►
From YouTube: Kubernetes Federation WG sync 20180117 part2
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
This
way,
which
you
know
much
of
that
work
has
already
been
done
in
kubernetes,
so
there's
a
lot
of
prior
art
to
lean
on
there,
and
hopefully
the
number
of
cases
where
we
think
we're
special
and
different
enough
from
kubernetes
that
we
need
to
solve
the
same
problem
in
a
different
way,
because
we're
the
ocean
will
be
small,
then
I'm,
not
excluding
the
possibility
of
in
such
cases,
but
but
I'm
hoping
it
will
be
relatively
small.
That
seems
like
a
more
likely
of
seed
procedure.
A
B
I
think
that
makes
a
lot
of
sense
and
I
I,
don't
see
any
rationale
and
starting
over
I
mean,
like
I
I,
really
think
that
a
lot
of
the
discussions
that
we've
had
about
what
what
these
api's
look
like,
they
like
I,
can't
point
to
a
single
thing
that
I'm,
like
ever
had
in
my
brain.
That
makes
me
feel
like
the
prior
discussions
that
we've
had
are
like
somehow
off
base,
so
I
think
I.
A
So
there's
you
you
guys
previously,
you
guys
being
redhead
kind
of
put
your
hand
up
to
do
some
of
that
work.
Are
you
in
a
position
to
put
forward
some
api's
for
review
and
to
form
the
basis
of
this
discussion?
I
know
your
phone's
been
doing
some,
so
we
can
either
you
know
kind
of
overlap
and
provide
two
proposals
for
the
same
type
or
we
can.
You
know,
split
divvy
the
work
up
and
say
you
know.
The
Red
Hat
group
focuses
on
type
a
B
and
C,
and
the
Huawei
group
focuses
on
the
yeah.
B
I
think
so
we,
the
exercise
that
we're
going
through,
has
been
trying
to
just
play
the
movie
forward
on
what
we've
already
discussed
in
in
the
context
of
this
group
and
what
I,
what
I
think
might
be
fruitful
is
is
maybe
F
on
and
some
folks
from
Red
Hat
and
anybody
else
that's
interested.
Maybe
we
can
sync
up
before
the
next
meeting.
I
suspect
that
we
probably
have
a
lot
of
the
same
things
in
mind
and
I.
Don't
think
that
there's
been
a
sizable
amount
of
work
in
any
case
that
that
would
be
duplicated.
C
B
C
B
C
C
Okay,
so
yeah
I
did
not
have
anything
else
apart
from
so
one
other
point,
I
just
wanted
to
bring
it's
like.
Okay,
oh
this
patrician
working
group
right
now
is
the
priority
about
editing
clusters
across
homogeneous
and
of
clusters
like
based
on
GC
or
just
based
on
a
the
players,
are
also
on
something
like.
C
A
I've
certainly
had
a
lot
of
requests
from
customers
for
specifically
hybrid
cloud,
it's
kind
of
like
a
hot
topic.
A
lot
of
big
companies
have
a
variety
of
reasons
to
12
months
across
multiple
cloud
providers
on
crime
and
public
being
the
most
prevalent,
but
also
multiple
public
clouds
is
the
other
one.
And
so
you
know
at
the
end
of
the
day,
somebody's
gonna
network,
and
there
are
challenges
that
are
associated
with
hybrid
cloud,
that
don't
necessarily
exist
this
and
extend
for
what
we
might
call
a
homogenous
cloud.
C
Yeah
sure,
and
basically
why
I
asked
this
like
is
that
any
challenge
to
something
like
a
federated
service
discovery
model?
Currently
we
do
that
based
on
DNS
right.
So
is
there
any
shortest
plan
or
something
to
implement
something
else,
or
are
they
going
forward,
but
directed
service
discovery
as
its.
D
No
problem,
just
clarifying
I
think
that
there's
work
in
progress
on
the
STS
I
to
do
multi
cloud
service,
discovery
and
I
think
that
sort
of
model
is
probably
a
good
alternative
for
people
who
need
those
capabilities.
Like
the
challenge
with
I,
don't
know,
there's
different
use
cases
I
think
we're
we're
probably
going
to
need
to
have
different
solutions
for
different
use
cases.
The
current
solution
works
for
some
use
cases.
D
B
Might
just
say
a
quick
word
on
hybrid
cloud:
that's
that's
something
that,
like
I've,
had
a
number
of
customer
requests
and
interest
so
back
to
I.
Think
that
I
and
to
do
plus
1
what
Quintin
said
there:
a
lot
of
extra
additions
to
the
problem
space
in
hybrid
cloud,
and
so
it
seems
to
me
that,
like
probably
as
as
we
chart,
you
know
future
directions
that.
B
Necessary,
or
at
least
and
one
of
the
ingredients
to
figuring
out
hybrid
cloud,
is
figuring
out,
homogeneous
cloud
or
homogeneous
Multi
cluster.
So
I
guess
what
I
mean
is
probably
watch
the
space
and
there
will
be
maybe
some
time
where
hybrid
cloud
is
not
the
predominant
focus,
but
I
think
that
we
should
be
mindful,
as
we
talk
through
like
more
basic
scenarios,
that
we're
not
painting
ourselves
into
a
corner
when
it
comes
to
hiding
that
makes
sense.
Yeah
yeah.
A
A
B
A
So
those
are
the
areas
for
my
experience.
Those
are
too
difficult
to
solve
areas,
but
I
also
think
that
that
we
don't
necessarily
need
to
solve
them
because
they
actually
have
whole
industries
around
both
of
those
problems
already,
and
we
should
the
same
way
that
we
mentioned
that
we
could
use
a
skier,
for
example,
for
cross
class.
Curves,
obviously
I
think
that's
the
right
approach
rather
than
to
try
and
like
invent
our
own
thing
of
using
and
I
think
the
same
applies
to
authentication.
If
and
where
there
are
third-party
tools.
A
B
Yeah
I
think
I
think
those
are
like,
probably
the
two
big
rocks
in
the
field.
There's
also
like
just
an
extra
element
to
the
hoth
one
like
once.
You
start
talking
about
hybrid
cloud
and
integrating
these
different
identity
providers
from
clouds.
There's
also
like
a
potential
entanglement
with
you
know.
Whatever
shows
an
organizational
identity.
B
Often
in
another
area
of
my
professional
life,
which
is
like
when
I
have
a
hybrid
cloud
solution,
there
are
service
offerings
from
different
clouds
that
are
in
similar
spaces.
There's
there's
not
a
that.
You
know
they
have
their
own
differentiators
and
a
lot
of
users
and
folks
in
the
community
have
asked
me
in
the
conflicts
Service
Catalog.