
it's
going
to
be
scalable.
So
there
will
be
a
lot
of
representations
for
that.
B
But
in
terms
of
aha
implementations,
a
lot
of
the
services
for
openstack
itself
are
inherently
designed
to
be
deployable
to
be
highly
available,
but
specifically
the
some
of
the
middleware
stuff
right
now
you
would
just
wind
up
kind
of
taking
these
things,
assign
them
to
multiple
nodes,
and
you
have
still
have
some
remaining
tasks
of
figuring
out.
You
know,
is
it
a
vip
and
a
load
balancer
or
is
it
what
are
hddl.
B
Of
in
discussion,
I
know
that
we
are
planning
on
excuses
planning
on
supporting
and
doing
the
the
clustering
of
the
various
pieces,
whether
it
be
the
horizon
or
the
controller.
F
B
A
Full-Blown
production
cluster.
F
Of
compute
nodes,
or
are
you
using
something
like
popular
chef.
A
B
So
our
initial
efforts
right
now
were
around
getting
crowmark
working,
which
included,
which
included
doing
chef
work,
for
that
does
not
necessarily
mean
that's
gonna,
be
the
end,
all
be
all
for
deployment
tools,
but
it
was
in
the
it
was
a
necessary
piece.
Recipe
will
work
without,
for
example,
and
so
I
know,
we've
got
puppet
in
other
places.
The
sections
are,
you
know
who's
the
manager,
which
is
our
you
know,
management,
tool
and
whatnot,
and
that
led
to
some
some
puppet
pieces
and
so
we're.
B
B
So
what
we're
looking
at
is
trying
to
figure
out
what
is
going
to
be
the
best
strategy
overall,
that
clustering
is
going
to
be
done
using
our
high
availability
stack.
B
Of
things,
but
but
how
we
go
about
doing
that
in
terms
of
strategy
and
and
how
much
of
that
is
going
to
be
done
internally
versus
you
know
and
release
out
to
everybody,
and
just
say
this
is
what
we're
doing
you're
welcome
to
jump
on
board
versus
what
we're
trying
to
do
is
avoid
that
and
actually
collaborate
with
the
community,
and
you
know,
make
decisions
together.
Okay,.
B
F
My
question
is:
openstack
is
an
infrastructure
and
there
are
lots
of
infrastructures.
You
know
my
sequel
well,
sql
period,
infrastructure
and-
and
there
are
tools
for
deployment
that
are
based
upon
these
things,
I'm
getting
the
impression
with
openstack,
if
they're,
just
one
after
another,
a
b,
c
d,
e,
f
and
g-
I
don't
know
all
the
names,
I'm
you
know,
I'm
wondering
why
are
there
number
one?
Why
are
there
so
many?
I
can
say
from
our
perspective,
our
customers
basically.
D
B
So
so
let
me
let
me
step
back
and
maybe
help
answer
some
of
that,
because
and
some
of
and
I'll
I'll
explain
a
little
bit
about
what
crowbar
does,
because
it's
it's
an
interesting
thing,
because
we
we
saw
a
very
similar
problem
when
we
were
first
starting
out
with
what
it
would
take
to
deploy
openstack
and
there
are
a
lot
of
pieces
and
there's
a
lot
of
codependencies
openstack
is
a
is
cloud
software.
B
It's
a
sophisticated
suite
of
interconnected
tools
that
provide
that
where
the
goal
is
effectively
to
replicate
the
same
set,
not
the
same
apis
exactly
but
the
same
api
capabilities
as
amazon
has
plus
new
capabilities
and
new
enhancements
right,
so
that
that's
the
bottom
line
goal
amazon
you
know
for
better
or
for
worse,
set
the
standard
for
what
it
was
to
have
api
based
infrastructure.
Okay,
it's
it's
worthless
to
try
and
pretend
anything
else.
They
have
done
an
amazing
job,
building
a
market
establishing
establishing
a
presence,
and
so
the
question
becomes.
B
If
you
want
to
build
an
infrastructure
for
api
accessible
or
if
you
want
to
build
out
operational
capability
for
for
api
driven
infrastructure,
which
is
really
what
cloud
is
remote
api
driven
infrastructure?
That's
one
depth,
one
definition,
and
if,
if
you
want
to
build
up
that
infrastructure,
it
is
a
lot
of
work,
it's
a
complex
thing
right.
You
have
to
have
message
buses.
You
have
to
have
api
servers,
you
have
to
have
load
balancers.
You
have
to
have
storage,
cache
and
block
storage
and
data
databases.
B
You
have
to
have
all
these
pieces
that
fit
together.
