►
From YouTube: GitLab runner direction 2021-01-03
Description
Overview of the GitLab Runner direction.
https://about.gitlab.com/direction/verify/runner/
A
A
Really
the
goal
is
about
provide
a
bit
more
context
in
terms
of
the
content
that
you
see
on
our
github
run
direction
page,
which
is
available
here
on
about
that
github.com
follower,
slash
direction
forward,
slash
verify
forward,
slash
run,
so
that's
it.
Let's
go
ahead
and
jump
right
into
the
presentation,
so
the
first
kind
of
like
just
to
set
the
stage
just
kind
of
talk
a
bit
about
how
I
get
lab.
We've
defined
our
strategic
client
cadence.
So
in
gitlab
we
have
mission,
vision
and
strategy
for
us
emission
is
30
years.
A
Vision
is
10
and
strategy
is
three,
and
so
I
think,
is
helpful
for
us,
as
we
go
into
the
next
couple
of
slides
to
kind
of
understand
the
time
that
we're
thinking
about
as
we
talk
about
that
long-term
product
vision.
As
we
talk
about
the
github
runner
vision,
how
it
aligns
that
long-term
product
vision
and
then
as
we
think
about
and
talk
about
our
near-term
three-year
strategy,
so
how
does
the
kit
lab
runner
vision
align
with
the
gitlab
product
version
right
so
as
defined
today?
A
The
gitlab
product
vision
is
to
replace
devops
tools,
chains
with
a
single
application
pre-configured
to
look
by
default
across
the
devops
life
cycle,
and
so
we're
thinking
about
that
vision.
Forget
that
product
we're
thinking
about
how
does
the
runner
play
internet?
What
does
the
runner
have
to
do
in
order
to
enable
that
vision
right
for
the
gitlab
runner?
We
support
gitlab.com.
A
We
are
the
engine
that
makes
gitlab
ci
and
get
that
back
on
work,
but
we're
also
the
engine
that
customers
use
and
deploy
if
they're
self-managing
their
own
gitlab
instances
or,
if
they're
self-managing,
their
own
runners
and
so
the
vision
for
running
in
order
to
enable
that
gitlab
product
vision
over
the
next
10
years.
Our
goal
is
to
provide
devops
team
devops
teams,
a
build
agent
that
works
seamlessly
on
tomorrow's
market,
leading
computing
platforms
and
the
tools
to
eliminate
ci
bill-free
operational
complexity
at
enterprise
scale,
that's
division.
A
A
So
at
a
high
level,
I'm
not
trying
to
cover
every
single
thing,
but
just
to
give
you
a
smattering
information.
This
is
the
current
technology
line
scheme
that
the
gitlab
runner
supports
from
architectural
architecture
perspective.
We
support
multiple
compute
architectures
on
x86
s64
image,
intel
ibm
z,
with
s390x
from
an
operating
system
perspective.
Obviously,
we
support
big
three
operating
systems:
linux
windows,
mac
ox
and,
of
course,
I'm
going
to
experience
different
distributions
of
index
right
and
then
from
an
executive
perspective,
I.e
the
execution
engineers
to
run
you
can
see.
A
We
support
a
multiple
number
of
different
executives,
a
day
right,
kubernetes,
docker
dock
machine
powershell,
shell
ssh
and
our
very
own
custom
executive,
which
gives
you
the
flexibility
to
use
the
running
environments
where
we
don't
have
an
in-executive
build
and
then
some
other
executives
as
well
parallels
and
virtualbox
right.
So
multiple
architectures,
multiple
operating
systems,
multiple
executives
and
then
finally,
multiple
major
comp
container
platforms
right.
A
We
support
eq
ec
amazon,
ecs
amazon,
eks
right
fargate
from
aws
red
hat
openshift
gke
on
aks
again
so
at
a
high
level
today
table
stakes
in
terms
of
the
runner
technology
landscape
in
terms
of
all
of
the
different
architectures
or
platforms
that
we
have
to
ensure
that
github
runner
works
seamlessly
on.
This
is
a
snapshot
of
that
kind
of
landscape
and
this
landscape.
As
you
will
see
in
a
subsequent
science,
what
is
informing?
A
A
What
is
what
are
we
thinking
what
happened
in
our
from
the
technology
landscape
perspective
perspective
in
x5
and
and
beyond,
and
how
does
that
inform
our
gitlab
product
strategy,
our
getting
our
vision
right,
so
the
high
level
level
from
an
operating
system's
perspective
and
as
we're
putting
together
this
line,
we're
thinking
in
the
future
we're
thinking
about
adding
more
details
so
as
to
to
provide
a
bit
more
context,
all
right
from
operating
perspective
perspective,
linux,
mac,
os
windows,
we
all
know
they're
mature
in
the
marketplace.
