►
Description
How CloudBees CI Brings Enterprise Scale to Jenkins - Doug Tidwell, CloudBees
With more than 70% market share, Jenkins is the #1 technology in the CI space. But how do you make it scalable and manageable? What about governance and compliance? In this video, you'll see how CloudBees CI solves those problems.
A
Hello,
this
is
doug
tedwell
for
cloudbees
over
the
next
few
minutes,
we'll
be
talking
about
cloud
bci,
which
you
may
know
as
cloudbees
core
cloud.
Bci
delivers
enterprise
jenkins
at
scale,
we'll
talk
about
some
of
the
most
common
problems
people
have
when
they
try
to
scale
jenkins
and
we'll
look
at
how
cloud
bci
solves
them.
A
We'll
start,
though,
with
a
few
numbers,
three
goals
and
three
unpleasant
truths.
Our
first
number
is
15.
Jenkins
is
more
than
15
years
old.
It
is
battle
tested.
It
is
used
every
day
by
hundreds
of
thousands
of
people
for
10
of
those
years
cloud.
Bees
has
been
helping
customers
scale
jenkins
and
make
the
most
of
this
technology
that
large
and
active
jenkins
community
has
created
more
than
sixteen
hundred
plug-ins.
A
That
means,
if
you
need
to
integrate
a
tool
with
your
build
pipeline,
there's
probably
an
existing
plugin.
That
will
do
that
for
you
for
cloudbees
customers,
there
is
the
cloudbees
assurance
program
which
ensures
and
reassures
you
that
the
most
popular
of
those
plugins
are
secure
and
stable
and
safe
to
use
in
your
production
environment.
A
Our
next
number
and
our
very
favorite
is
the
number
one
with
more
than
70
percent
market
share.
Jenkins
is
the
number
one
technology
in
the
ci
space
for
the
open
source
version
of
jenkins.
In
addition
to
building
cloud
bci
cloud
b's,
employees
contribute
more
than
80
of
the
code
changes
to
the
open
source
jenkins
project.
A
B
A
A
A
We'll
talk
about
these
goals
and
how
cloud
bci
can
help
you
achieve
them
now
for
our
three
unpleasant
truths.
Our
first
unpleasant
truth
comes
from
our
friends
at
forester.
They
have
found
that
48
of
all
organizations
have
more
than
100
tools
in
their
various
tool
chains.
So,
if
you
look
across
the
organization,
there
are
lots
of
tools
that
have
to
work
together
to
automate,
builds
and
deliver
software.
A
A
A
A
A
The
first
problem
is
the
problem
of
hidden
costs,
especially
this
is
an
opportunity
cost,
because
probably
the
best
engineers
on
each
team
are
keeping
the
jenkins
server
up
and
running
and
patched
keeping
the
latest
levels
of
plugins
installed
on
them
and
so
forth.
That
means
they're,
not
innovating,
so
you're
actually
spending
a
lot
of
time.
A
lot
of
money
and
wasting
a
lot
of
resource
by
keeping
all
of
these
separate
jenkins
servers
up
and
running.
A
A
It
gets
back
to
the
80
percent
of
devops
efforts
that
won't
scale,
but
the
biggest
problem
is
that
we
have
no
governance
here,
this
team,
being
the
forward
thinking
engineers
that
they
are,
they
make
sure
that
every
pipeline
running
on
that
jenkins
server
requires
a
code
build
to
include
a
security
scan.
That's
a
best
practice,
that's
great,
but
from
an
administrative
point
of
view,
we
have
no
idea
that
that's
happening
here.
A
We
have
no
idea
if
it's
happening
on
the
other
servers,
we
have
no
oversight,
no
control,
no
idea
what
these
individual
jenkins
servers
are
doing.
So
this
is
a
problem.
We
call
the
islands
of
jenkins
problem,
because
what
we
end
up
with
is
lots
of
servers,
each
of
which
has
overhead,
each
of
which
is
completely
different,
and
we
have
no
no
policies,
no
guidelines,
no
governance
in
place
to
make
sure
everyone's
doing
the
same
thing.
