►
From YouTube: CNCF Storage Working Group - 2018-04-25
Description
Join us for KubeCon + CloudNativeCon in Barcelona May 20 - 23, Shanghai June 24 - 26, and San Diego November 18 - 21! Learn more at https://kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy and all of the other CNCF-hosted projects.
A
A
A
A
A
A
A
A
B
B
B
D
B
B
All
right
looks
like
we've
got
about
14
15
people
like
weekly
have
about
20,
but
let's
kick
it
off
this
morning,
I've
I
think
we've
got
a
pretty
light
agenda.
The
the
plan
for
today
was
a
kind
of
one
key
thing
which
was
just
reviewing
with
the
group,
the
slides
that
we
put
together
and
the
plan
for
our
cute
Connie
you
next
week,
and
we
do
have
open
space
here
today
for
other
topics
that
folks
wanted
to
cover,
and
we
didn't
get
a
presenter
to
talk
about
a
project
for
this
week.
B
But
if
you
wanted
to
chat
about
something
feel
free
to
go
to
the
agenda
and
add
your
topics
there,
you
can
fill
them
the
end
of
the
time
or
we
can
cut
it
short
today.
The
the
first
thing
that
we
need
that
I've
thrown
the
agenda
was
wanting
to
get
Quentin
to
to
kind
of
speak
a
little
bit
about
his
new
role
as
a
TOC
member
at
the
CAF
and
talk
a
little
bit
about
you
know
what
he
wants
to
do
here
within
the
swg.
E
Thank
you
yeah,
so
some
of
you
may
or
may
not
be
aware.
I
was
asked
to
join
the
TRC
a
couple
of
weeks
ago
now,
which
I
accepted,
and
one
of
the
things
they
have
asked
me
to
do
is
help
you
guys
get
some
clarity
on
what
the
TRC
would
like
out
of
this
group.
I
understand
that
historically,
that's
been
little
challenging.
It's
been
not
entirely
clear
what
the
TOC
wanted
you
guys
to
do,
and
I'm
very
new
into
all
of
this
stuff.
E
So
please
do
feel
free
to
disagree
or
correct
me
along
the
way,
if
any
of
the
repressions
I've
gathered
are
not
accurate
or
different
from
with
the
impressions
in
this
group.
So
yeah,
my
thinking-
and
this
is
still
pretty
informative-
is
to
just
put
a
proposal
together
as
to
what
I
think
would
be
both
useful
and
I
think
ideally
relatively
unconscious,
initially
just
to
get
a
get.
Some
groundwork
laid
that
we
can
then
build
on
top
of
and
I
understand.
E
Cool,
ok,
be
we
have
any
questions
or
comments
for
Quentin
I
actually
have
a
request
are
not
that
familiar
with
the
people
in
this
group.
I
know
some
of
you,
but
not
all
of
you.
Would
it
be
too
much
to
ask
for
to
do
a
quick
round
table
and
have
a
quick
intro
of
the
sort
of
active
members
of
the
group
yeah.
G
I
H
J
L
D
You
deep
I
think
unit
and
then
needed
it's
okay.
If
you
don't
want
to
introduce
yourself
all
good
yeah.
D
M
N
C
A
Yes,
I'm
tipo,
Lee,
I
I
work
with
a
few
people
on
the
I
scuzzy
Tanya
library,
in
the
cloud
space
and
we
are
working
to
add
C
SCSI
to
the
front
end
and
we
wrapping
up
the
back
end
with
a
few
back
pocket
storages
and
in
progress
is
we
are
doing
the
gateway
portal,
the
protocol
gateway
for
ice
Kathy,
and
we
have.
We
are
interested
in
seeing
what
this
group
is
to
us.
Thank
you
and
what
company
are
you
with
people?
Oh,
we
are
not
associated
with
any
company.
P
Q
Q
Q
Q
E
O
B
B
B
Q
E
Bus
I
know
well,
is
Mike
Ruben
involved
here
much
now.
