►
From YouTube: Observability Service for Cloud-scale Apps John Kendall (Datadog) OpenShift Commons Gathering 2021
Description
OpenShift Commons Gathering 2021
Lightning Talk: Observability Service for Cloud-scale Apps
Speaker: John Kendall (Datadog)
https://commons.openshift.org/index.html#join
A
Hello,
everybody,
my
name
is
john
kendall
and
I
lead
the
container
monitoring
product
at
datadog
and
for
those
who
don't
know
what
datadog
is.
Datadog
is
an
observability
service
for
your
cloud
scale,
applications
and
every
year
we
deliver
the
container
report,
which
I'm
excited
to
share
with
you
today,
where
we
examine
over
1.5
billion
containers
from
tens
of
thousands
of
customers,
and
we
share
some
trends
that
we're
seeing
in
the
container
ecosystem
that
we
all
participate
in.
A
A
A
As
a
matter
of
fact,
nearly
90
percent
of
manage
of
organizations
are
using
managed,
kubernetes
services,
so
thinking
eks,
gke
aks,
and
they
are
letting
cloud
providers
do
the
control
plane
operations
for
them.
Moving
on
looking
at
pod,
auto
scaling,
so
we
examine
both
horizontal
pod,
auto
scaling
and
vertical
pod.
A
Auto
scaling
horizontal
pod,
auto
scaling,
meaning
you're,
changing
the
number
of
pods
identical
pods
that
you're,
using
in
response
to
fluctuations
in
demand
vertical
pod,
auto
scaling
talking
about
how
you
can
change
the
characteristics
of
your
pods
in
terms
of
cpu
in
memory
and
considering
horizontal
pod,
auto
scaling.
A
Another
question:
would
you
say
that
over
10
of
organizations
are
using
horizontal,
auto
scaling
or
less
than
10?
So
by
show
of
hand
over
10
are
using
10
of
organizations
are
using
horizontal
pod,
auto
scaling,
okay,
not
many,
and
how
many
of
you
think
less
than
10
are
using
horizontal
pod.
A
Auto
scaling
yeah
about
40
percent
of
organizations
are
using
horizontal
pod,
auto
scaling
and
are
really
becoming
more
sophisticated
and
they're
being
comfortable
having
their
metric
services
emit
what
they
need
to
watch
for
to
change
the
number
of
pods
in
response
to
demands
in
traffic.
A
Datadog
is
an
observability
service,
so
people
use
monitors
and
we
looked
at
not
only
organizations
using
containers
but
organizations
not
using
containers
and
what
are
your
thoughts
on?
Would
you
say
that
organizations
using
containers
in
terms
of
the
number
of
monitors
for
organization
organizations
using
containers
versus
not
using
containers
which
type
of
organization
typically
had
more
monitors
so
by
show
of
hands,
non-containerized
organizations,
more
monitors
and
containerized
organizations
using
more
monitors?
A
We
appear
to
be
split
down.
The
middle
turns
out
that
organizations
using
containers
had
more
monitors
as
of
2021
by
a
meaningful
amount.
A
It
will
be
removed
and
for
context.
Docker
shim
is
an
underlying
component
within
kubernetes
that
enables
users
to
be
able
to
use
the
docker
container
runtime
with
kubernetes,
then
it's
being
removed
and
users
are
being
are
migrating
to
other
container
runtimes
such
as
container
d,
and
this
is
something
that
I'm
sure
a
lot
of
you
are
curious
about
is
where
this
is
going.
So
by
a
show
of
hands
around.
A
Would
you
say
over
10
at
this
point
have
migrated
away
from
docker
container
runtime
or
less
than
10,
so
over
10
percent,
okay
and
less
than
10
percent
yeah.
It
seemed
like
we're
kind
of
split
down
the
middle
yeah.
We
have
some
ways
to
go
in
this
migration.
We
found
that
at
the
moment
about
since
january,
about
seven
percent
have
migrated
to
container
d
in
this
context
and
to
wrap
things
up
with
the
last
theme
that
we're
looking
at
considering
I'm.
A
Cool,
so
the
percentage
growth
of
users
using
openshift
how
much
by
show
of
hands
over
10
have
migrated
the
percentage
growth
of
users
using
openshift
and
less
than
10
percent
growth.
A
B
A
So
this
report
was
published
today,
and
this
is
talking
about
as
of
2021,
so
to
wrap
things
up,
I
invite
you
to
come
join
us
at
our
major
booth
at
the
at
the
expo
area.
We
can
talk
about
the
report
and
you're
welcome
to
receive
a
demo
of
datadog.
Thank
you
for
listening.