►
From YouTube: Keynote Kubernetes Big Picture, 1 22 and Beyond
Description
Kubernetes Community Days Bengaluru'21
Kubernetes has been growing by leaps and bounds over time. In this talk, you will hear about the big picture of how Kubernetes offers a fresh Cloud-Native perspective for Infrastructure and Applications. You will learn about the ecosystem and community being built in the Kubernetes and larger CNCF. You will also get to know about what the community is planning for the upcoming release 1.22 and future plans that we are cooking up. There are quite a few experiments and proof of concepts we are building that will shape the future to come. We will celebrate the work of folks from India who are diligently working in the community as well.
Slides: https://drive.google.com/file/d/1TM4e2_lRZQtvKT-ZGJKqKb1Gw-JRRnS0/view?usp=sharing
A
Hi,
everyone
welcome
to
kubernetes
communities
days
in
bangalore,
2021
and
let's
get
started.
So
what
I
wanted
to
start
today
was
to
by
paying
tribute
to
folks
that
we
lost
recently
dan
cohen
and
phupta
dan
was
instrumental
in
making
cncf
what
it
is
today
and
pierce
was
a
valued
member
of
the
community
who
passed
away
recently
due
to
covet.
A
Please
click
on
the
urls
below
basically
scan
code,
the
qr
code
to
us
to
relive
some
memories,
and
you
know,
leave
your
thoughts
there.
A
little
bit
about
myself.
I
was
born
and
brought
up
in
india
ended
up
in
boston.
I've
been
doing
open
source
for
a
long
time.
I
wear
a
few
hats
in
the
community.
I
am
employed
by
vmware
and
I'm
easily
reachable
on
twitter,
slack,
github,
etc.
A
A
The
world
by
you
know,
has
changed
life
for
everybody,
as
we
know
it,
and
some
of
the
things
that
we
are
seeing
is
that
a
lot
of
growth
in
sas,
for
example,
people
moving
to
public
clouds,
people
trying
to
use
multiple
cloud
vendors
also
edge
the
on
premise
on
premise:
installation
of
kubernetes
may
be
a
little
bit
lower,
but
overall,
I
think
it's
it's.
The
industry
itself
has,
you
know,
changed
quite
a
lot
in
the
last
year.
A
A
So
if
you
look
at
the
cloud
native
report
from
cncf,
you
know
the
number
of
people
that
are
developing
on
kubernetes
using
kubernetes
is
growing
by
leaps
and
bounds.
The
last
one
is
from
2019
for
obvious
reasons.
We
didn't
run
a
report
last
year,
but
you
know
even
the
last
report
from
2019
show
is
showing
a
whole
bunch
of
usage
of
kubernetes
cloud
container,
orchestration
technologies,
and
you
know
cncf
projects
in
general.
A
There
is
also
significant
developments
happening
around
serverless,
active
architectures
and
cloud
functions
as
well,
and
a
whole
bunch
of
serverless
users
are
using
things
like
aws
lambda
as
well.
Now,
what
is
cloud
native?
This
is
the
precise
definition
from
cncf
as
to
what
cloud
native
is,
and
this
basically
tells
us
that
you
know
people
should
start
rethinking
their
software
development
practices
and
how
they
approach
architecting
software
and
make
them
you
know,
run
well
on
public
clouds
and
not
try
to
design
things.
A
A
A
So
some
of
the
projects
you
might
recognize,
for
example,
k3s
keyword,
no
and
things
like
that
cni
genie,
argo
volcano-
are
you
know.
Smi
is
a
new
thing
as
well
that
we
are
trying
to
see
if
people
are
interested
in
is
important
for
security.
You
know
going
forward.
So
there
is
a
lot
of
momentum
building
around
you
know
not
just
kubernetes.
A
In
general,
but
cloud
native
technologies
engine
as
well
so
in
terms
of
graduations,
we
graduated
five
different
projects
and
a
whole
bunch
of
prior
projects
also
moved
from
sandbox
into
in
incubation.