B
No
okay,
I
think
that
that
probably
goes
you
know.
Quentin's
intro
goes
pretty
well
into
the
next
thing
that
we
had
on
the
agenda,
which
is
talking
about
the
advanced
session
at
cube
con.
What
we,
what
we
said
talked
about
on
this
call,
was
trying
to
get
a
someone
from
the
TOC
to
come
in
chapter
with
us.
B
We
could
have
a
nice
ground
table
discussing
you
know
the
future
of
the
STB
key
and
you'll
see
wanted
from
us
think
wit
has
already
gotten
back
some
of
that
information
because
he's
been
working
with
them
to
you
know,
agree
on
that,
but
we
are
gonna,
have
Alexis
Richardson
who's
gonna
come
join
us
at
the
conference
during
our
advanced
session.
So
we'll
have
you
know
not
only
Quinton
as
a
TOC
member,
but
also
Alexis
who's
gonna,
be
for
a
long
time
giving
us
just
some
more
guidance
on
what
we're
we're.
B
E
Some
writing
here,
which
is
a
proposal
that
I've
sort
of
passed
it
passed.
Some
of
you
I,
think
I.
Guess
some
of
the
previous
meetings
I've
been
to
have
not
included
everyone.
I,
don't
know
if
those
were
like
a
splinter
group
specifically
for
the
presentation
or
whether
I
missed
something
but
yeah
I
flirted
the
idea
of
what
I
think
a
useful
first
step
would
be
and
it
seemed
to
be
reasonably
well
received,
so
I've
actually
passed
that
by
the
TRC
and
that's
being
well
received
there
too.
E
E
B
E
E
Think
some
of
them
spoke
to
some
of
you
guys
as
well
and
got
some
feedback,
and
so
this
is
kind
of
an
attempt
to
distill
all
of
that
wisdom
into
a
what.
If
I
hope,
is
a
Eva
Belov
goals
that
will
be
useful
to
the
group
and
to
the
CNN
CF
in
general
and
which
would
be
considered.
You
know
an
interesting
thing.
Would
you
use
Fri
you
guys,
which
is
you
know
equally
important?
E
So
just
to
be
clear
about
the
non
goals,
so
these
are
not
non
goals
forever,
they're
just
non
goals
for
what
I
would
term
we
can
call
this
phase,
one
so
phase
one.
We
tackled
our
goals
and
we
explicitly
don't
tackle
the
non
goals
and
then,
at
the
end
of
phase
one,
we
decide
what
the
goals
for
next
for
the
next
phase
are,
and
maybe
those
include
what
are
currently
labeled
as
non
goals.
E
So
maybe
actually
we
should
make
that
explicit
and
say
non
goals
for
phase
one
just
so
that
nobody
gets
confused
because
ultimately,
I
do
think.
We
need
to
come
up
with
some
kind
of
idea
that
that,
in
a
cloud
native
environment,
you
know
these
things,
work
well
or
had
more
more
pros
than
cons,
and
these
other
things
tend
not
to
work
well
and
I.
E
Think
most
of
us
have
intuition
as
to
what
those
groups
look
like,
but
until
I
think
we
have
a
common
understanding
of
what's
actually
done
in
in
cloud
native
environments
at
the
moment
and
what
all
the
terminology
means
and
how
it
relates
to
each
other
I
think
it's
difficult
to
have
the
discussions
that
are
required
in
the
non
goals
section
and
my
take
on
some
of
the
stumbling
blocks.
Up
to
now
have
been
that
we
may
be
tackled
what
are
listed
as
non
goals
before
we
had
a
common
understanding
of
the
stuff
listed
under
goals.
E
E
E
Obviously
you
know
one
has
to
limit
it.
One
can't
go
into
excruciating
detail
on
all
these
areas,
upfront
because
you
end
up
with
a
500
page
document
that
nobody
will
read,
but
what
I
have
in
mind
and
I?
