►
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
Keynote: Kubernetes and Other CNCF Projects Overview - Liz Rice, Technical Evangelist, Aqua Security & Janet Kuo, Software Engineer, Google
For more info click here: https://sched.co/FzD2
A
So
I'm
Jenna
Cole
I'm,
a
software
engineer
at
Google
I'm
co-chair
of
this
conference
and
I
had
a
great
time
putting
together
the
schedule
with
Liz
Rice.
We've
got
lots
of
great
sessions
scheduled
for
you
today,
I'm
going
to
give
you
an
overview
of
kubernetes
and
let's
first
take
a
look
at
how
popular
kubernetes
is
today.
A
A
Combining
the
worldwide
user
searches
results,
you
can
see
that
kubernetes
is
clearly
training
as
and
has
grown
exponentially.
I
started
working
on
kubernetes
in
mid-2015,
and
that's
me
and
surely
after
I
joined,
kubernetes,
1.0
was
released
and
CNCs
was
formed
and
it
was
all
before
the
first
ever
chicken.
A
A
A
So
it
is
very
clear
that
kubernetes
has
transitioned
from
early
market
of
early
adopters
to
mainstream
market
of
early
majority
users.
Those
were
early
majority
users
have
very
different
expectations.
They
didn't
want
to
learn
the
Aqsa
t
that
the
ins
and
outs
of
urban
areas
anymore.
They
wanted
to
just
work
so
that
they
can
focus
on
what
they
actually
care
about,
which
is
delivering
their
business
values
and
react
with
market
faster,
and
the
good
news
is
that
urban
Ares
is
now
so
great,
so
solid
and
so
mature
that
it
is
now
very,
very
boring.
B
A
Don't
get
me
wrong,
boring
is
good,
for
it
means
that
a
lot
of
people
are
already
using
it
and
it
just
works
it
also.
It's
also
good
for
mainstream
market
users,
because
it
allows
them
to
run
their
business
without
worrying,
about
operating,
proven
edits.
So
when
I
was
preparing
for
this
keynote,
I
looked
at
the
reason.
Kubernetes
features
and
I
felt
that
tremendous
now
focus
on
two
things.
A
Now,
let's
first
take
a
look
at
open
standard
standard
gives
developers
expectations
around
how
a
system
will
the
behavior
of
the
system
and
how
their
applications
were
wrong.
So
kubernetes
provides
a
set
of
standard
built-in
api's
and
those
api's
are
and
provides
a
convenient
abstraction
layer
over
infrastructure
so
that
your
applications
are
portable
across
different
kubernetes
implementations.
A
So
because
of
that,
you
can
run
kubernetes
in
all
different
of
different
environments.
You
can
run
it
in
hybrid
and
multi
cloud
or
even
edge,
and
you
can
find
sessions
of
our
running
kubernetes
in
different
environments
in
this
conference.
The
next
standard
is
conformance,
so
then
just
mention
the
certificate.
So
when
you
see
this
certified
kubernetes
logo,
you
can
be
sure
that
when
you
choose
this
service
providers
version
of
kubernetes
implementation,
you
will
get
the
required
built-in
api's
and
you
will
get
consistent
behavior
that
helps
you
to
make
your
application
truly
portable
across
different
environment.
A
So,
to
recap
with
kubernetes
open
standard,
you
can
run
kubernetes
anywhere
and
you
will
get
consistent
behavior
because
of
conformance.
So
we
just
talk
about
open
standards.
Now,
let's
look
at
extensibility,
so
kubernetes
is
extensible.
It
takes
future
growth
into
consideration
when,
when
we
decided
so
extensibility
has
basically
two
parts.
The
first
is
infrastructure
extensibility
and
the
second
is
API
extensibility
infrastructure
extensibility
means
that
kubernetes
with
kubernetes
you
can
customize
how
kubernetes
consumes
the
underlying
infrastructure.
A
A
Another
extensibility
is
API
extensibility,
so
I
just
briefly
talked
about
the
building
api's,
they
are
generic
and
they
cover
about
80%
of
use
cases,
but
you
sometimes
you
need
to
customize
for
your
own
business
needs
right.
So
with
this,
you
are
empowered
to
create
your
own
custom
API,
and
you
can
write
your
own
controllers
that
automates
the
underlying
text
for
you.