You
have
to
have
centralized
authentication.
You
have
to
have
a
way
to
store
big
blocks
of
data.
You
have
to
have
something
to
actually
manage,
manages
virtual
machines.
You
have
to
have
something
that
that
keeps
your
networking
in
insane
way
and
you
and
then
to
do
it
in
a
way
that
actually
satisfies
a
broad
market
right.
There's
three
networking
modes
built
in
openstack
today:
quantum
there's
a
whole
new
programmable
networking
model.
That's
coming
called
quantum.
B
B
B
That's
smaller
also,
so
it's
an
object,
storage
system
dashboard
is
the
ui,
also
known
as
horizon,
where
you
can
actually
interact
with
all
these
pieces
and
and
horizon
actually
is
a
recent
addition
to
openstack
as
an
official
as
a
release
product,
but
it's
a
hugely
critical
thing,
because
it
lets
you
that
you
have
a
way
to
interact
with
the
cloud
without
needing
to
know
the
apis
right
up
until
this
release,
the
official
way
to
interact
with
openstack,
just
like
the
origin
original
way
to
interact
with
amazon,
was
via
apis
only
and
then
keystone
is
an
integrative
piece
that
brings
it
all
together
by
allowing
you
to
have
centralized
authentication.
A
B
Well,
but
hold
up
so
but
there's
a
problem.
The
problem
here
is
complexity.
Right
for
anybody
in
this
room
to
get
up
and
running
with,
openstack
requires
a
level
of
understanding
that
you
know
frankly
takes
time
to
build.
It
can
be
a
daunting
task,
the
the
reason
why
we
started
our
project.
The
reason
we
got
involved
is
we
started
and
frankly,
we
started
with
puppet.
B
We
started
building
doing
our
work
with
puppet
when
we
were
first
starting
out
is
the
level
of
information
that
you
had
to
have
just
to
use
the
type
of
great
coding
work
that
dan
is
doing
right
or
the
the
puppet
the
chef
guys
had
been
doing.
Also,
then
any
just
to
leverage.
B
Yes,
and
it
also
required
you
to
spend
a
lot
of
time.
Just
you
know
installing
operating
systems
and
network
modes-
and
we
have
you-
know
significant
stories
of
people
failing
their
whole
cloud,
failing
because
they
didn't
know
how
to
configure
their
switch
correctly
or
that
their
switch
wasn't
configured
correctly.
B
So
the
the
work
we
did
was
to
help
people
bootstrap
into
that
state.
Right,
crowbar's,
real
job
is,
is
not
you
know,
we
build
the
whole
cloud
and
end
and.
B
B
B
A
A
C
B
A
B
I
would
say
that
it
has
the
momentum
in
terms
of
drawing
talent,
to
the
community
to
become
the
thing
that
can
compete
with
amazon,
and
I
think
that
it's
I
that
seems
like
that's
true
as
of
the
last
release,
which
is
something
stable
and
but
something
that
still
needs
to
get
better.
But
I
think
that
there's
enough
there
for
that
to
the
last
release,
his
code
name
was
essex.
The
one
before
that
was
diablo.
C
B
Right
there,
so
the
the
here's,
the
here's,
the
thing
about
maturity
for
openstack
I
there
are
people
there
have
been
people
who
have
been
deploying
it
since
the
the
c
release
the
cactus
release
and
and
diablo,
and
then
you
know
definitely
essex.
B
Every
release
that
comes
along
makes
it
easier
for
people
to
make
a
smaller
knowledge,
investment
to
be
productive
and-
and
that's
really
what
the
trend
is.
So
you
know
jim's
right
we're,
seeing
a
tremendous
adoption
that
the
market
drive
around
openstack
is
is
very
large
right.
People
are
definitely
we're
building
an
ecosystem.
Things
are,
things
are
moving
very
quickly
and
the
we
in
this
case
is
the
openstack
community.
B
The
the
the
challenge.
The
challenge
is
that
every
as
we
turn
this
crank
we're
trying
to
make
it
easier
and
easier
for
people
to
come
in
right
so
that
that's
been
frankly
dell's
role
in
openstack
from
the
very
first
day
that
we
started
right
in
the
austin
with
what
was
the
austin
conference,
the
first
the
pre-release
conference.
B
We
signed
up
to
say
that
that
we
want
our
customers
to
be
able
to
use
this.
Actually,
it's
for
us,
it's
community,
because
we
have
an
open
source
project
to
be
able
to
use
openstack
with
the
minimal
investment
minimal
time
commitment.