Don't
know
if
any
of
you
have
read
the
service
working
groups,
white
paper
that
they
published
a
few
months
ago
and
they
they
took
on
a
similar
goal
within
serverless,
and
that
was
quite
successful,
so
I
would
actually
recommend
I
can
try
and
pull
out
the
link.
Now.
E
B
E
B
And
on
the
you
know,
the
depth,
but
I
think
that
you
know
kind
of
defining.
This
is
trying
to
go
wide
to
be
comprehensive
for
storage
and
I.
Think
I
agree
with
you
there.
What
I'm
concerned
with
is
the
depth
like
if
we
create
a
white
paper?
Well,
if
we
look
at
the
surrealist
white
paper,
a
lot
of
that
territory
was
was
new,
I
feel,
and
so
it
made
for
interesting
reading
along
the
way,
because
you're
not
covering
things
that
have
been.
B
You
know,
solidified
over
less
yeah,
30
years,
attack
yeah
and
in
this
case
we're
talking
about
storage,
like
the
fundamentals
of
it.
I
think
that,
whether
it's
block
stores,
key
values,
databases
like
a
lot
of
that
assessment
covered
in
a
bunch
of
different
places,
and
so
what
I
would
I'd
be
worried
about
is
is
creating
that
before
that
much
of
was
already
out
there
or
what
you
know
tries
to
clarify
too
much
of
what's
already
on
paper
in
other
places
like.
B
E
There's
you
know
various
different
permutations
and
some
of
those
are
relatively
recent
and
then
they're
also,
you
know
the
different
variants
that
exist
out
there
today
are
sometimes
built
in
different
ways.
Some
things
are
built
on
top
of
local
block
stores.
Some
things
are
built
on
top
of
distributed
block
stores.
Some
things
are
built
on
top
of
distributed
file
systems,
and
they
things
like
fuse.
E
I
mean
there
are
a
lot
of
way
places
where
the
sort
of
classic
layering
of
a
storage
system
is
not
adhere
to
anymore,
for
sometimes
good
reasons,
so
I
think
just
clarifying
all
of
that
stuff,
and
maybe
I,
didn't
make
it
clear
here
that
that
I
think
there's
quite
a
lot
in
each
one
of
those
areas
in
terms
of
as
I
say,
distributed
versus
non
distributed,
enterprise
versus
commodity
hardware.
You
know
all
of
these
different
dimensions
where
it
would
be
worth
at
least
having
a
paragraph
or
two
on
each
of
these
saying
this
is
a.
E
This
is
a
thing
it.
It
has
these.
This
is
why
people
do
it,
because
it
has
these
strengths.
It
has
these
drawbacks.
This
is
you
know
how
people
mitigate
some
of
those
drawbacks,
etc
and
just
go
through
the
list
like
that,
just
to
explain
why
all
these
things
exist
and
and
and
how
they
relate
to
each
other
yeah.
B
E
Know
I'm
sure
there'll
be
some
debate
and
that's
why
I
proposed
that
we
focus
on
on
stuff
that
actually
exists
and
stuff
that
is
in
use
and
and
if
there's
any
kind
of
debate
as
to
how
this
should
be
or
whether
this
is
better
or
not,
it
boils
down
to
well,
you
know
who's
doing
this
like
show
us
a
case
where
this
is
done,
and
if
there
is
a
case
where
this
is
done,
then
let's
say
this:
this
is
done
in
this
case
and
ideally
give
you
know
some
reasons
why?
Whoever
did
it
that
way?
C
I
think,
though,
you
comes
out
of
perhaps
standardizing
interfaces
to
these
classes
of
storage
or
stateful
services,
and
once
you
go
get
into
implementation
details,
those
could
be
endless
and
amount
to
vendor
sales,
pitches
and
things,
and
it's
potentially
really
treacherous
to
do
that.
Deep
dive
into
these
I
mean
I,
wouldn't
be
against
having
occasional
sessions
and
presentations
for
those
interested
but
kind
of
cross
industry
discussions
going
into
deep
dives
of
particular
implementations.