What's
more,
you
can
create
your
own
API
policy
so
that
you
can
modify
your
API
object
and
you
can
do
some
checks
on
them.
A
Today.
There
are
already
more
than
a
thousand
github
projects
that
leverage
API
extensibility.
For
example,
you
can
use
it
for
running
machine
learning,
workloads
for
managing
databases,
for
monitoring,
for
building
servers
frameworks
or
even
for
managing
resources
outside
of
kubernetes,
so
it
makes
kubernetes
with
kubernetes.
You
can
use
a
single
config
language
to
manage
your
in-house
software
and
all
operand
software.
A
So
to
recap:
kubernetes,
open
standard
and
extensibility
makes
it
a
ubiquitous
general
purpose
platform.
For
you
to
build
your
own
platform,
it
has
a
set
of
standardized
abstractions
that
allows
you
to
make
your
applications
portable.
You
will
get
consistent
behavior
because
of
conformance
and
kubernetes
is
extensible,
so
you
can
customize
coverage.
However,
you
want,
and
kubernetes
is
open
and
compatible,
so
you
are
free
to
choose.
A
A
B
Thank
You,
Janet
and
hi
everyone
I'm
so
happy
to
be
here
at
the
very
first
cube
con
I'm
cloud
native,
can't
here
in
China,
so
as
Janet
says,
I'm
going
to
be
giving
you
an
update
on
the
major
news.
That's
been
happening
in
the
CNCs
project
since
our
last
event
in
Copenhagen
in
May
just
a
few
months
ago
this
year,
and
a
lot
has
happened,
but
before
we
dive
into
the
individual
projects.
Let's
just
recap
on
what
we're
trying
to
achieve
with
these
projects.
We
adopt
them
to
the
CNCs.
B
The
goal
is
to
create
a
full
stack
of
the
components
that
you
need
to
run
your
applications
cloud
natively
as
an
end
user.
You
should
be
able
to
build
and
run
your
applications
in
a
fully
scalable
way,
with
automated
infrastructure
and
self-healing
systems
running
in
any
cloud,
whether
it's
public,
private
or
some
combination.
B
Let's
use
the
CNC
F
trail
map
to
walk
through
the
different
projects
and
how
many
like
what's
been
happening
over
the
last
few
months,
so
the
first
steps
you'll
take
as
an
organization
moving
to
cloud
native,
is
to
decide
to
containerize
your
applications
and
dependencies
and
start
building
them
automatically
using
a
CI
and
maybe
CDE
platform
and
you'll
be
using
an
Orchestrator.
Janet
has
been
telling
you
about.
What's
happened
in
cuba
nexus
recently.
B
You
also
need
a
way
to
define
your
applications
so
that
they
can
run
in
kubernetes,
and
that
is
the
goal
of
the
helm
project.
Now
helm
used
to
be
a
sub
project
under
kubernetes,
but
it
recently
moved
to
becoming
an
incubation
project
in
its
own
right
and
you
can
use
hound
charts
to
define
even
the
most
complex
of
applications
to
run
under
kubernetes.
B
Having
deployed
an
application,
we
need
to
be
able
to
observe
its
behavior
and
there
are
several
projects
under
the
cnc
f
in
this
observability
category.
Prometheus
is
a
service
monitoring
system
and
time
series
database.
That
seemed
huge
adoption
in
recent
times,
as
dan
mentioned
earlier
in
August
this
year
it
became
the
second
project
in
the
CNC
F
to
graduate
and
that's
a
really
significant
milestone,
leading
up
to
graduation.
The
team
focused
on
stability,
and
you
can
now
reliably
do
more
than
a
million
data
points
per
second
on
any
modern
machine
and
sustain
that
rate.
B
Now,
let's
move
on
to
thinking
about
how
the
different
application
services
can
discover
and
communicate
with
each
other,
envoi
or
envoy,
depending
on
how
you
pronounce
it
is
a
high-performance
service
mesh
proxy.
It's
a
very
active
project.
There
have
been
over
a
thousand
improvements
from
over
a
hundred
contributors
since
Copenhagen,
with
improvements
to
health
checks,
TLS
support,
load,
balancer
flexibility
and
lots
more
in
June.
B
Services
need
to
be
able
to
discover
each
other
and
core
DNS
is
a
DNS
server
that
can
be
used
by
kubernetes
as
the
clustered
DNS
service
and,
in
fact,
from
version
1
point
11
of
kubernetes.