We
had
a
four-hour
goal
of
being
able
to
be
have
an
open,
stack,
totally
deployed
from
nothing
so
and
then
we
actually
exceeded
that
goal
to
be
less
much
less
than
two
hours
much
less
than
four
hours,
but
those
those
types
of
things
those
invest.
B
Incredible
site:
that's
run
by,
and
gentle
who
often
comes
to
these
meetups
trying
to
make
things
much
easier
to
use
and
go
hp
has
been
contributing.
The
operating
systems
has
so
yeah,
so
I
mean
canonical.
Suse
and
red
hat
have
have
all
contributed
code
att
ntt
in
japan.
It's
an
international
effort
and
I'll
say
from
our
customer
customer
flow.
We
see
worldwide
adoption
of
this
product.
It's
it's
truly
staggering.
F
H
F
A
B
F
B
A
F
B
E
C
B
Do
and
actually
thank
you
for
the
segway,
so
I
was
going
to
hand
it
over
to
correct
so
I'll
if
you
want
I'll,
give
you
a
short,
a
very
short
intro
in
what
the
topic
is
and
then
I'll.
Let
you
kick
over
from
there
go
ahead.
B
E
Well,
good
evening,
everyone,
hello,
I'm
craig
and
I'm
pleased
to
actually
be
talking
a
little
bit
about
devstack
night.
I
have
hope
and
had
hoped
to
be
doing
this
for
a
couple
months
now
I
volunteered,
I
guess,
background
the
first
or
second
meeting,
partly
as
a
way
to
get
used
to
doing
presentations
again.
So
if
it's
always
I'm
not
the
best
public
speaker
in
the
world,
that's
the
reason
why.
The
first
thing
I
wanted
to
start
with
was
just
to
throw
something
up
to
kind
of
answer
your
questions.
E
And
so
the
other
question
I
was
going
to
get
up
and
draw
for
you.
I
guess
how
you
do
it
offline
afterwards
is
so
the
idea
is
kind
of
like.
D
E
All
of
those
things
you
mentioned,
like
storage
and
authentication
and
virtualize
them
so
that
you're
at
the
end
of
the
day,
all
we
care
about
is
applications
running
on
our
computers
right
and
so
the
whole
point
is
to
enable
your
development
team
or
whatnot
to
concern
themselves
less
with
os
details
and
having
to
manage
hardware
and
bring
systems
up
and
and
make
it
more
so
that
they
just
they
deal
with
a
known
blank
image,
and
then
they
use
puppet
or
whatnot
to
bring
that
image
up
to
a
deployable
state.
E
And
so
that's
exactly
a
bizarre
voice.
I
guess
one
of
the
kind
of
two
ways
that
we're
doing
we're
thinking
about
doing
dev
stack
this
one
just
to
manage
our
infrastructure
so
that
we
don't
have
the
hassles
of
dealing
with
individual
systems
anymore
and
but
in
general
in
general.
We're
moving
a
lot
of
stuff
to
the
cloud
exactly
for
that
reason,
so
we
can
say
well
that
node's,
obviously
just
behaving
cool,
kill
it
and
let
let
another
one
come
up
and
take
its
place.
E
E
E
Is
actually
with
real
hardware,
so
that's
going
to
be
the
unfortunate
thing
in
the
past
couple
months,
I've
switched
jobs,
and
so
now
I
had
to
kind
of
shoehorn
things
into
my
virtual
machine
running
on
my
laptop.
Where
did
you
go
from.
E
And
you
can
see
that
basically,
they
have
a
couple
of
different
ways
that
you
can
download
and
run
it.
So
you
run
it
on
vmware
vms,
which
is
what
I'll
try
and
do
tonight.
E
So
dev
stack
is
yeah,
so
it's
a
basically
a
way.
It's
meant
to
be
a
way
for
people
that
are
developing
on
openstack
to
to
actually
have
that
employment
environment
without
having
to
flush
with
the
consultant
deal
and
actually
deal
with
it,
the
openstack
infrastructure
and
having
all
these
individual
nodes.
So
it's
a
way.
D
B
D
E
Actually
answer
that
on
the
fact
about
what
it
is
and
why
someone
asks
well,
why
not
puppet
or
why
not
promo
yeah,
the
idea
is
basically
the
intent
is
for
developers
basically
to
have
a
good
clean
development
environment
and
when
you
actually
use
it
in
the
all
on
ones
all
on
one
system
method
of
using
the
openstack.
Actually
they
even
mentioned
that
you
could
be
on
an
airplane
encoding
and
basically,
how
you
have
your
you:
have
your
infrastructure
there
running,
but
you're.
