►
From YouTube: Keynote: Spinnaker - The Future of Your Enterprise CD? Three Reasons Why - Gopal Dommety, CEO, OpsMx
Description
For more Continuous Delivery Foundation content, check out our blog: https://cd.foundation/blog/
Keynote: Spinnaker - The Future of Your Enterprise CD? Three Reasons Why - Gopal Dommety, CEO, OpsMx
Continuous Delivery is a uniquely exciting space to be working in today, sitting at the intersection of so many trends and innovation. While Spinnaker has proved to be the robust CD platform that enterprises need, what will be Spinnaker’s role going forward in a world increasingly dominated by containers, Kubernetes, and multi-cloud deployments?
In this presentation Gopal will share his perspective on the future of Spinnaker based on his work with some of the world’s largest Spinnaker deployments. He will highlight the important efforts now underway to make Spinnaker easier to adopt at scale, improve the developer experience with smart GitOps, and add control to emerging new tools like Argo CD. Together, these initiatives promise to keep Spinnaker at the center of your future CD strategy.
A
A
A
It
made
it
easier
to
adopt
at
scale,
and
you
know,
thanks
to
all
of
you,
whoever
you
are,
whichever
company
you
work
for
open
source
contributors,
commercial
support,
vendors,
Armory
officer,
Max
and
all
the
users
and
all
the
companies
that
got
to
that
scale,
because
I
know
how
much
hard
work
you've
put
in
so
big.
Thank
you
today.
A
A
It
was
originally
designed
that
way,
giving
the
ability
to
enhance
the
Enterprise
capabilities
to
meet
the
future
needs
of
an
Enterprise.
You
know
four
or
five
years
ago,
Argo
wasn't
on
the
scene,
so
an
example
is
Argo
CD.
Today
it
has
new
kubernetes
capabilities
and
I'll
share
with
you.
How
Ops,
Max
and
others
are
working
to
integrate.
Spinnaker
are
good
together.
A
Enterprise
githubs
will
go
a
long
way
in
simplifying
the
Spinnaker
developer
experience,
which
is
something
that
we
should
all
strive
to
work
towards,
and
each
of
these
is
Big
topics.
So
the
interest
of
time
I'll
keep
the
conversation
at
a
high
level,
I'm
happy
to
chat
more
so
first,
a
Spinnaker
has
made
continuous
delivery,
a
simpler,
secure
and
scalable.
A
How
did
it
make
it
simple
made
it
easy
to
use
lots
of
enhancements?
We
ourselves
have
operator
one
command
installation
for
for
beginners,
just
one
single
command
configuration
through
everything
through
a
UI
Etc,
easy
to
operate
and
I'll
talk
about
Spinnaker,
SAS
and
then
easy
to
extend
and
I'll
talk
about.
You
know
optimax
Center
of
Excellence.
There
are
a
lot
of
things,
Community
has
done
together
and
we're
only
just
I'm
going
to
only
scratch
the
surface
here.
A
Spinnaker
says
you
know:
optimax
has
a
Spinnaker
SAS,
it's
simple
to
get
started.
All
you
need
is
the
account.
If
you're
behind
a
firewall,
you
need
a
remote
CG
agent,
you
deploy
it
and
once
you
deploy
it,
it
helps
you
access
all
the
resources,
deploy,
resources
behind
firewalls
and
also
any
Enterprise
tool.
That's
behind
the
firewall
can
be
will
be
accessible
through
the
SAS.
A
The
the
other
feature
is
it's
a
scalable
security.
Essentially,
you
can
configure
your
keys
security
credentials
all
on
the
local
agent,
not
necessarily
on
in
the
server
it.
This
allows
for
autonomous
teams,
meaning
that
each
team
is
isolated.
You
can
have
thousands
of
teams,
thousands,
tens
of
thousands
of
kubernetes
clusters
all
working
autonomously,
every
every
team
using
its
own
tool
chain,
Etc
I-
think
it's
it's.
It's
actually
super
simplifies
the
scaling
and
operation
of
Spinnaker
and
security,
so
self-service
for
developers
allowing
them
to
manage
targets
and
tools.
A
So
this
goes
at
least.