Just
sounds
like
you
know.
My
mine
can
beat
up
yours
or
something
and
I
don't
know
how
yeah.
E
E
Want
to
avoid
that
I
agree
and
just
just
to
kind
of
clarify
in
everyone's
minds
what
I
have
here.
Maybe
we
should
put
it
explicitly
so
that
the
server
list
document
I
believe
is
in
the
region
of
30
pages
long,
and
so
you
know
I
can
imagine.
If
you
take.
Let's
say
there
are
sort
of
10
key
concepts
across
all
of
these
things.
E
We're
talking
about
a
couple
of
pages
about
blocks
tours
and
a
couple
of
pages
about
key
value
stores
and
I
and
and
it's
by
no
means
gonna,
be
exhaustive
and
it's
by
no
means
going
to
be
a
deep
dive
into
all
the
possible
implementations
of
these
things.
But
it
is
hopefully
gonna
clarify
the
terminology
and
clarify
you
know
to
whatever
extent
one
can
do
in
that
amount
of
space.
How
these
things
tied
together,
yeah.
C
E
E
F
Sort
of
things
are
playing
and
the
actual
things
apart,
you
know.
Is
it
it's
four
devices
if
they're
not
chip
store,
is
it
accessible
over
an
API?
It's
a
participant
assistant
interface?
Is
it
a
local
class
system?
This
is
the
shared
resistance
for
the
distributed
file,
system
or
sort
of
pages,
and
then
there
are
the
the
control
line,
type
stuff
and
that's
going
to
become
2d
interfaces
and
how
we
get
integration
and
your
constraints
and
things
like
CSI
and
of
sorts
of
things,
and
we
may.
F
We
may
also
want
to
discuss
about
how
it's,
how
the
different
solutions
or
how
the
different
options
are
provided
so
that
perhaps
you
know
differentiating
between
you
know
software
options
and
and
harder
options
and
cloud
options
and
versus
on-premise
options,
for
example,
also
whether
the
solution
can
be
managed
by
an
orchestration,
you
know
didn't
work
more
and
more.
There
are
lots
of
solutions.
F
You
know
like
rogue,
for
example,
where
it's
a
software
solution
and
it's
being
deployed
unmanaged
body
nor
no
straighter
and
that
that's
what
in
pointing
out
as
well
but
I
think
if
we
cover
those
those
those
areas.
Maybe
maybe
we
need
to
kind
of
split
up
those
up
it's
and
and
and
gets
a
couple
of
people
to
preserve
unions,
so
actions
of
something
yeah.
B
I
feel,
like
the
mean
you
just
went
through
a
lot
of
stuff
and
I,
think
you
know.
Some
of
that
is
what
we've
been
talking
about
along
the
way,
and
you
know
I
feel
like
one
of
the
most
important
things
that
we
need
to
do
when
we
start
is
that
we
narrow
down
the
scope
and
we
figure
out
like
who's
the
actual
audience
and
how
deep
he's
gonna
go
because
from
a
cloud
native
perspective,
if
I'm
just
looking
at
the
public
clouds
right,
they
provided
all
these
different
data
services.
B
Yet
I
have
no
idea
how
they're
operated
and
maybe
I
shouldn't
care.
I
have
no
idea.
You
know
they
tell
me
what
the
availability
is,
but
I
really
don't
know
the
details
of
how
that
happens
behind
the
scenes-
and
maybe
that's
just
enough
information
now
from
that
consumers
perspective
they've
got
a
service
model,
is.
B
And
with
some
availability-
and
it
provides
this
certain
functionality
to
support
its
application
and
that
may
be
the
level
of
information
they
care
about.
You
know
if
you're
lower
down
on
the
stack
and
you're
thinking
about
running
these
data
services,
yourselves
in
your
own
cloud,
then
you
may
have
different
concerns
about
you
know
what
to
consider
and
how
to
actually
build
and
operate
these
things.