If
you're
installing
using
cube
ADM
core
DNS,
will
now
be
the
default.
So
congratulations
to
the
core
DNS
on
achieving
that
link.
Adi
is
a
full-service
mesh
implementation
and
version.
2
was
announced
in
September
with
a
complete
rewrite
in
rust
for
the
data
plane
and
go
for
the
control
plane.
B
Link
Adi
can
operate
as
a
service
sidecar
that
doesn't
require
cluster
wide
installation.
So,
as
a
service
developer,
you
can
access
link
Adi
features
without
needing
access
to
the
entire
kubernetes
cluster
link.
Adeem
now
supports
zero
downtime
configuration
changes
using
a
feature
that
allows
multiple
processes
to
bind
to
the
same
port.
So
you
can
start
a
new
link.
Id
process
wait
for
it
to
start
handling
traffic
and
then
terminate
the
old
version
moving
on
along
the
trail
map,
we'll
skip
past
networking
and
the
container
network
interface,
because
they've
remained
pretty
stable.
B
So
we'll
move
on
to
the
question
of
storing
data
in
a
cloud
native
application.
Vitesse
is
a
database
clustering
system
for
horizontal
scaling
of
my
sink.
My
sequel
there's
a
session
tomorrow
discussing
how
JD
comm
have
been
using
kubernetes
and
Vitesse
for
over
a
year
with
a
massive
system
of
over
4,000
shards,
the
rook
project
moved
out
of
the
sandbox
and
became
an
incubating
project
in
September.
It's
a
file
block
and
object.
Storage
solution
for
kubernetes
they've
also
released
a
rook
framework
to
make
a
general-purpose
solution
into
which
storage
providers
can
integrate
their
storage
solutions.
B
Moving
on
along
the
trail
map,
the
cncs
also
supports
projects
that
enable
cloud
native
application
components
to
communicate
with
each
other.
G
RPC
is
a
high-performance
RPC
framework
based
on
HTTP
2
they've,
recently
released
GRP
see
web
that
allows
web
clients
to
communicate
directly
with
a
kind
web
services.
G
RPC
now
supports
16
different
languages
and
has
quickstarts
for
12
of
them.
B
B
There
have
been
recent
announcements,
my
Microsoft
Azure
and
data
dog
that
they're
now
using
tough.
So
that
brings
us
to
the
end
of
the
trail
map,
but
there
are
other
projects
that
the
CNC
F
looks
after
in.
What's
called
the
sandbox,
because
the
projects
in
incubation
and
graduation
are
not
yet
a
comprehensive
solution
for
the
entire
stack
that
we're
aiming
to
build.
So
the
sandbox
gives
us
the
opportunity
to
support
experimental
projects.
We've
welcomed
lots
of
projects
into
the
sandbox,
and
there
are
two
projects
I
want
to
mention.
B
Although
the
first
we
just
haven't,
moved
it
onto
the
trail
map.
Yeah
we've
already
heard
about
harbor,
which
was
in
the
sandbox
until
five
minutes
ago,
and
I
also
wanted
to
mention
ta
TI
kV,
which
is
a
distributed,
transactional
key
value
database
written
in
rust
and
also
with
its
roots
right
here
in
China.
B
B
So
you're
here
at
cube,
con
and
cloud
native
con
and
it's
a
great
opportunity
for
you
to
learn
more
and
get
involved
with
these
projects
over
the
next
two
days.
There
will
be
plenty
of
sessions
which
Janet
and
I
had
the
pleasure
of
curating
and
they'll
teach
you
lot
about
the
individual
projects
and
user
stories
and
all
sorts
of
interesting
things
that
are
happening
around
the
world
of
cloud
native.
B
B
You
learn
more
about
their
projects
and
whether
you're
an
end
user
or
a
developer,
help
you
get
involved
or
hear
your
ideas
and
then
finally,
don't
forget
the
hallway
tract
one
of
the
main
benefits
of
being
physically
here
together
is
that
we
can
have
face-to-face
discussions
and
get
to
meet
each
other
personally,
so
I
hope
you
have
a
fantastic
time
over
the
next
couple
of
days.
Hope
you
really
enjoy
your
time
here
at
Q,
con
cloud
Mesa
con
here
in
Shanghai.