B
A
E
I
mean
actually
that
was
my
intent
earlier
when
I
first
got
interested
in
this.
I
think,
because
I
had
all
this
hardware
at
my
last
job.
It
was
just
sitting
around
to
be
used
as
test
equipment.
B
B
You
know
give
me
these
stable
lists
of
packages.
B
B
I
say
the
point
in
time:
it's
not
necessarily
okay,
this
is
the
essex
or
the
final
diablo
release,
but
more
like
you
know,
for
example,
when
we,
when
susan
generates
its
its
product,
it's
gonna
have
okay,
specific
versions
of
these
specific
packages.
B
We've
all
been
testing
together,
we
know
it
all
works
together,
they're,
coming
together
and
building
this
product,
you
can
add
additional
stuff
on
after
the
fact
right,
but
in
terms
of
a
stable,
essential
release,
quality
thing
the
the
dev
status
is
wearing
is
what
people
who
are
developing,
so
they
need
to
have
the
latest
and
greatest
leading
edge
code.
They
haven't
been
necessarily
through
all
the
the
rigors.
E
Great
thanks
so
actually
downloaded.
They
made
it
very
used
user
friendly
these
days,
so
there
is
basically
what
they
call
this
user
data
script,
and
you
can
see
this
is
on
overview
by
the
way
they
actually
support,
fedora
and,
I
believe
something
else,
but
it's
a
good
chance
for
me
to
actually
download
and
get
used
to
looks.
E
Basically,
you
just
run
this
script
and
you
can
see
it
downloads.
The
git
repository
sets
up
a
sort
of
minimal
environment
which
they
call
a
local
rc,
and
it's
just
as
simple
as
running
this.
Shell
script
stack
top
shell
and
we're
actually
going
to
have
to
run
that.
I'm
not
actually
quite
sure
what
will
happen
at
this
point,
but
you
can
see
the
dev
stack
holder
is
actually
downloaded.
E
Unfortunately,
I
think
when
I
put
this
to
sleep,
the
virtual
machine
got
confused,
so
we'll
just
kind
of
mimic
running
that
stack.
Shell.
B
E
Yeah,
it
was
actually
quite
nice.
There
we
go.
Maybe
I
mean
the
end
result
is
at
this
point
which
was
running
earlier.
Is
you
get?
This?
Is
the
nice
little
dashboard
that
everyone
was
talking
about
earlier
and
at
this
point
we
can
start
to
bring
machines
up?
I
don't
know
if
it's
gonna
actually
work.
E
Maybe
so
it
looks
like
it
is
trying
to
download
something,
but
this
devsec
dev
stock
desktop
shell
is
actually
pretty
good,
so
it
downloads
all
the
components
that
I
needed
to
actually
get
this
running
and
another
link
that
I'm
going
to
actually
show
is.
E
Another
web
page
that
recently
I
found
that
actually
shows
answers
the
question
that
about
how
you
would
actually
download
the
individual's
openstack
source
code.
So
in
that
local
model
rc,
you
can
actually
you
can
actually
tell
it
all
the
components
that
you
want
to
download,
and
so
this
web
page
actually
is
a
pretty
nice.
D
E
E
I
I'll
be
happy
to
test
that
if
you
want
offline
yeah.
F
B
You
just
have
to
get
the
right
branch
when
you
check
out
the
code,
so
part
right.
Part
of
part
of
the
goal
with
this
is
that
you
can
actually
get
these
get
dev
stack
connected
to
the
the
code
as
it's
being
checked
in
so
right.
This
is,
this
is
has
been
a
challenge
for
us
as
a
consumer
of
openstack.
If
a
developer
changes,
something
that
new
check-in
might
change
the
whole
deployment
or
change
configuration
settings
or
something
like
that,
so
it
varies
bran.
A
B
B
The
the
key
is
that
a
lot
of
the
deployment
mechanisms
that
we're
using
require
somebody
to
do
a
package
so
build
all
this
code,
put
it
into
a
dev
or
rpm,
or
something
like
that
that
you
can
pull
into
your
deployment
for
crowbar.
I
don't
know
if
dan,
if
you're
doing
the
same
thing
yeah,
I
suspect
you
are
you?
Actually
you
actually
use.
B
You
know
you
use
those
packaged
repositories
for
the
install,
so
devstack
is
very
different
because
it
actually
pulls
in
the
python
code
and
then
deploys
the
python
code
and
being
able
to
do
that.
Allows
you
to
be
much
more
aggressive
about
staying
on
trunk
and
getting
the
very
latest
code,
which
is
why
it
becomes
the
gate
for
check-ins.