So
it's
like
just
different
audiences
and
I've
been
capturing
both
and
one
white
characters.
B
F
If
we,
if
we
stick
to
to
sort
of
three
main
categories
of
information,
then
one
is
the
control
plane.
How
are
you
going
to
request
an
interface,
and
you
know,
orchestrates
the
storage
and
then
the
second
point
being
the
data
plane,
which
is
how
you
access
it
any
gasifier
block
and
file
and
object
and
various
other
things
and
databases
and
key
value
stores
and
whatever
else,
and
then
the
third
bit
is
you
know
where
it's
available
so
is
it?
Is
it
on-premises
in
in
the
clouds?
This
is
a
cloud
or
only
offering
is
it
software?
F
Is
it
hard
or
whatever?
And
if
we
stick
to
those
those
are
probably
less
contentious,
and
then
you
know
if
we
want
to
go
any
further,
we
can
talk
about.
We
can
talk
about,
affords,
consider
ear,
just
or
theaters
and
stuff
like
that.
You
know,
have
the
half
the
pros
and
cons
of
centralized
versus
decentralized
versus
distributed
and
whatever
else,
and
you
know,
talk
about
latency
and
access
models
and
things
like
that,
but
that
could
easily
be
like
a
face
to
day.
E
That
sounds
like
a
like
a
very
good
starting
point
to
me
and
I
mean
I
can
imagine.
We
may
want
to
add
to
that.
So
I
mentioned
in
one
of
the
last
item
on
the
goals
there
is
to
compare
and
contrast
to
save,
to
take
a
fundamental
set
of
properties,
of
storage,
availability,
scalability,
consistency,
durability,
performance
and
without
referring
to
you
know,
specific
implementations,
but
certain
certain
approaches
are
fundamentally,
you
know
higher
performance
and
lower
availability,
for
example,
or
less
scalable,
but
more
consistent
or
whatever
the
case
may
be.
E
That
I
think
that
could
be
a
subsequent
step.
I
think
what
you
sketched
out
as
a
first
bunch
of
steps
is,
is
very
good
one
and
we
may
call
it
a
day
at
that
point
and
say
we
reduced
the
scope
of
phase
1
and
we
produced
something
useful
and
we
left
out.
You
know,
goals,
3
and
4
or
whatever
the
case
may
be,
and
we're
going
to
tackle
those
as
phase
2.
F
B
E
Open
to
suggestions
I
mean
in
my
mind,
I've
got
that
this
is
maybe
like
a
three
month
exercise,
but
you
know,
if
you
guys
think
differently.
Let
me
know,
and
what
I
have
in
my
mind
is,
is
maybe
a
small
number
of
well
we'll
just
call
them
primary
authors,
the
let's
say
to
next
to
throttle
to
make
sure
that
this
document
gets
done
in
three
months
and
they
would
be
kind
of
responsible
for
putting
the
framework
together.
E
So
let's
decide
what
the
first
step
is
in
the
next
three
weeks,
we're
going
to
try
and
define
you
know
the
following
terms
and
we,
inter
put
them
in
the
document
and
we're
going
to
review
them,
and
you
know,
come
to
conclusion
and
then
we're
going
to
you
know,
carry
on
from
there
and
then
as
many
and
and
maybe
maybe
a
step
in
the
process.
This
is
to
break
it
up
once
we've,
you
know
taken
a
first
pass
at
what
the
structure
of
the
information
looks
like
we've
mapped
carpet
up
and
say
right.
E
This
block
store
expert,
goes
and
writes
two
pages
on
block
stores
and
explains
to
us.
You
know
pros
cons.
Whatever,
whatever
we
decide
this
important
there
and
they
can
fill
in
all
the
meet
there
and
then
bring
it
back
to
the
to
the
main
duct
for
inclusion
and
review,
because
that
does
that
sound
like
a
reasonable
approach,
mm-hmm
and
there's
sort
of
three
months
to
produce
a
30
page
document,
I
mean
it
might
be
a
little
optimistic
given
people's
time
availability
and
cute
cut
in
between.
