►
From YouTube: CNCF Storage WG Meeting - 2019-03-27
Description
Join us for Kubernetes Forums Seoul, Sydney, Bengaluru and Delhi - learn more at kubecon.io
Don't miss KubeCon + CloudNativeCon 2020 events in Amsterdam March 30 - April 2, Shanghai July 28-30 and Boston November 17-20! Learn more at kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects
A
C
Sure
hi
Alex
hi
everybody-
this
is
Evan
and
yeah.
Really.
We
will
of
course,
focus
on
the
demo
and
the
how
right
what
is
open
EBS
and
how
does
it
work
just
quickly,
I
think
here,
ins
sharing
the
slides,
but
the
company
is
called
Maya
data
and
has
been
around
for
a
number
of
years
now,
open
EBS
was
started
in
late
2016
and
really
our
our
mission
is
something
we
call
data
agility
and
we
can
talk
a
little
bit
more
about
that
and
happy
to
answer
any.
C
You
know
questions
about
the
business,
but
it,
but
it's
an
open
source
project.
It's
a
little
more
open.
Abs
is
a
little
more
than
two
years
old.
Now
and
now
it
does
have
a
bunch
of
users.
We
are
by
far
the
largest
contributor,
although
there
are
we're
about
70
people
to
give
you
an
idea
and
there's
350
total
contributors.
Many
of
those
are,
you
know,
small
contributors,
so
one
reason
we
would
want
to
contribute
the
project
potentially,
of
course,
is
make
it
more
welcoming
to
you
know
to
other
other
companies
and
other
contributors,
yeah
Karen.
A
D
D
D
A
D
D
D
Since
then,
a
lot
of
things
have
happened
in
terms
of
stabilizing
the
product,
and
then
we
do
have
a
lot
of
users
using
it
in
production.
Right
now,
in
the
United
States
we
were
taking.
You
know
it's
actually
some
additional
feedback
from
this
group
and
see
what
the
group
has
to
say
about
pushing
it
into
CNC
of
sandbox,
all
right
emulation
of
how
it
works,
and
then
we
will
dive
into
components
or
the
architecture
of
opening.
Yes
right,
so
just
convert
it
into
a
slight
mode.
D
Typically,
you
start
off
with
the
kubernetes
cluster.
There
are
some
how
this
running
in
the
cloud
some
kind
of
attach
to
end.
So
the
question
stated
typically
comes
in
and
says:
okay,
let
me
just
set
up
open
areas
which
installs
some
control
components
as
I
call
them.
These
are
clustered
about
components,
and
example.
No
disk
manager
is
one
of
the
component
that
actually
runs
as
a
diamond
set
on
all
the
nodes
which
can
help
discover
the
devices
attached
to
the
node.
It
goes
ahead
and
actually
creates
resources
for
the
radius
resources.
D
There
are
multiple
data
engines
that
we
support,
so
see
story
is
one
of
them,
so
can
run
by
making
use
of
one
or
more
disks
that
are
attached
to
the
node
and
they
and
if
it
created
them
all
the
nodes.
This
and
abused
is
a
shade
edge
storage
model.
Once
you
do
that,
like
any
other
storage
provider,
we
have
to
create
a
storage
class
that
says
power
into
one.
D
Different
storage
pools-
and
this
is
totally
controlled
by
some
policies
that
you
can
set
for
a
storage
class
in
terms
for
me,
because
we
want
and
cease
to
target
is
the
one
that
actually
does
a
synchronous
replication
to
the
differential
to
us
at
once.
The
replicas
are
rushed
to
the
target.
The
provisional
goes
ahead
and
creates
a
3d
object
and
uses
the
entry,
as
does
he
want
him
right
now
to
attach.
D
D
D
D
D
D
D
He's
actually
getting
down
in
the
bandage
one
of
the
reasons.
Why
do
we
didn't
go
ahead
with
that
is
beyond
completely
kubernetes
Native
communities
provided
in
the
external
storage,
provisional,
we're
good
enough
and
was
pretty
awkward
compatible
with
all
the
versions
we
do
have
customers
that
are
running
still
in
questions
so
definitely
and
just
to
make
sure
we
are
those
customers.
D
A
D
D
D
Store
actually
uses
this,
it
puts
in
a
request
for
devices
and
that
will
be
taken
up
by
the
indium
operator
to
look
at
what
are
the
nodes
devices
are
available
and
then
associate
them
with
the
device
request
and
see.
Store
operator
basically
uses
these
disks
right
so
as
C
stored
in
the
same
open,
BS
cluster.