You
must
run
on
devstack
before
you
can
check
in,
because
everybody
can
then
test
your
code
before
they
accept
the
push
and
the
folks
in
that
run.
E
B
E
Looks
like
it's
having
problems,
I
think
part
of
the
thing
is
is
because
I
didn't
want
to
actually
have
to
install
an
operating
system.
I
went
down
there's
a
site
that
actually
is
distributed
images,
but
it
looks
like
the
guy
that
did
this
is
in
italy,
and
so
that
was
already
a
big
challenge
having
to
figure
out
how
to
remap.
B
E
E
B
E
C
F
Should
can
I
install
this
on
a
free,
nas
machine
which
I
think
is
previously.
B
B
E
Fact
the
dev
stack
script
itself
will
it'll.
If
you
don't
have,
I
guess
the
the
footwork
that
they
support.
They
actually
will
tell
you.
You
have
to
force
it
to
run,
and
I
found
pretty
quickly
when
I
downloaded,
I
think
1104.
E
For
example,
all
of
the
packages
were
wrong,
so
I
mean
it's,
it's
pretty.
You
can
actually
know
at
some
point
someplace
on
this
website.
They
actually
walk
you
through
that
that
shell
script,
that
that
stack.sh
script
line
by
line-
and
it's
it's
pretty
precise
about
what
it's
expecting,
and
so
I
don't
know,
could
you
hack
together?
You
know
a
bsd
version
of
it
probably,
but.
B
D
B
Is
doing
a
lot
of
work
to
make
it
equal
or
on
par
experience?
And
that's
part
of
that's
part
of
maturing
the
environment,
and
we
we
need
to
invest
in,
is
you'll,
be
able
add
in
the
repository
whether
it's
you
know
either
a
cd
tray
or
a
network
repository
and
then
just
go
in
and
click
the
check.
A
E
He
goes
and
that's
a
supported
model
here
too,
if
you
had,
I
guess,
a
machine
that
was
adjustable
to
one
of
these
and
then
they
have
the
all
in
one
machine.
I
just
have
to
have
windows
because
we
were
very
heavily
invested
in
outlook.
So
I,
if
I'm
going
to
have
outlook,
I
prefer
to
actually
manage
it
with
proper
tools.
B
Supposedly
so
the
other
topic
that
we
had
for
tonight,
although
usually
we
stop
at
eight,
but
I'm
certainly
going
to
be
done.
It
was
to
talk
about
the
folsom
conference
right,
the
folsom
summit
conference
and
summit's
the
right
term.
Do
people
want
to
go
a
little
bit
longer
and
we
need
to
talk
about
folsom
for
a
little.
C
B
So
I'm
not
I'm
not
sure
that
we,
you
know,
I'm
not
sure
the
three-way
dialogue
about
what
we
saw.
Is
it's
going
to
be
that
useful,
but
we
can
talk
about
it
briefly.
So
here's
what
I
would
suggest
we
can
do
like
five
minutes.
You
probably
you
want
to
talk
here.
We
want,
we
can
do
like
five
minutes
of
impressions
about
it
and
that
and
sort
of
sum
it
up.
That
way,
there's
actually
been
some
some
posts,
a
lot
of
great
posts
about
what
happened
at
the
summit
and
some
impressions
and
ideas.
B
Was
is
there
something
that
stuck
out
in
your
mind
in
the
conference?
It's
just
so
many
smart
people
that
are
experts
and
stuff
that
I
don't
understand
like.
I
think
I
feel
like
I
bring
my
expertise
to
the
table,
but
so
many
people
are
collectively
there
that
bring
their
their
expertise
to
the
table
that
it's
it
was
amazing.
Is
there
a
session
that
that
stuck
out
in
your
mind,
just
being
really
good?
I
actually
hung
out
most
of
the
time
in
the
developer,
room
and
just
talked.
A
B
It
was
awesome,
I
I
I
did
a
lot
more
in
the
summit
than
in
the
conference
right.
That's
that's
much
more
interesting
and
from
the
poor.
How
would
you
describe
the
format
to
people
so
that,
because
it's
a
different
type
of
event,
if
you're
not
used
to
it?
Well,
how
would
you-
and
this
was
your
first
comment-
yeah?
I
was
at
the
conference
in
boston,
but
not
at
the
summit.
Okay,
I'd
say:
the
format
is
just
packed.
It
was
so
packed
with
with
sessions
and
ideas.
B
It
was
hard
to
know
what
to
pick,
but
it
was.