A
A
The
first
problem
is
the
issue
of
scalability
how
many
cpus,
how
much
ram,
how
much
disk
space
do
we
need
to
throw
at
this
monolithic
server,
to
keep
it
up
and
running
and
make
sure
it
performs
to
our
needs,
and
even
if
we
do
that
for
today,
will
it
be
enough
a
month
from
now?
Will
it
be
enough
on
the
last
day
of
a
sprint,
that
can
be
a
difficult
problem
to
solve.
A
The
second
problem,
which
is
somewhat
related
to
that,
is
that
this
can
become
a
single
point
of
failure
for
the
entire
organization.
If
your
jenkins
server
goes
up
in,
flames
goes
up
in
flames,
that
was
fun.
Wasn't
it
if
your
jigging
server
goes
up
in
flames?
No
one
in
your
organization
is
building
anything
until
it's
back
online.
A
A
A
A
I
have
islands
of
jenkins
that
are
not
governed,
I'm
not
sure,
what's
happening,
and
I'm
spending
a
lot
of
time
and
resource
maintaining
those
instead
of
innovating
and
I've
got
a
monolithic
server
which
takes
substantial
resources
to
keep
up
and
running.
But
it's
not
meeting
everyone's
needs
either.
However,
shall
we
solve
this.
A
Problem
the
answer
to
our
scalability
problems
and
governance
issues
is
cloud
bci.
The
reason
that
we're
here
with
cloud
bci,
you
have
some
number
of
managed
masters.
We
have
master
number
one
master
number,
two
master
number
three
and
so
on.
Each
of
these
is
independent,
so
they
can
have
different
software
stacks
different
tools,
different
plugins.
A
Each
of
these
can
meet
the
needs
of
an
individual
organization
or
project.
They
are
also
all
managed
by
cloud
bci.
So
we
have
central
management
here
which
can
give
us
the
governance
and
compliance
that
we
need.
A
couple
of
things
that
are
great
about
this
one
is
that
onboarding
becomes
really
easy.
If
we
have
a
new
project
or
a
new
team,
we
just
create
a
new
master
and
everything
is
good.
That
team
is
up
and
running
very
quickly.
A
A
A
A
A
Secondly,
at
surah,
when
we
talk
about
reducing
risk,
their
use
of
cloud
bci
led
to
a
roughly
75
percent
decrease
in
defects.
That's
amazing
and
again
lowers
risk
by
taking
more
and
more
defects
out
of
the
picture
and
finally,
at
cintine
they
worked
with
clubby
support
and
services
to
move
from
their
islands
of
jenkins
to
cloud
bci.
A
A
A
C
Hey
how's
everybody
doing
today,
hey
matt,
can
you
hear
me
I
can?
Can
you
hear
me?
Yes,
I
can
yeah
good
stuff,
isn't
it
yeah
enterprise
jenkins?
So
do
we
have
any
questions
here?
Let
me
see
the
chat
here
or
anybody
has
any
questions.
There
was
one
question
from
sebastian
about.
Do
we
have
anybody
in
quebec?
C
I
did
answer
it
on
the
chat,
but
I
can
definitely
speak
to
her
here.
We
do
have
a
global
presence.
So
if
you
have
anybody,
it
may
not
be
properly
in
that
area.
But
you
know,
if
you
do
have
a
need.
You
can
always.
You
know,
go
to
cloudbees.com
contact
us
or
even
if
you
have
any
questions
right
now,
we
can
bring
it
to
the
table
and
we
can
discuss
it,
but
actually
here
to
help.
If
there's
any
questions
you
may
have.
B
So
on,
while
we're
waiting
for
people
to
chime
in
with
questions
in
the
chat,
I
was
just
curious,
so
we
watched
doug
walk
through
the
cloud,
bci
architecture
and
kind
of
how
it
addresses
some
of
the
challenges
that
that
organizations
face
with
scaling
open
source
jenkins.