This
kind
of
helps
to
boundaries.
D
D
D
D
B
A
D
D
All
right,
so
this
one
just
gives
the
same
sorry
I,
think
I
says:
control
brain
okay.
The
last
piece
was
this
operator
that
which
is
talked
about
that
staring
at
it
and
a
custom
resource
for
open
base
is
available.
That
will
help
with
what
data
engines
to
enable
or
ability
to
actually
doing
get
open
views
to
check
the
status
of
the
various
components
that
kind
of
stuff.
D
Engine
box,
so
you
really
start
with
the
storage
node
that
has
four
spots
and
is
launched.
It
starts
a
TiVo
bundle
topic,
which
is
the
a
service
which
has
a
sidecar
for
my
matrix
exporter
and
then,
depending
on
the
replicas.
The
jigglypuff
thoughts
will
be
not
lost
on
different
modes
which
make
use
of
that
host
path
right
and
the
stateful
obligations
bought
itself
will
make
use
of
the
filter
it
and
the
CSH
entice
the
signal
shelter,
one
of
them,
depending
on
what
connect
that
we
are
using
to
connect
to
the
service.
D
I,
see
tablets
a
wrist
and
provide
the
data
right.
So
one
of
the
reasons
for
using
the
service
is
the
model
itself
is
stateless,
so
in
case
a
known
face
that
can
actually
get
scheduled
on
to
some
other
node,
and
the
cubelet
tower,
like
the
initiator,
will
be
only
talking
to
the
service
ib
instead
of
the
actual
target
IP.
D
A
Kiran,
maybe
this
will
become
a
bit
clearer
on
the
in
the
demo
but
I'm
sort
of
struggling
a
little
bit
to
understand
which
bits
are
kind
of
like
internal
components
and
software
in
which
bits
are
sort
of
logical
components.
I
mean
when
we
say
the
c-store
state
engine.
Are
we
talking
about
container
per
nodes,
a
container
per
pool
a
container
per
volume
or
handle
hands
that
look?
Okay,.
D
D
D
D
D
D
D
D
D
D
So
the
next
one
is
the
school
system,
booth
management.
This
is
the
sidecar
that
actually
catches
the
details
from
the
CSP
custom
resource.
Other
storage
configuration
typically
in
this
case
there
was
only
one
disk,
but
it
to
lists
that
you
want
a
middle
configuration
on
the
disks
itself.
Do
you
want
to
enable
some
other
storage
level
parameters?
Those
things
will
be
available
in
the
CSV
object
and
that'll
be
the
end,
and
then
the
pool
will
be
computed
along
with.
There
is
another
side
context
for
matrix.
D
D
D
So
we
had
a
bunch
of
controlling
components
like
the
provisioner
and
the
storage
API
server,
and
then
we
saw
that
there
are
three
ports
that
God
created
for
each
of
the
pools
which
had
data
container
and
the
management
container
I'm
watching
this
one.
So
this
one
creates
a
PVC
called
back
on
offense,
basically
using
a
sisters,
pass
phone
and
your
warning
as
we
created
PVC
c901.
So
for
this
PVC
and
news
East
on
top
and
when
you
get
created
or
then
three
beers
look
at
this
one.
D
A
A
D
A
A
D
D
D
Good,
so
we
have
all
these
system
tool
pots
that
are
already
running
when
the
new
PVC
is
created.
We
create
a
system
bonding
target
pod,
that's
basically
getting
the
I
will
package
from
the
especi
from
the
straight
full
application
and
for
each
of
the
package
that
it
receives
it.
Basically,
synchronously
replicates
to
the
different
pods
and
as
part
of
synchronous
replication.
It
also
actually
happens
a
little
bit
of
a
header
that
the
sister
bonding
target
understands
simplicity
sake.
A
D
A
D
D
External
stories
snapshot
API,
so
when
the
snapshot
EPA
comes
in,
the
provisional
actually
sends
the
request
to
the
storage,
controller
and
storage
controller
accesses
these
Eastern
target
to
take
the
snapshot
and
cease
to
target
internally
will
synchronously
make
the
call
to,
or
it
basically
pauses,
the
I
go,
and
then
each
of
the
see
story
to
take
a
snapshot
on
the
corresponding
volumes.
Replicas
right
and
in
case
one
of
the
node
is
not
available
when
the
replica
was
taken
when
it
comes
back
up
it.
Basically,
there
are
reasons
the
data
and
then
takes
the
snapshot.