I
was
mentioning
our
book
before.
That
was
one,
so
we
also
have
observable
tools
like
thanos
and
cortex
in
the
mix,
and
we
do
have
kubernetes
at
the
edge
with
cube
edge.
A
Now
what
is
kubernetes
itself
right,
and
why
are
you
all
here?
All
of
you
here
are
because
of
kubernetes,
and
you
know,
kubernetes
is
used
for
managing
containers
applications
and
it
helps
you
spread
the
containerized
applications
across
multiple
hosts
and
we
can
do
scaling
and
deployment
and
maintenance
of
things
really
easily,
and
it
gives
you
like
a
platform
layer
over
and
above
different
cloud
providers
or
even
on-prem
as
well.
A
A
This
is
media
mentions
and
other
things
like
that
cncf
collects
information
from
different
sources
and
uses
that
to
construct
this
graph
in
terms
of
contributions
to
from
various
folks.
You
know,
google,
vmware,
red
hat
microsoft,
ibm
all
the
people
that
you
expect
amazon
intel,
suse
apple,
huawei,
alibaba,
everybody
is
there
in
the
ecosystem
and
all
are
collaborating
well
and
trying
to
get
things
done
in
the
community.
A
A
We
were
just
talking
about
how
many
different
companies
are
invested
in
it,
and
it's
not
just
the
companies
themselves,
but
you
know
people
contribute
as
individuals,
people
and
there's
a
long
tail
of
contributors.
Coming
all
from
all
over
the
world
from
all
geographical
locations
working
on
kubernetes,
that
is
definitely
a
game
changer.
A
The
way
a
kubernetes
community
works
has
has
been
eye-opening
as
well.
You
know
right
at
the
beginning,
you
know
things
got
added
into
the
same
repository
the
big
repositories,
but
over
a
period
of
time
we've
been
able
to
like
harvard
pieces
like
the
cni
interface
for
networking,
the
cri
interface
for
runtimes,
the
csi
for
the
storage
and
for
the
cloud
providers.
A
We
have
the
cpis,
so
we
kind
of
like
split
things
apart
and
users
in
use
interfaces
to
to
essentially
allow
people
to
experiment
enhance
in
their
own
repositories
on
their
own
time,
and
you
know
not
worry
about
the
changes
that
are
happening
in
kubernetes
kubernetes
itself,
so
crds,
aggregated
apis
have
had
a
huge
effect
in
building
the
ecosystem.
You
take
any
of
the
the
projects
that
we
were
talking
about.
Qno,
you
know,
falco,
you
take
any
of
those
projects
that
we
were
talking
about.
A
All
of
them
end
up
using
crds
of
some
sort,
because
crds
are
the
native
way
to
talk
to
kubernetes
resources
using
the
cube,
ctl
command
lines.
The
command
line
interface-
and
you
know
it-
makes
the
crds
make
it
really
easy
for
people
to
not
only
interact
with
the
customer
resources,
but
also
to
write
code
right
operators
that
work
against
the
crds
to
get
things
done.
A
It's
been
battle,
tested,
reliable,
scalable
and
we've
been
putting
out
releases,
we
are
at
like
121
and
122
is
in
progress
so
and
don't
forget
all
the
alpha
beta
and
the
release
candidates
that
we've
released
over
a
period
of
time,
so
there's
a
huge
ecosystem
that
tests
validates
runs
things.
So
you
know
you
can
rely
on
kubernetes
for
your
own
needs
as
well.
So
we
talked
a
little
bit
about
the
declarative
apis
and
eventual
consistency
now.
So
what
are
we
talking
about
here?
A
So
what
we
are
talking
about
is
control
loops.
So
there
is
this
concept
of
level
versus
edge
based
triggering
as
well,
and
what
we
are
trying
to
say
here
is
you
know
you
as
a
user?