It
was
great
in
that
there
were
so
many
people
presenting
on
so
many
different
kinds
of
specific
messages
that
there
is
it's
potentially
a
great
way
to
at
least
skim
the
surface
on
a
lot
of
topics,
but
even
more
so
to
make
relationships,
you
can
help
build
expertise
as
well.
That's
one
of
the
things
that
is
important
to
understand
about
the
summit
is
a
lot
of
times.
B
It's
a
relationship
and
you
need
somebody
where
real
work
and
discussions
often
happen
after
the
summit
on
irc
and
chat
and
scheduled
meetings.
So
if
you're,
not
at
the
summit,
you
didn't
miss
anything,
but
you
didn't
miss
the
chance
to
have
sort
of
the
one-on-one
or
facetime
with
people
who
are
making
the
decisions.
C
A
Trunk,
like
the
compute
trunk
and
the
swift
chunk
and
the
identity
child
meet
with
people
who
they
might
not
otherwise
talk
to
and
get
feedback
from
them
and
from
everyone
on
how
to
design
the
next
version.
A
So
it
had
the
essence
six
months
ago,
in
boston,
there
was
still
quite
a
bit
of
debate
about
how
to
make
openstack
succeeds
that
we
needed
to
put
it
in
the
foundation
and
we
needed
to
do
this.
We
needed
to
do
that
if
we
didn't
do
these
various
nonsense
money
months
at
this
design
summit,
the
whole
system
in
san.
B
A
B
B
C
B
B
D
A
A
E
A
B
B
Yeah,
it's
so
a
lot
of
some
of
those
things
in
openstack
that
are
that
are
required
to
that
are
required
in
order
to
be
successful
with
the
cloud
deployment
right,
they're,
all
personal,
so
favorite
session
session.
That
stands
out
in
your
mind,
of
the
entire
quantum.
B
A
A
That,
if
you
wanted
a
private
cloud,
your
real
choice
would
be,
and
we
wanted
to
offer
that
to
it.
B
B
So
the
track
out
the
track
I
was
following
a
lot
was
the
the
block
storage
track
so
I've?
I
I
fully
agree
with
you
about
openstack
having
having
crossed
certain
thresholds.
B
Block
storage
is
potentially
the
weakest
right
now
component,
because
the
architecture
for
for
well
the
architecture
for
openstack
and
hyperscale
clouds
has
been
local
storage,
predominantly
not
remote
or
network
storage.
B
So
we're
seeing
that
as
a
significant
place
of
growth
and
opportunity
that
our
customers
are
asking
for,
and
there
was
a
really
good
session
and
so
there's
two
types
of
sessions
at
the
openstack
conference
and
very
little
in
between
there's
the
this
general.
B
We
are
going
to
create
world
peace
with
the
cloud
by
doing
load
balancers
as
a
service,
and
here
is
the
state
diagram
of
the
idealized
version
and
those
have
like
200
people
sitting
in
them
and
some
guys
up
front
and
not
much
happens
like
it's
the
incubation
of
something
cool,
and
then
you
have
the
opposite
side,
where
it's
10
people
clustered
up
around
the
front.
Saying
all
right.
B
I'm
going
to
write
this
subroutine
and
you're
going
to
write
that
subroutine
and
then
we
have
to
consider
this
and
so
the
block
storage
guys
sort
of
started
up
there
and
then
they
snuck
off
in
a
room
and
there
I
was
part
of
a
really
good
session
that
was
five
or
six
companies
xenta
solidfire
saf
dell
was
there
cisco.
B
Is
there
the
cup
two
or
three
others
that
I'm
not
remembering
off
top
of
my
head,
but
so
literally
competitors
in
the
in
in
the
block
storage
space
sitting
down
together,
saying
all
right
we're
going
to
build
this
api
with
this
code.
We
have
to
consider
this.
We
can't
break
very
significantly
and
we
can't
break
the
way
it
works,
so
it
was
really
significant
from
collaboration
from
dedication
to
not
breaking
things
and
and
and
basically
what
happened
is-
and
this
is
like.
B
F
Are
you
saying
openstack
is
weak
in
the
iscsi
area
or
in
the
definition
of
abstracting
block.
B
How
do
I
in
a
generalized
abstracted
way,
allow
you
to
create,
create
a
volume
and
attach
a
virtual
machine
to
it
and
that
the
apis
to
do
that
are
not
in
such
a
way
that
they
can
be
abstract
and
it
gets
that
that
conversation
specifically
got
really
interesting,
because
also
people
like
okay,
wait
a
second.