A
The
nexus
has
been
around
since
1991
mac,
os
different
time
frame,
windows,
obviously,
and
windows
server
obviously
been
around
for
quite
some
time.
We
expect
we
forecast
that
these
three
major
operations
operating
systems
being
considered
mature,
will
continue
to
have
a
significant
market
presence
over
the
next
five
years.
We
don't
see
a
significant
change
in
that
sort
of
landscape
from
an
architectural
perspective
marketplace.
Forecasters
and
industry
analysts
are
saying
that
x64
there's
the
tiny
market
share
for
x64
on
personal
computers.
A
We
don't
necessarily
see
that
having
a
significant
impact
in
terms
of
what
we
have
to
support
yet
from
a
gitlab
running
perspective,
so
some
that
that,
as
noted
but
again,
we
expect
x64
to
be
relevant
in
the
market
for
next
five
to
10
years,
even
though
it
has
declined
in
market
share
for
personal
computers
as
the
rise
of
risk,
as
you
can
see
in
the
quadrant
below,
continues
to
continue
to
gain
share
in
the
market
I'm
arc
from
r
from
r,
as
well
as
a
new
apple,
m1,
chipsets
right.
A
We
expect
that
to
continue
to
increase
share
in
the
marketplace.
We
expect
that
the
apple
m1
chips
will
be
released
in
2020
or
in
terms
of
the
mac.
The
macbook,
the
apple,
the
macbooks
and
macbook
pros
will
continue
to
see
growth
in
terms
of
their
adoption
next
three
years
on
the
app
in
terms
of
apple
laptops
and,
of
course,
and
I
would
expect
apple
servers
as
well,
and
then
you
know
entrance
to
that
maturity
phase
in
2024
2025,
right
so
from
an
architecture
and
operating
systems
back
to
a
high
level.
A
There
will
be
changes
in
the
market,
but
we
can
see
that
there's
evolution
in
the
marketplace
in
terms
of
arm
gaining
more
market
share,
but
at
high
level
we
expect
that,
from
a
runner
perspective,
we'll
have
to
continue
to
support
very
well
of
this
architecture,
the
computing
architecture
and
this
operating
system
landscape
over
the
next
five
years
and
then
from
a
container
platform
perspective.
A
Right
now,
as
of
today
all
eyes
on
kubernetes,
it's
got
the
most
market
traction.
We
think
that's
where
the
market
it
will
continue
to
invest
and
where
things
are
heading,
and
that
will
be
the
technology
of
choice
for
the
next
five
years.
A
A
These
major
container
platforms
that
we
support
today
from
ecs
all
the
way
through
you
know
whether
openshift,
google
and
sure
we
expect
that
there
will
be
shifts
in
terms
of
adoption
market,
share,
wins
and
losses,
but,
generally
speaking
within
the
next
five
years,
our
expectations
are
that
that
there
will
that
these
players
will
continue
to
be
relevant
in
the
marketplace
in
terms
of
container
platforms
right
and
that
our
customers
will
continue
to
choose
solutions.
A
One
of
the
solutions
here
in
terms
of
their
kubernetes
offerings
and
so
from
a
runner
perspective.
This
is
kind
of
what
we're
thinking
about
we'll
have
to
ensure
that
we
support
very
well
over
the
next
five
years.
So
there's
the
operating
systems,
no
major
change.
Obviously,
different
variants
of
linux
will
be
out
there
and
you
know
different
life
cycles
for
those,
but
generally
speaking,
operating
systems
and
these
architectures
and
these
container
platforms
is
what
we
expect
to
be
relevant
and
what
we
expect
to
have
to
continue
support
over
the
next
five
years.
A
So
now,
what
does
that
mean
from
a
runner
near
some
strategy
perspective?
What
are
we
thinking
about
doing
in
terms
of
natural
strategy,
from
themes
for
the
runner,
on
our
way
to
that
vision,
right
that
version
of
being
able
to
provide
devops
teams
that
build
agent
that
works
seamlessly
on
tomorrow's
marketing,
compute
platforms
right
and
the
tools
to
eliminate
operational
complexity
right
and
so
in
2021?
Some
of
the
major
themes
that
we
have
on
the
docker
as
follows:
security,
a
really
basic
table
stakes
team.
A
We
have
to
get
continued
to
enhance
our
security
flipping
because
by
ensuring
that
we
have
the
right
security
foundation
allows
us
to
unlock
that
enterprise
management,
that
enterprise
management
goal
right
and
so
security
major
theme
first
in
2021
and
the
other
major
theme
around
enterprise
management
is
the
first
pillar.
If
you
will,
on
that
on
that
journey
towards
seamless
operation
or
frictionless,
operation
or
zero
operation
is
what
we're
calling
that
new
runner
admin
view
in
our
get
lab
ui
and
there's
some
slides
and
everyone.