You
want
to
say
I
want
something
done
right
and
then
the
controller
will
go.
Look
at
it
and
say
what
is
the
current
state
of
the
world
and
what
is
the
state
that
you
want
and
if
it
is
same,
we
are
all
done.
A
A
So
if
the
layered
architecture
makes
it
really
easy
to
understand,
as
well
as
try
to
figure
out
like
how,
where
things
belong
and
how
to
interact
with
things
so
in
terms
of
the
components
itself,
you
know
there's
this
handful
of
like
six,
seven
components
that
form
the
heart
of
the
system.
We
have
the
api,
server,
controller
manager
and
cubelet
that
are
really
important
and
scheduler
is
used
for
scheduling
the
pods.
A
The
q
proxy
is
kind
of
an
optional
thing,
because
there
are
some
cni
providers
that
offer
a
replacement
for
q
proxy.
That
works
as
well.
Well
as
as
well,
and
the
the
main
thing
that
qrintus
is
able
to
do
is
when
you
express
an
intent
as
a
resource,
then
we
end
up
turning
it
into
things
that
get
scheduled
on
the
node
and
cubelet
is
the
heart
of
things
running
the
things
on
each
node.
A
We
do
store
things
in
hcd,
which
is
our
external
persistent
store.
Some
people
have,
you
know,
talked
about
replacing
hcd
with
other
things.
You
know,
and
there
are
projects
in
the
community
as
well.
For
example,
k3
is
use,
uses
kind
and
can
work
with
other
storage
providers
as
well.
So
this
is
the
heart
of
kubernetes,
these
five
six
processes.
A
A
The
community
is
divided
into
special
interest
groups
and
work
groups,
and
we've
had
like
a
really
good
experience.
In
terms
of
you
know
the
number
of
contributors
and
the
number
of
sigs
and
working
groups-
and
we
love
having
new
leaders,
step
up
and
old
leaders
making
way
after
providing
mentoring.
A
So
emeritus
basically
means
that
the
folks
that
were
running
things
are
stepping
down
and
they
are
able
to.
You
know,
groom
new
people
to
take
over
all
the
things
that
they
were
doing.
So
one
important
milestone
is
we
crossed
100
000,
pull
requests
in
kubernetes
kubernetes
repository.
This
is
a
huge
milestone.
This
is
probably
the
second
project
to
cross
this
milestone
and
we
also
passed
the
5
50
000
contributors
mark
as
well.
A
Now
I'm
having
a
little
trouble
converting
it
into
lags,
but
you
can
do
the
math.
So
then,
in
terms
of
like
coverage
in
terms
of
like
conformance
testing,
you
know
how
do
you
know
what
you're
running
from
a
specific
vendor
supports
all
the
things
that
you
can
do
with
kubernetes?
You
know
we
have
test
coverage
for
like
75
percent
of
the
api
endpoints
and
we
are
rapidly
catching
up
to
do
full
conformance
as
well.
There's
lots
of
sub
projects.
A
A
A
Now
when
things
are
in
beta,
they
are
turned
off
when
by
default
and
when
they
are
in
alpha,
they
are
turned
off
by
default.
So
if
there
is
something
that
is
marked
alpha,
then
if
you
want
to
use
it,
you
switch
things
on
in
various.
You
know:
kubernetes
components,
command
line,
command
lines,
for
example,
abi
server
and
cubelet
have
feature
flags
that
you
can
turn
on.
A
A
So
in
terms
of
talking
promotion
to
stable,
we
have
a
whole
bunch
of
things,
but
these
are
the
most
important
ones
that
are
going
from
beta
to
stable.
Some
of
the
things
that
I'm
really
happy
about
are
the
warnings
mechanism,
especially
because
you
know
people
don't
read
the
release
notes
and
when
they're
on
the
cube,
ctl
command
line.
You
know
they
get
the
warning,
saying:
hey
you're,
trying
to
do
something.