Our
view
goes
a
long
way
in
simplifying
operations,
and
the
third
is,
you
know:
we
started
a
center
of
excellence,
you
know
and
large
Enterprises
need
capabilities,
not
in
the
open
source.
Spinnaker-
and
you
know,
Optima
centrifice
Excellence
can
develop
new
features.
Customized
for
you
examples
are
examples.
Are
gcp
Cloud
run
load
balancer,
Cloud
driver
starting
are
all
in
these
various
releases,
so
so
this
is
how
we've
simplified
Spinnaker.
A
Now,
how
did
you
know?
Spinnaker
is
safer
because
it's
a
safer
platform
fixed
a
lot
of
vulnerabilities
overall,
the
usage
in
in
the
last
two
years
have
made
it
safe,
say
for
delivery
process.
We
now
have
much
more
controls,
control,
Gates
stages,
Integrations
best
practices
coded
as
templates.
All
of
this
makes
the
delivery
process
itself
is
safer
and
I
also
alluded
to
how
the
act
of
accessing
Enterprise
resources
through
Spinnaker,
be
it
as
a
SAS
or
on-prem.
When
you
have
a
large
large
implementation
becomes
extremely
simple,
safe.
A
The
third
is
about
how
to
make
it
scalable
in
the
in
the
law.
As
I
said
in
the
last
12
12
months,
many
Enterprises
went
to
scale.
Community
has
made
lots
of
enhancements.
One
specific
problem
that
we
focused
on
is
large-scale.
Spinnaker
deployments
have
had
the
bottleneck
of
Spinnaker
Cloud
drivers
service,
and
we
now
have
a
service
called
storm
driver
to
solve
this
problem,
and
so
we
are
open
sourcing
Strom
drivers
from
driver
is
a
new
service
for
cloud
driver,
especially
yes,
it's
efficient.
It
takes.
A
A
So
so,
if
you
know
this
is
how
we
as
a
community
are
making
Spinnaker
the
first
sort
of
the
reason
why
I
think
we
have
a
good
base.
The
second
is
now
I
want
to
talk
about.
The
second
reason:
how
Spinnaker
is
able
to
successfully
integrate
with
emerging
sort
of
tools
like
Argo?
You
know
part
of
what
makes
software
delivery
exciting.
A
You
know
some
of
us
are
old
enough
to
have
been
pre,
pre-kubernetes
and
sort
of
kubernetes
came
in,
and
so
an
example
now
is
Argo
Argo
has
got
good
adoption,
and
so
there
will
be
many
many
Innovations
to
come
and
developers
are
driving
adoption
or
Fargo
due
to
its
Simplicity
and
and
support
for
kubernetes.
Most
developers
don't
want
to
deal
with
policies
or
scaling
issues.
Security
operations
need
to
step
in
to
provide
these
controls,
so
we're
integrating
Spinnaker
with
Argo
to
support
the
choice
for
these
developers
and
to
embrace
that
sort
of
adoption.
A
So-
and
you
know
some
of
them
want
to
leverage
Argo
rollout
capability
with
Spinnaker
and
others
want
to
benefit
from
having
Spinnaker's
control
in
an
Argo
CD
environment.
So
if
you
so,
what
does
it
mean
to
integrate
Argo
and
Spinnaker
or
interoperate?
There's
sort
of
different
levels
of
integration
and
operation
available,
Loosely
coupled
to
to
more
integrated,
so
some
start
with
adding
Argo
rollouts
as
a
mechanism
for
kubernetes
deployments.
A
So
that's
the
sort
of
the
beginning.
The
second
is,
you
know:
Argo
users,
especially
if
you
have
a
whole
Argo.
Your
student
deployment
need
mechanisms
to
promote
between
environments,
enforce
organization
policies,
control,
especially
control
access
to
other,
actually,
access
other
Enterprise
tools,
control
them
and
make
annotations,
and
all
of
that
stuff.
I
think
so.
Spinnaker
adds
that
delivery
control
to
Argo,
CD
that
and
and
finally,
it's
about
both
Spinnaker
and
Argo
are
affecting
effect
operating
each
other.
A
So
users
need
the
end-to-end
sort
of
visibility
of
all
the
stages,
especially
if
you
have
mix
and
match
of
them
and
then
in
in-depth
audit
across
all
deployments.