We
talk
a
bit
more
about
that.
A
So
enterprise
match
a
major
theme
for
us
in
2021.
There
are
a
theme
for
us
in
2021
around
self-managing
one
is
at
scale.
It's
run
as
auto
scaling,
kubernetes
platforms
in
fifth
note:
7.
We
release
a
runner
operator
for
red
hat
open
and
they
have
got
an
open
ship
operator
hub
so
in
in
2021,
we'll
be
moving
our
operator
to
ga
and
then
we'll
continue
to
invest
in
the
evolution
of
our
wondrous
kubernetes
executive.
A
The
other
thing
that
we
will
be
looking
at
as
well
as
from
a
theme
perspective,
2021
israeli,
auto
scaling
on
public
cloud
provider,
virtual
machines,
because
we
recognize
that
that
all
of
our
customers
are
mind
reading
or
adopting
kubernetes.
As
a
platform
of
choice,
as
a
matter
of
fact,
some
industrial
research
has
suggested
that
the
current
trends
are
for
typically
medium
to
large
enterprises
being
the
forefront
of
the
adopters
of
kubernetes,
and
we
can
see
obvious
reasons
for
that
probably
have
more
resources
and
so
on,
and
so
we
can.
A
We
can
expect
that
over
the
next
couple
of
years,
those
trends
to
continue
so
maybe
larger
customers
will
be
adopting
communities
faster
than
smaller
customers.
So,
from
an
auto
scaling
perspective
on
public
cloud
providers,
we
have
to
continue
to
support
communities
and
we
have
to
continue
to
support
giving
customers
the
ability
to
secure
github
runners
on
public
power,
virtual
machines
and
the
next
big
theme
for
us
in
calendar
year.
A
2021
is
run
a
build
club
first
moving
our
windows
runners
on
gitlab.com
to
ga
and
then
introducing
our
macbook
os
build
called
first
as
beta
ng
and
more
specific
and
more
interestingly,
both
on
x64,
as
well
as
on
a
new
apple
camera
on
chipset
right.
And
so
we
really
think
there's
a
really
interesting
opportunity
for
gitlab
to
provide
something
new
and
novel
as
it
relates
to
and
really
kind
of
tying
in
that
concept
of
the
one
devops
platform.
A
With
this
mac
os
field
called
offering
so
security
enterprise
management
actually
key
pillars
for
us
that
helps
us
think
about
on
the
journey
towards
getting
to
that
10-year
vision
and
then
along
those
along
the
way,
auto
scaling,
kubernetes
or
skinning
public
power
providers.
Kind
of
like
wrongs
out
that
some
of
the
the
themes
in
support
of
that
vision
and
then
2022
and
2023
kind
of,
if
you
probably
all
heard
of
the
fuzzy
front
end
things
are
a
bit
more
fuzzy
right
now.
A
So
in
the
next
few
months,
as
we
revamp
these
slides
and
we
continue
to
evolve
our
thinking,
we
might
have
a
bit
more
detail
more
context
in
terms
of
what
we
foresee
in
2022
and
2023
in
terms
of
themes,
but
at
high
level
right
now
we're
thinking
about
continuing
to
evolve.
Our
enterprise
management
capability
sets
continue
to
invest
and
evolve,
kubernetes
executive
to
ensure
that
we
can
scale
the
runs
and
various
communities
platforms
in
a
seamless
manner
and
adding
public
public
cloud
provider.
A
A
Can
we
have
automated
bills
for
all
mac
os
low
cost
so
that
customers
on
github.com
that
are
building
for
the
macbook
ecosystem
can
literally
get
to
the
you
know
that
full
screen
build
without
any
setup
at
all
right,
and
so
can
we
get
to
the
zero
setup
sort
of
like
vision
for
all
mac
os
built
again
tie
into
that
overall
kit
flat
version
of
that
one
application
to
replace
the
entire
device
for
the
developed
cycle
right,
and
so
those
are
the
major
themes
that
are
kind
of
driving
our
three-year
strategy
for
gitlab
running
right.
A
So
this
is
a
the
strategic
items.
The
items
before
dimension
here,
our
tablestake
islands,
making
sure
that
we
continue
to
support
very
well
these
container
platforms
with
our
kubernetes
offerings.
These
architectures
the
operating
systems.
Again
this
is
stable
state.
This
is
keep
the
lights
on
and
then
these
themes
are
some
of
the
things
we're
thinking
about.
In
terms
of
how
do
we
continue
to
evolve?
The
run-up
solution
set
right
to
ensure
that
we
enable
that
long-term
vision.
A
So
now
I'm
talking
about
enterprise
management
and
some
of
the
pain
that
we've
heard
from
customers
around
enterprise
management-
and
this
is
just
a
smattering
of
a
few
quotes
from
customers
around
what
they're,
seeing
as
they
are
now
managing
launch
installations
of
either
github
instance
or
get
lab
runners
right.