B
E
B
How
about
a
comment
from
anyone
else
in
the
call
you
know
how
we
use
the
the
SM
Aug
top
time
in
the
next
few
months,
because
we've
been
doing
presentations
from
the
community
about
different
storage,
related
projects
and
I've
I've
enjoyed
that,
but
if
others
want
to
divert
at
the
time
and
just
make
it
all
about
working
on
the
paper,
that's
also
an
option
to
how
does
anyone?
How
does
everyone
feel
about
you
know
the
time
spent
in
swg
the
next
few
months,
I.
O
G
It
yes
from
the
community,
especially
if
you're,
covering
a
broad
spectrum
of
data,
store
solutions
assuming
the
main
purpose
of
the
white
representation.
You
know,
I
think
we
can
really
talk
about
technologies
in
abstract
way,
not
referring
to
specific
solutions.
I
think
what
the
endless
last
time
was
I've
got
quite
contentious
was
basically
turning
the
white
paper
into
a
cloud
native
landscape,
where
we
have
logos
and
four
different
vendors
and
solutions
and
try
to
say
what
one
is
better
than
the
other
I
think.
G
We
should
probably
avoid
that
as
much
as
possible
focus
mostly
on
technology
and
even
there
it
can
be
quite
contentious
with
somebody
if
you
prescribe,
let's
say
if
you
have
to
use
a
distributed
fastest,
any
sort
of
key
value
store
or
some
other
solution,
or
no
sequels
to
ask
for
where's
the
sequel
or
you
know.
So
we
have
to
be
quite
careful
about
how
we
navigate
this
document.
I.
F
Agree,
but
you
know
I
think
what
Quinton
clarified
for
us
is
that
we're
not
about
reducing
scope
or
saying
what
is
or
what
isn't
I
painted
were
we're,
describing
you
know
the
terminology
and
two
different
categories
of
interfaces
and
the
different
theories,
and
you
know
accessing
it,
even
a
sort
of
thing
and
I
think
that
it
should
be
unpretentious
because
it
could
is
a
list
of
stuff
right.
So
so
it's
not
it's
not.
This
is
better
than
that.
The
other
thing
yeah.
G
E
I
just
want
to
had
a
quick
question,
so
have
you
guys
been
meeting
once
a
month
or
more
frequently
the
mess
up
to
now
by
weekly,
okay,
so
every
two
weeks,
so
I
guess
one
one
approach
might
be
to
have
a
monthly
presentation
and
the
intermediate
bi-weekly
meetings
being
work
on
this
document.
Just
a
thought,
mm-hmm
or
you
could
split
each
meeting
in
half
and
a
half.
You
know
one
presentation
and
some
work
on
the
doc.
E
What
I
would
suggest
just
based
on
previous
experience,
is
that
the
actual
work
on
the
dark
have
happened
not
in
those
scheduled
meetings,
because
otherwise
you
probably
won't
get
anywhere,
but
you
know
people
need
to
go
out
and-
and
this
is
sort
of
these
primary
authors
that
I
mentioned
there's
this
background
work.
You
know,
I
think
it's
usually
better
to
write
these
kind
of
documents
in
a
quiet
place
when
you
can
think
clearly
and
put
some
thoughts
on
paper,
and
then
you
know
bring
that
back
to
the
group
for
review.
J
E
Know
I
agree,
one
approach
might
be
to
for
anyone
who's
interested
in
being
the
primary
author
to
go
off
and
produce
a
kind
of
wireframe
of
what
they
think.
The
outline
of
the
document
might
look
like
and
perhaps
have
more
than
two
weeks
time,
and
then
we
can
look
at
those
things
and
you
know.
Maybe
this
is
just
a
brainstorm.