If
I
have
a
storage
system
in
which
I
can
create
shadow
copies
of
something
right.
So
if
I
have
a
sand
and
instead
of
making
a
full
copy,
I
just
need
a
little
bit
of
a
copy.
B
My
google
hangout
that
so
I
I
just
I
want
to
make
shadow
copy
well,
if
you
store
a
copy
of
that
server,
and
then
somebody
makes
a
copy,
you
want
to
leverage
that
you
don't
want
to
make
full
copies
of
things
as
you
replicate
them
around,
so
the
people
who
have
that
capability
want
to
leverage
it
use
it
make
it
part
of
it.
So
you
get
into
this
sort
of
interesting.
B
B
B
B
B
I
don't
know-
and
this
I
don't
think
we
are-
and
one
of
the
things
that's
really
significant
here-
is
that
I
spent
a
lot
of
time
doing
vmware
clusters
and
things
like
that
before
for
this,
and
there
is
a
real
desire
to
not
repeat
the
a
server
runs
to
a
sand
and
you
get
a
block
of
servers
running
to
one
san
and
then,
when
you
outgrow
that
you
you
flip
over
right,
the
design
is,
I
have
compute
and
I
have
store
block
store
and
they
need
to
be
connected
by
network.
B
But
if
I
want
to,
if
I
run
out
of
block
storage
space,
I
can
send
up
a
new
block
storage
system,
or
if
I
want
to
offer
high
performance,
block,
storage
and
low
performance,
they
can
sit
next
to
each
other
and
vms
can
go
against
whichever
ones
they
need.
It's
not.
You
know,
because
of
the
way
openstack
needs
to
scale.
You
have
to
be
able
to
expand
out
without
having
that
tight
coupling.
B
I
have
failures
and
anticipated
an
expected
event,
and
so
everything
has
to
be
okay.
We
have
to
be
able
to
expect
that
this
can
fail
and
we
have
to
be
able
to
expect
that
we
can
just
throw
in
another
note
at
any
given
point
in
time
we'll
be
able
to
join
in,
and
then
you
have
to
be
able
to,
as
they
said,
programmatically
address
these
things
that
you
can't
have
something
where
we
have
to
go
and
just
you
know,
pull
in
the
block
devices
assigned
to
this
thing
and
then
fill
all
this
stuff
out.
B
B
This
section
of
it
is
leaning
much
more
towards
the
phd
section,
just
because
it's
still
it's
earlier
in
its
development
stage
than
the
other
pieces,
nova
and
glance,
and
these
kind
of
things
are
fairly
stable,
mature
at
this
point,
so
that
makes
it
easy
to
go
in
and
build
some
of
these
deployment
tools
to
take
some
of
this
pain
out.
But
things
like
the
swift,
with
the
with
the
object
stores
and
now
quantum
these
kind
of
things.
These
are
very
much
in
that
much
more
raw.
B
Rapidly
changing
nobody
wants
to
write
deployment
tools
because
they're
not
necessarily
sure
what
the
final
form
is
going
to
be,
and
then
it
will
have
to
rewrite
it.
Let.
B
F
B
B
Is
the
this
is
this?
Actually,
we
could
do
a
whole
meet
up
on
block
storage
and
actually
it's
just
it's
a
good
future
topic
just
to
talk
about
block
storage,
because
and
actually
what
I'll
do
is.
I
know
greg's
listening
so
greg,
I'm
going
to
ask
you
at
some
point
in
the
future
to
lead
meet
up
on
block
storage,
which
the
code
name
for
black
storage
is
cinder
yeah.
I
believe
that's
official
I'll
look
to
greg
to
give
me
a
yes
or
no,
that
that
was
officially
the
the
name
of
the
project.
B
B
You
know
it's
openstack
is:
is
we're
playing
there's
a
lot
more
big
players
involved
now,
so
there's
real
business,
real
money
involved.
The
thing
that
that
I'm
getting
used
to
expecting
at
the
summit
for
me
is:
I
don't
go
to
a
lot
of
sessions.
B
I
make
it
through
the
hallway
and
I
get
pulled
into
a
conversation
because
as
much
as
the
community
is
doing,
development
work
and
planning
development
around
what's
coming
in
the
next
release,
you
know
our
our
job.
My
job
is
to
help
get
the
products
and
the
ecosystems
and
the
fits
and
the
pieces
and
do
a
lot
of
coordination.
So
I'm
coordinating
with
our
partners
to
figure
out
you're
doing
this,
I'm
doing
that.
If
we
collaborate
together,
we
can
accomplish
more.