So
I
was
wondering
if
you
could
maybe
share
some
experiences
with
you
know
what
sorts
of
challenges
you've
seen
customers
run
into
with
trying
to
scale
jenkins
and
maybe
how
cloud
bases
has
stepped
in
to
help
with
those
those
issues.
C
So
some
teams
would
be
very
successful
in
you
know,
building
sort
of
a
jenkins,
I
would
say
almost
a
practice
for
a
component
for
an
application,
maybe
a
smaller
size
teams
when
they
start
quickly,
realizing
the
the
goodness,
as
I
like
to
call
it
of
jenkins
and
scale
it
to
multiple
different
teams.
They
start
stumbling
upon
many
different
problems.
It
could
be
a
performance
issue.
C
They
have
multiple
different
teams
that
have
come
together
on
that
same
jenkins.
We
all
love
and
then
they
start
hating
it
because
guess
what
you
know
there's
an
application,
there's
certain
performance
it
has
and
naturally,
if
you're
not
getting
the
right
feedback
loop
as
a
developer,
you
start
hating
that
application
right.
It
could
be
jenkins,
it
could
be
something
else.
So
that's
what
we
saw
as
a
pattern
where
we
had
these
pockets
of
success,
so
I'm
just
giving
you
some
examples
of
a
large
financial
company.
C
I
was
working
with
just
last
week
and
their
goal
was
that
we
had
these
teams.
They
all
have
the
same
power.
They
all
had
jenkins.
They
all
had
certain
sort
of
tool
set.
You
know
some
artifact
repository,
but
they
were
just
not
able
to
scale
so
really
what
it
came
down
to
to
your
point
is
we
provided
some
of
those
practices?
For
instance,
how
do
you
make
sure
you
spend
less
time
on
administration,
so
one
of
the
big
things
that
they
got
got
benefit
out.
The
gate
was
didn't
have
to
change
anything.
C
You
know
they
use
jenkins.
They
continue
to
use
jenkins
with
global
ci,
because
they're
built
on
that
jenkins
platform,
which
is
what
we
do,
though,
is
we
do
we
did
validate
and
certify,
and
on
top
of
it,
then
you
provided
templates.
I
think
that
was
a
big
thing
for
them.
You
know
if
you
have
these
pockets
of
success.
If
you
want
to
replicate
that
across
the
organization
you
can
templatize
it.
So
if
you
think
of
it,
what
I'm
talking
about
is
we
were
spending
less
time
on
administrating
jenkins.
C
The
teams
are
spending
more
time
on
building
pipelines.
I
think
that
was
a
quick
win
that
they
got.
We
did
this
in
a
span
of
about,
I
would
say,
less
than
a
month
to
get
running,
and
then
we
had
other
teams
onboard,
but
that's
how
we
do
it
so
just
about
when
I
say
scaling
the
scale.
We
mean
by
focusing
less
time
on
administration,
focusing
more
time
on
the
value-added
pipelines,
and
then
you
scale
that
successes
through
the
organization.
B
Yeah
that
totally
makes
sense
and
is
totally
in
line
with
you
know
things
that
I've
seen
as
well.
You
know
at
that
point
about
it
kind
of
how
do
you
translate
that
the
pockets
of
success
into
in
organization-wide
enterprise,
ready
kind
of
recipe
for
success
at
scale
is
often
a
pretty
serious
undertaking?
C
C
But
yes
thank
you,
but
at
the
same
time
he's
saying
did
he
have
to
did
he
have
to
really
write
backwards?
Isn't
that
funny?
Whenever
I
see
those
videos,
I
always
wonder
exactly
how
that's
working,
but
it
is
some
neat
technology.
I
think
you
know
when
I
looked
at
it
and
yes,
he
did
practice
sliding
backwards.