E
E
Exactly
and-
and
you
know-
hopefully,
we
won't
be
terribly
contentious
and
I'm
sure
some
of
them
will
look
similar,
assuming
that
people
have
kind
of
digested.
What
I
put
them
writing
here
and
hopefully
I
haven't
left
out
too
much.
But
you
know
ultimately
something
in
the
region
of
a
30.
Page
document
have
a
look
at
the
service
white
paper.
E
If
you
want
to
get
a
flavor
of
the
level
of
detail
and
the
kinds
of
headings
and
things
and
then
with
a
bit
of
luck,
we'll
end
up
with
a
few
of
those
that
look
not
too
different
from
each
other,
I
mean
if
they
do
look
very
different
than
we
have
two
camps
of
people
who
line
up
behind
very
different
documents.
We
will
gonna
have
to
result
that,
but
hopefully
I
would
assume
that
is
unlikely
to
happen.
B
Great
stuff,
okay,
so
let's
move
on
to
the
next
thing
on
the
agenda
here,
I
put
links
into
the
the
agenda
under
the
cute
Connie
you
under
the
intro
session,
so
we
did
have
that
subgroup
that
formed
and
thank
you
for
Alex
for
actually
putting
together
this
slide.
So
we
now
have
there's
what
the
first
link
is.
The
meeting
notes
where
we
just
captured
some
the
ideas
and
then
the
second
is
the
actual
slides
that
were
be
using
for
the
the
coupon
intercession
today.
B
B
F
The
main
thing
that
needs
to
be
decided
is
who's
actually
and
who's
actually
gonna,
be
there.
Unfortunately,
something's
come
up
and
I'm,
not
gonna,
actually
be
able
to
travel
next
week.
So
I'm
going
to
be
into
two.
She
presents
this
or
be
involved,
unfortunately,
but
we
need
some
volunteers
to
actually
be
able
to
go
through
this.
Okay,
I.
Q
C
E
B
Alright,
so
maybe
the
so
it
sounds
like
we
got.
You
know
three
to
four.
You
know,
folks
that
can
play
a
part
in
this
and
you
guys
want
to
sync
up
later
this
week
separately
or
do
we
want
to
spend
the
next
10
minutes
and
just
kind
of
nail
down?
You
know
what
pieces
of
this
presentation
that
that
we
can
do
or
what
you
guys
want
to
do
here.
B
B
B
Know
that
so
Ardell
on
inside
you
guys
were
both
on
the
the
planning
policy,
but
I.
Don't
think
that
you
were.
B
O
B
Okay,
so
we
go
from
a
section
of
hey
what
is
for
the
storage
what's
working
group
and
then
we
go
straight
to
projects
and
architecture
and
then
there's
patterns
Ardell
on
that
it
sounds
like
side,
maybe
like
slides.
What
like
three
to
two
six
are
kind
of
grouped
together:
yeah
that
sound
right,
yeah.
B
O
So
actually
I'm
thinking
does
it
make
sense
to
have
multiple
presenters
or
do
just
one
presenter
and
then
have
a
panel
for
QA
I
find
that
when
you
try
to
break
it
up
into
multiple
presenters,
it
kind
of
just
gets
a
little
messy
and
especially
if
you're,
switching
back
and
forth
I'd
be
happy
letting
you
do.
The
president,
the
full
presentation,
if
you're,
ok
with
a
Clint
saying
here
or
I,
can
volunteer,
don't.
E
B
F
O
C
I
B
E
Q
Q
E
B
B
All
right
sounds
good
anything
else,
any
other
comments
about
the
deck
nope,
all
right,
I,
think
in
terms
of
the
agenda.
I,
think
that
we're
we're
done
any
other
questions
that
people
wanted
to
throw
in
here
last
minute,
nope
all
right!
Well,
thank
you.
Everyone
for
joining
today,
looking
forward
to
seeing
everybody
out
of
it
at
a
cute
con,
and
you
got
about
seven
minutes
back
in
your
day.
Okay
might.