A
A
You
know
auto
scaling
apis
and
external
credential
cloud
providers,
and
I
can
go
on
and
on
about
each
of
these
things,
but
I
don't
think
we
have
the
time
so
watch
out
earlier.
We
used
to
do
four
releases
a
year.
Now,
we've
got
dropping
down
to
three.
There
are
important
things
like
docker
shim
that
are
getting
deprecated.
So
please
stop
using
docker
as
a
runtime.
A
A
A
A
Can
you
re-imagine
how
we
would
write
distributed
system
code
when
we're
not
talking
about
like
pods
and
deployments
and
those
kinds
of
things
but
brand
new,
different
fields,
all
right?
What
would
it
look
like?
Can
you
still
use
cucurl
to
control
like
microwave
or
whatever
right,
so
whatever
we
can
come
up
with?
So
this
is
very
exciting.
You
know
thinking
about
like
how
we
can
reuse
all
the
declarative
apis
and
things
like
that
in
a
non-uh
container
ecosystem.
A
So
the
main
thing
that
I'm
really
happy
about
is
all
the
people
that
I
get
to
work
with
on
a
daily
basis
like
mother,
helps
us
out
with
container
controller
on
time
test.
Infra
premca
helps
out
with
python
client.
Aditi
is
working
hard
on
sig
node
and
she's
written
some
caps,
or
helping
with
some
caps.
Naveed
has
been
helping
me
in
code
organization.
A
Humbly
is
working
on
six
storage.
Sheikh
shekhar
raja
he's
been
writing
a
website
for
the
enhancements
proposal
from
scratch.
Sladen
has
been
helping
with
sig
windows
and
ci.
Signal
sonia
is
on
the
release.
Team
arsh
was
an
intern
and
we
worked
together
for
three
weeks
on
something
really
interesting.
That
provost
provides
a
lot
of
value
to
us.
Dvr
has
been
working
on
the
release.
Team
ayush
has
been
working
on
six
storage
and
csi
drivers.
A
Long
has
been
helping
with
api
and
since
yaml
as
well,
rajula
has
been
helping
with
contrabax
and
release.
Burun
has
been
helping
with
test
infrastructure.
Kunal
has
been
doing
an
awesome
work.
Getting
all
the
students
interested
in
the
things
that
we
do
in
cncf
and
spreading
the
word
shweta
has
been
working
on
an
e2e
framework.
Shivani
and
ankita
have
been
helping
out
with
cluster
api.
A
Rajas
has
been
helping
out
in
different
things,
and
the
most
recent
thing
that
he
was
helping
out
with
was
s390
results
to
test
rate.
Navarone
is
all
over
the
place.
Secretly
apa
missionary
core
organization
contrabax,
you
name
it
and
he's
there
last,
but
not
the
least.
My
co-conspirator
nikita
has
been
awesome
and
I
love
working
with
her
and
she's.
You
know
we
talk
about
all
kinds
of
things
all
the
time
and
I
love
every
every
chance
I
get
to
talk
to
to
this
awesome
team.
A
Now,
how
can
you
join
us?
It's
not
just
the
people
that
I
showed
you.
The
pictures
of
we
have
a
whole
bunch
of
more
people
from
all
over
india
who
work
in
the
community
and
who
do
things
in
the
community
so
come
join
us.
We
have
a
channel
slack
channel
on
kubernetes
slack
in
dev
and
we
try
to
have
fun
as
well
when
we
are
able
to
get
together.
These
are
a
pre
coverage,
so
don't
worry
about
it
if
you're
not
wearing
masks
in
these.
A
A
The
first
is
the
first
link
so
hit
the
first
link
to
get
this
presentation.
You
can
hit
me
up
on
github,
twitter
slack
and
the
most
important
of
all.
If
you
want
to
get
started.
The
last
url
has
a
whole
bunch
of
links,
including
videos
and
presentations
and
hands-on
labs,
and
things
like
that.
So
don't
be
shy.