C
I'm
connect
there's
a
there's
a
little
solution
for
that,
but
it
is
pretty
nice
when
you
see
it
and
why,
whatever
big
on
white
boards
as
well
lucas,
so
I
love
when
I
see
something
like
this,
but
I
think
the
other
question
you
are
having
lucas
is.
Can
you
give
me
some
detail
an
example
of
how
templates
can
be
used
to
promote
best
practices?
C
Yeah,
absolutely
I
mean
the
fact
is
you
know,
we've
all
been
trying
to
do
that.
Isn't
it
lucas
we
build
something
and
we
feel
that
pipeline
is
a
success.
To
be
very
frank
with
you
before
we
had
this
template
functionality
in
cloud
vci
the
enterprise
jenkins
we
used
to
see
when
I
go
to
customers
even
today.
I
see
a
lot
of
these
shared
libraries
getting
created
and
you
know
customers
go
and
they
start
getting.
Those
shared
libraries
so
going
back
to
an
example
of
a
template
would
be
a
very
simple
example.
C
You
know
one
of
the
most
common
examples
you
can
see
in
the
jenkins
world
is
like
a
java
maven
pipeline,
a
build
pipeline
right
and
you
want
to
almost
do
a
continuous
integration
pipeline,
which
does,
let's
say,
unit
testing.
You
know
some
sort
of
code
scanning
whatever
that
you
have
to
do.
Basically,
that's
your
continuous
education
pipeline
and
how
different
is
it
when
you
think
of
like
these?
You
know
artifacts
getting
built
in
this
java.
Maven
space
I
mean
it's,
it's
the
same
should
be
the
same
process.
Isn't
it
you?
C
You
have
a
repository
or
a
set
of
repositories
and
you
point
to
the
repository
and
you
want
those
artifacts
to
get
built
and
then
published
to
the
artifact
repository
so
lucas
to
your
point
or
the
question
you're
asking
about
template.
That's
a
good
example.
All
you
want
to
do
is
build
a
pipeline
in
such
a
way
where
you
can
abstract
those
variables
to
point
to
a
repository
to
point
to
where
your
artifacts
are
published,
and
then
you
rinse
and
repeat
that
in
such
a
way
that
you
can
create
those
templates.
C
So
I'm
hoping
that
helps
you.
You
know,
of
course
you
could
put
it
on
chat,
but
the
template
is
that
how
we
build
templates,
we
actually
provide
something
called
in
copy
ci.
We
have
something
we
have
called
pipeline
template
catalogs
that
works
hand
in
hand
with
jenkins
pipelines
with
shared
libraries,
to
make
sure
that
our
customers
can
provide
those
pipeline
template
catalogs
across
the
enterprise
as
well.
So
matt.
C
B
Yeah,
I
think
that's
a
great
summary
that
you
know,
as
you
start,
to
build
out
your
collection
of
of
pipelines.
That
patterns
are
going
to
start
to
emerge
and
your
different
project
types
and
language
types
and
etc,
where
it
starts
to
make
sense
where
you
can
capture
those
in
a
template
that
then
just
get
parameterized
as
you.
You
share
these
across
different
instances.
B
The
other
thing
I'll
add
there
is,
I,
I
think,
of
templates
in
kind
of
two
different
layers
within
cloud
vci.
So
there's
the
the
job
level
templates,
but
also
templates
in
the
sense
of
setting
up
your
jenkins
instances,
your
controllers,
so
there
are
also
mechanisms
within
cloud
vci
that
facilitate
that
so
being
able
to
set
up
what
sorts
of
plugins
you
want
installed.
What
your
system
level
configuration
you
want
to
be
being
able
to
kind
of
set
that
up
prior
to
provisioning
is,
is
also
something
that
we
find.
C
Yeah
great
point
matt
exactly:
we
cannot
forget
the
the
controllers,
in
fact,
speaking
of
controllers,
the,
as
we
all
know,
we
are
not
calling
the
jenkins
masters
anymore
because
of
diversity
reasons
as
the
community
voted
for
it
to
be
called
controller.