B
B
The
most
collaborative
challenging
yeah
just
constantly
frenetic
pace
event
that
I
go
to
at
all.
Similarly,.
B
Struck
me
was
that
it
was
announced
that
it
was.
A
B
Of
that
effort
was
even
quite
impressive
and
that's,
I
think
one
of
the
challenges
for
meetup
groups
like
us
is
to
turn
us
into
working
groups,
because
it's
going
to
be
impossible
for
everybody
to
be
represented
in
those
conferences.
So
I
think
one
of
the
evolutions
I
see
for
the
meetup
group
and
we
have
to
help
new
people
get
used
to
openstack,
but
also
keep
the
people
who
are
really
contributing
and
collaborating
already
at
the
expert
level.
Right
like
what
dan's
been
talking
about
and
the
types
of
things
we're
doing.
D
B
In
the
middle
of
this,
we
need
to
make
sure
that
we
we
create
these
nexuses
of
collaboration
within
our
regions,
so
that
when
we
so
that
our
represented
this
was
a
theme
before
that
before
went
to
the
summit,
that
our
views
are
represented
at
the
summit
that
people
people
have
had
a
chance
to
talk
things
through
even
before
we
reach
that
the
good
conversations
at
the
summit
were
ones
where
people
have
already
gotten
a
certain
way
down
the
field
right.
It's
a
very
careful
balance.
B
If
you
come
to
prepared
and
you're
just
showing
oh
yeah,
I
did
this
code,
it's
ready
to
be
integrated
into
openstack.
It's
not
that
helpful.
But
if
you
show
up
saying
I
have
this
great
idea
that
I
haven't
done
any
work
on
it.
It's
also
not
very
helpful,
so
we
as
a
community
have
to
start
bringing
bringing
things
to
a
level
of
maturity
that
they
can
really
be
taken
up
at
summit.
B
All
right,
let
me
do
that,
so
I'm
going
to
I'll
switch
that
and
we'll
probably
close
on
this
video,
which
I
think
is
a
good,
a
good
wrap
the
conference,
the
close
yeah.
This
was
the
last
session,
so
you
know
I
would
encourage.
I
would
encourage
people
read
some
of
the
materials
about
the
summit
come
back
in
and
you
know
we
can
dedicate
some
time
to
q
a
you
know,
we're
obviously
very
freeform
on
that.
So
please
come
back
and
and
think
about
that
you're
welcome
to
post
links
into
the
meetup
group.
B
If
you
find
things
that
summarize
the
conference
really
well,
I
know
that.
Why
am
I
blanking
on
the
same?
The
community
manager?
Excuse
me:
that's.
B
B
Yes,
yes,
that
posted
a
list
of
links
that
had
people
summaries
about
the
openstack
summit.
That
was
some
good
listing
all
right.
So
with
that
I'm
gonna
flip
the
camera
and
we'll
show
the
show
the
video
and
then
we'll
close
so
take
pizza.
And
if,
if
your
company
wants
to
sponsor
ping
me-
and
I
will-
I
will
hook
you
up
with
the
right
people-
t-shirts
as
well-
oh
yeah,
you've
got
way
too
many
ships
so.
D
D
D
D
A
C
B
C
Awesome,
that's
all
I
needed
all
right
great.
Thank
you
very
much.
So
who
are
you
and
what
do
you
do?
I
am
large
lucky,
I'm
a
common
venture
partners.
We
do
early
stage
software
investment
and
you
were
just
on
a
really
interesting
panel
here
at
the
openstack
conference.
Talking
about
the
investment
prospects
for
companies
involved
in
openstack,
would
you
say
those
prospects
are
pretty
good?
I
think
it's
a
it's
a
great
time.
It's
a
great
community
we're
seeing
an
enormous
amount
of
interest
in
the
openstack.
C
I
think
you
know
the
classic
we're
inside
the
tornado
of
a
pretty
interesting
technology
trend
right
now,
I'm
an
investor
in
piston
cloud.
So
I'm
a
believer
in
openstack
and
how
that's
going
to
change
the
future
of
computing
excellent.
Now
one
of
the
other
people
said
that
he
wouldn't
bet
against
openstack
at
this
point,
would
you
agree
I
I
would
agree.
C
I
think
it's
it's
the
community
and
the
market
force
behind
this
is
is
a
wave
that
is
more
powerful
than
sort
of
the
market
forces
of
any
individual
company
to
try
to
stop
what's
happening
with
openstack
awesome.
Well,
thank
you
very
much
for
your
time,
sir,
all
right
all
right,
sir,
who
are
you
and
what
do
you
do.