Audit
of
pre
during
deploy
and
post
deployment
audit
and
we're
able
to
provide
that
with
Spinnaker
and
I
I
mean
I
think
that
Spinnaker's
ability
to
integrate
with
new
Innovations,
like
this
will
is
a
is
a
big
sort
of
a
capability
that
enterprises
would
love.
A
You
know
all
of
the
you
know
about
this
new
model
of
of
operation
that
has
gained
mind,
share
and
traction
to
some
extent.
I
think
it
can
go
a
long
way
in
simplifying
the
developer
experience
if
we
adopt
it
properly.
Yeah.
Let
me
share
my
perspective
and
especially
Enterprise
githubs
can
be
made
fully
comprehensive
with
leveraging
Spinnaker.
A
So
if
you
step
back
and
good
Ops
has
started
as
a
sort
of
a
declarative
spec
for
infrastructure,
as
code
for
deployments,
primarily
in
the
kubernetes
environment
and
with
the
source
of
Truth
forget,
although
sort
of
git
sorry
drops
was
popularized
in
kubernetes
environments,
users
are
looking
to
expand
that
model,
Beyond
kubernetes
two
VMS
to
serverless
to
and
so
on.
So
essentially
it's
like
the
first
step
is:
how
do
we
get
to
Beyond
kubernetes?
A
Let
me
share
with
you
how
we
are
thinking
about
it,
and-
and
hopefully
we
should
all
work
together
on
it
so
and
the
first
is
githubs
for
multi-cloud,
and
this
is
nothing
new
to
this
community.
There's
always
been
Keel,
so
Enterprise
our
view.
At
least
you
know
we
are
leveraging.
Keel
Keel
is
a
good
sort
of
a
framework
developed
by
Netflix.
You
know
Extended
by
AWS,
and
we
are
using
that
to
extend
to
other
environments
it.
A
So
essentially
it
provides
a
declarative
framework
for
deployments.
You
know
a
skill,
database
and
I
think
you
all
are
familiar
with
it
and
so
we're
looking
to
support
kale
to
go
beyond
now.
There's
a
talk
later
today,
Autodesk
and
AWS
I
think
we're
going
to
touch
upon
this,
the
the
so
so.
The
the
second
step
is:
how
do
we
take
githubs,
Beyond
deployments?
A
Let
me
talk
about
declarative
functions.
Supported
by
git
today
is
deployments
in
a
delivery
process.
There
are
other
functions
as
I
mentioned
that
are
outside
the
scope
of
current
devops
view.
If
you
look
at
these
functions,
like
policies
verifications
approvals
promotions
to
other
functions,
they
can
all
be
first,
you
know
specified
in
the
git
now
you
can
some
of
them.
A
You
can
specify
them
as
goals
and
then
declare
it
will
implement
it,
and
so
we
are
looking
at
verification
policies
as
declarative
constructs,
and
so
that's
where
we
will
start
I
think
if
we
can
sort
of
implement
the
extend,
git
Ops
and
have
declarative
versions
of
these
functions,
we
will
be
able
to
realize
the
vision
of
fully
operating
or
fully
delivering
without
any
human
intervention.
You
know
one
of
the
goals,
one
of
the
dreams
of
when
we
started
optimax.
Okay,
Ops
stands
for
operations
machine,
you
know
it's
it's.
A
A
So
the
third
sort
of
is
is
a
yeah,
so
so
yeah.
So
when
we
look
at
looks
like
a
Skips
a
slide,
yes,
so
what?
What
we're
doing
is
to
take
all
of
these
functions
and
have
the
declarative,
spec
foreign
of
these
functions,
and
and
so
that's,
what
we're
working
on
and
have
implementations
of
the
controllers
for
them
and
and
a
big
part
of
fully
automating
software
delivery
is
making
sure
that
github's
model
incorporates
Specialties
of
concern
Beyond
developers.
A
So
if
you
look
at
existing
githubs
for
developers,
there
are
SEC
security
teams
for
complex
deployments.
You
need
security
teams
that
need
to
do
guard
rails,
Security
checks,
Ops
teams
that
need
to
do
sizing,
monitoring
and
so
on
and
so
forth.
So
there
are
two
sort
of
enhancements
needed
to
to
make
this
happen.