So
we
at
cloud
b
is
also
at
the
forefront,
make
sure
that
we
start
actually
practicing
ourselves
to
call
it
controllers.
So
whenever,
let's
say
you
hear
me
or
matt
talking
about
jenkins
masters,
in
fact,
is
jenkins
controllers
going
forward.
C
So,
but
the
fact
is
the
same
thing
as
matt
said
is
about,
as
you
think,
of
templatizing,
your
pipelines,
think
of
almost
templatizing
and
standardizing
on
your
controllers
as
well,
and
then
you
can
start
really
winning
that
game
of
spending
less
time
and
administration
and
more
times
on
pipelines,
very
good
yeah
keep
those
questions
coming.
This
is
this
is
one
I
think
you
have
a
few
minutes
to
go
it's
about
four
or
five
minutes
remaining.
C
So
matt
you
had
some
good
questions.
What
what
else?
What
are
the
questions
that
sort
of
comes
about
any
interesting
engagements?
You
had
recently
any
customer
scenarios.
I
mean
he
was
a
cuban
administrator
as
well
right,
certified
cka
very
sought
after
so
when
you
think
of
it.
This
way,
any
recent
engagements
where
customers
are
deploying
jenkins
and
kubernetes
and
what
the
experience
has
been
and
how
we
help
them,
get
the
benefits
of
kubernetes
but
then
scale.
B
Yeah,
absolutely
so,
it's
definitely
something
that
that
comes
up
pretty
frequently,
and
you
know
one
of
the
the
other
aspects
of
this
cloud.
Vci
platform
is
it's
kind
of
hybrid
nature
and
being
able
to
bridge
that
gap
between
you
know
your
more
traditional
vm-based,
jenkins,
implementations
and
kind
of
moving
forward
into
that
cloud,
native
kubernetes,
based
type
installation,
so
that
sort
of
conversation
is
something
that
we
we
have
all
the
time
and
can
be
a
pretty
fun
exercise
to
look
at.
B
You
know
what
is
a
customer's
current
approach
to
pipeline
authoring
and
how
do
we
easily
translate
that
to
running
on?
Kubernetes
is
with
as
minimal
disruption
as
possible,
so
making
sure
that
you
know
there's
no
interruption
in
workloads
and
that
you're
able
to
use
the
same
labels,
the
same
tools
that
you're
calling
from
your
pipelines,
but
starting
to
kind
of
embrace
that
paradigm
shift
where
everything
is
now
a
container
or
a
pot
in
kubernetes
and
of
course,
there
are
numerous
different
benefits
that
come
from
running
your
your
jobs
and
jenkins.
B
B
C
Yeah,
I
love
it
matt.
I
think
that's
really
where
I
think
cloudbees,
as
as
a
partner
shines
with
the
customer,
I
feel
is
just
imagine
being
in
a
jenkins
administrative
shoes
and
we
have
the
complexities
to
deal
with
now.
There
is
the
of
course,
the
pipeline
development.
There
is
the
maintenance
of
the
infrastructure,
and
then
you
have
kubernetes,
and
you
know
many
times
they
may
actually
learn
and
themselves,
but
then
you
start
combining
those
things
it
becomes
super
complicated,
but
at
the
same
time
you
want
to
get
the
value
from
it.
C
C
C
You
know
normally
your
aks
as
well,
but
the
fact
is,
you
know
many
different
platforms,
it
just
you
put
it
on
you,
get
it
up
and
running,
and
then
you
basically
start
focusing
on
how
you
can
get
that
value-added
concept
as
well
so
lucas
one
minute
ago,
but
you
had
a
question
which
is:
is
this
will
be
a
cr
solution,
somehow
related
to
jenkins
x,
so
we're
looking
into
that?
I
mean
techton
is
one
of
those
things
which
is
sort
of
underlying
underpinning
to
jenkins
x.
Have
we
invested
in
jenkins
x?