E
D
E
E
A
D
So
what
we
do
is,
we
said
the
toleration
levels
to
kind
of
restart
very
quickly
for
these
target.
Pods
I
have
not
started
using
the
still
exploring
using
the
port
priority
again.
Some
of
these
things
are
going
to
come
as
the
features
get
into
beta
and
are
available
around
the
clusters.
Some
of
the
clusters,
where
is
used
or
string.
A
D
All
three
layers
can
be
independent
by
default.
They
are
independent,
but
you
can
also
four
depending
on
the
story.
That
means
they
can
actually
Punk
it
at
the
policies
to
say
that
the
workloads
need
to
be
on
one
of
the
storage
nodes
and
also
the
top.
It
has
to
be
only
on
those
storage
nodes,
all
co-located,
always
with
the
application,
these
all
translated
to
or
definitely
anti
affinity.
D
D
I'm,
a
little
bit
on
the
coconutty
slack
channel
as
well
as
me
upon
the
email
I'm
on
the
CNC
F
we
mail
address
will
probably
take
more
questions.
They
do
not.
Let
me
come
back
to
you
on
that
one
question
and
then
one
of
the
other
things
that
we
need
to
check
on
Alexis
the
licenses.
If
you
only
talk
to
you
with
the
different
each
of
the
projects
that
we
haven't
repository
and
what
is
the
current
license
scheme
that
they
have
and
what
are
the
implications
for
that.
A
B
D
D
E
Just
just
by
the
way,
the
licensing
stuff,
usually
the
CN
CF,
the
Linux,
Foundation
staff,
kind
of
sort
of
a
licensing
stuff.
They
have
lawyers
and
they
they
understand
open
source
licenses
way
better
than
us
engineers.
So
so
we
we
can
totally
just
hand
over
that
stuff
to
them
and
tell
them
to
so
today.
It
won't.
You
know.
Obviously
the
project
won't
be
formally
incorporated
into
the
CN
CF
unless
it
can
be
changed
to
a
compatible
license,
and
sometimes
that
has
takes
a
while,
but
but
yeah
in
some
sense.
E
Gotcha
one
final
question,
which
I
think
you
guys
are
answered
to
me
on
a
previous
call,
but
just
for
the
rest
of
the
people
on
this
call.
Could
you
just
give
us
some
sort
of
indication
of
your
main
reason
for
wanting
to
Prince,
or
particularly
to
donate,
was
to
CN
CF,
n
and
open
source
for
that
matter.
You
know,
people
are
gonna,
ask
how
how
you
gonna
make
money.
Are
you
doing
you
know
open
core
or
you
know
how
do
we
assume
that
you're
still
gonna
be
around
next
year?
If
you.
D
C
Mean
I'm
happy
to
but
yeah
it's
a
good
question,
of
course,
but
no,
we
we
do
have
a
model
which
says
for
this,
let's
say
persona
who
tends
to
want
to
grab
this
and
self,
adopt,
give
them
100%,
open
source
solution
for
the
enterprise
and
then,
let's
say
the
VP's.
You
want
to
see
everything
maybe
begin
to
do
some
things
like
programmatic
controls
and
so
forth.
We
have
something
we
call
today.
We
call
it
Maya
online,
which
is
a
SAS
solution
that
gives
you
a
you,
know
beautiful
GUI
and
we
think
beautiful
management.
C
E
C
Hoping
we're
hoping
Quinton
and
it's
a
little
hard
to
know
to
what
extent
folks
have
been
inhibited
by
us
being
a
sole
let's
say:
I
saw
under
project,
but
that's
the
thesis
here
is
that
that
the
you
know
there
have.
Certainly
we
have
like
single
double
triple
contributions
by
some
folks
who
could
do
a
lot
of
help,
so
the
goal
would
be
to
unlock
some
of
that
and
then
just
yeah,
I,
guess,
I.
Suppose
the
marketing
or
awareness
piece
that
that
you
mentioned
as
well.
Okay,.
E
C
E
It's
all
right,
cool,
I,
think
we're
officially
out
of
time.
That
was
really
great.
Thank
you.
Everyone
very
informative,
I,
think,
presentation
and
yeah.
We
can
take
this
forward.
I
think
the
next
step
is
to
actually
put
a
formal
proposal
together
and
get
some
CNCs
sponsors
and
I.
Can
some
TRC
sponsors
I
can
help
doing
that
and
this
anyone
has
any.
You
know
major
objections.
You
can
speak
to
me
about
it,
but
that's
what
I
would
suggest
our
next
steps
are.