You
know
in
some
cases
hundreds
of
teams,
hundreds
of
projects,
hundreds
of
runs
and
so
on.
Right,
as
one
customer
says,
hey,
I
need
an
easy
way
to
determine
measuring
group
inheritance,
another
customer.
A
As
a
github
administrator,
I
can't
find
a
means
by
which
to
identify
the
environment
on
which
your
runner
is
installed.
Where
is
that
going
to
install
that
right?
Very
really,
key
questions
and
administrator
of
something
as
important
as
your
develop
storage?
You
need
to
be
able
to
answer
and
as
a
gift
of
administrator,
it's
difficult
for
me
to
find
and
delete
all
runners,
no
longer
active,
quote,
unquote
dead,
but
still
sure
is
registered
added
to
a
group
or
a
project
right.
A
There
are
very
likely
runners
on
all
versions,
iic
109
at
5.1,
but
we
are
on
gitlab.
You
know,
12.10
and
further.
The
method
of
deploying
configuring
and
manual
runners
is
widely
different
from
team
to
team
increasing
operation
complexity
of
these
teams.
I
think
this
is
just
a
small
ring
of
some
of
the
themes
and
some
of
the
the
feedback
we've
heard
from
customers
that
have
already
embarked
on
on
down
the
path
of
enterprise
management,
with
managing
a
bill
fleet
at
scale,
and
these
are
some
of
the
problems
we're
trying
to
solve
for
us.
A
We
think
about
that
long-term
vision
of
can
we
reduce
operation
complexity.
While
we
ensure
that
we
can
support
the
execution
of
the
run
on
multiple
container
platforms
and
multiple
architectures
and
multiple
operating
systems
again
table
stakes
right
can
we
do
so
in
a
way
that,
while
you're
doing
it
at
scale
that
management
of
that
definite
skill
is
as
seamless
or
operationally
zero
operational
costs
as
possible?
A
And
so
this
is
just
initial
markup
of
one
potential
evolution
of
that
runner
admin
view
here,
we're
kind
of
exploring
some
initial
ideas
to
help
us
kind
of
figure
out
capabilities
or
features
that
we
may
need
to
add
along
the
way
of
solving
that
problem.
Again,
the
goal
is:
can
we
get
to
a
zero
sort
of
a
zero
administration
footprint
right?
A
zero
operational
overhead
footprint?
A
Is
that
long-term
vision,
and
so
you
can
see
on
the
screen
we're
thinking
about
it
doesn't
make
sense
in
this
view,
to
add,
like
some
statuses,
like
online
runners,
right
account
of
online
runners,
account
of
crash
runners,
lost
runners,
outdated
runners
and
so
on,
right
and
and
from
within
this
view,
add
the
ability
ability
for
users
or
administrators
to
take
proactive
action.
Not
just
have
this
view
be
informational
only
but
be
actionable.
A
You
can
see
here
another
concept
view
where
we're
in
it
we're
thinking
about.
Should
we
enable
capabilities
to
allow
an
administrator
to
set
a
number
of
versions
of
a
runner
that
can
be
used
in
the
environment
in
terms
of
conversion,
offsets
and
so
on,
and
so.
A
We
expect
that,
and
we
talk
about
it
quite
a
bit
on
our
direction
page,
that
there
will
continue
to
be
choice
in
the
marketplace.
Five.
Ten
years
from
now,
as
there
has
been
choice
in
the
marketplace
in
the
past,
from
a
computing
architecture
from
a
computing
operating
system
perspective
right
from
a
public
cloud
provider
perspective
and
so
first
path.
A
First
pill
in
our
strategy
is
to
ensure
that
we
support
those
market
leading
computing
platforms
for
our
self-managed
customers,
while
at
the
same
time
ensuring
that
we
provide
the
tool
sets
and
the
capabilities
so
that
managing
github
runners
managing
that
build
fee.
There
is
zero,
if
possibly
getting
that
nirvana
of
zero
operational
maintenance
overhead,
and
so
this
is
kind
of
that
overview.
That's
for
us
kind
of
what's
driving
the
github
runner
version
and
direction
in
a
future
video.
A
A
My
hey
what's
actually
happening
what
might
actually
be
developed
in
the
next
release
or
deliver
so
in
the
next
release
from
our
features
of
liberality's
perspective,
so
talking
more
tactically
about
the
near
term
roadmap
and
things
that
might
actually
be
delivered,
as
well
as,
depending
on
how
things
change
and
our
thoughts
about
this
vision,
the
shot
at
the
unlike
strategy
and
some
of
the
near-term
changes.
How
all
those
things
may
have
evolved
since
this
first
video?
So
with
that
said,
that's
a
wrap.