►
Description
Don’t miss out! Join us at our upcoming event: KubeCon + CloudNativeCon Europe 2023 in Amsterdam, The Netherlands from April 17-21. Learn more at https://kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects.
Keynote: Observability Simplified - Eduardo Silva, Calyptia
Observability is a journey that brings excellent values, at a current cost of an operational complexity associated with it, until now.
In this presentation, we will go through our lessons in the logging space by building the Fluent Bit project and how extending the ecosystem to support Metrics, and Traces has helped shape a simplified user observability experience.
Calyptia, one of the primary sponsors of Fluent Bit, is redefining the observability experience by removing the operational burden; come to learn more about how you can reduce your "time to observe" and "costs" in a vendor-neutral way.
A
Well,
thanks
everybody
as
a
quick
intro
again,
my
name
is
Eduardo
Silva
I'm,
the
co-founder
of
this
company
called
kalitia
and
and
the
CEO-
and
this
is
a
short
presentation
about
the
technology
that
we
have
been
working
on.
That
starts
from
a
open
source,
but
also
we
are
in
an
observability
conference
right.
A
So
but
observability
is
it's
like
a
trending
topic,
everybody
talks
about
observability,
but
a
few
ones
are
talking
about
how
to
make
it
simple,
because
actually
it's
really
hard
right
and
it's
really
good
and
easy
to
really
fun,
to
observe
and
extract
the
insights
from
the
data
right,
but
to
get
there
there's
a
journey
and
this
journey
at
least
from
the
open
source
ecosystem.
We
started
it
like
many
years
ago
when
we
created
this
project
called
a
fluency
which
originally
was
a
solution
for
monolithic
systems
to
extract
logs
or
move
any
type
of
data.
A
From
point
A
to
point
B
and,
as
I
said
earlier,
this
quickly
evolved
for
a
more
Cloud
native
solution
that
we
call
it
a
fluent
bit
where
today
we
maintain
both
projects
and
with
a
huge
community
and
they're
very
active,
and
this
both
projects
are
hosted
by
the
cncf
and
in
a
graduated
state.
That
means
that
our
production
and
production
are
ready,
and
when
you
talk
about
observability,
there
are
certain
expectations
right
and
in
distributed
systems,
things
get
complex,
do
observability
or
maybe
just
monitor.
A
One
system
monitor
is,
is
not
the
same
as
observability
and
there's
a
certain
expectations
and
things
can
go
around
or
fine,
and
actually
this
is
expectation
that
everybody
has
right.
Your
cluster
is
working
full
of
dashboards
and
everything
is
working
smoothly,
and
actually
it's
not
like
that
right.
There
are
many
struggles
in
the
middle,
but
this
is
where
everybody
wants
to
get.
A
Everybody
want
to
go
it's
here
and
when
we
talk
about
observability,
if
you
think
as
a
if
you
forget
your
background
in
I.T,
you
forget
your
background,
maybe
as
a
developer
or
system
administrator
your
final
goal
as
a
user
that
is
doing
observability.
Your
action
is
trying
to
observe,
try
to
gather
insights
to
take
Better
Business
decisions.
That's
your
final
goal.
The
final
goal
is
not
do
anything
else
right.
Actually,
I
just
knew
few
people
who
really
loves
to
manage
observability,
tooling
right.
A
Everybody
just
want
to
get
the
right
insights
right
away,
but
when
you
get
into
observability,
you
started
with
this
mindset.
I
need
to
get
this
information
to
take
this
kind
of
decisions,
and
then
you
get
into
the
next
step.
Okay,
you
decide
mentally
what
kind
of
a
stack
platform
or
SAS
cloud
service
you're
going
to
use,
and
there
are
many
right
and
the
primary
goal
of
these
platforms
is
allow
you
to
correlate
different
type
of
data
that
come
from
different
sources.
You
can
query
logs,
you
can
correlate
metrics
and
traces
and
there
are
many
flavors.
A
There
are
many
solutions
and
everyone
has
pros
and
cons,
but
your
final
goal
is
observability
and
you
decide
to
take
a
bet
on
one
of
the
solutions
available
in
the
market,
which
is
fine,
but
then
you
get
to
the
next
step,
which
means
okay.
If
I'm
going
to
use
this
platform,
first
I
need
to
take
care
of
how
to
send
the
data
to
this
platform.
So
I
can
do
this
correlations
and
then
I
can
observe
this
data.
A
And
then
you
realize
that
you
are
in
a
distributed
system
with
containers
with
data
come
from
different
places
and
we
have
hundreds
of
developers
that
thinks
differently
about
how
things
should
be
done.
If
you
think
about
standardization
at
this
level
today
we
have
some
initiatives
initiative
like
open
Telemetry.
A
A
A
Calypco
comes
with
a
full
UI
where
you
can
manage
all
your
data
pipelines.
You
can
configure
it,
and
also
we
have
another
option
to
bring
your
own
business
logic
into
the
pipeline,
which
means,
if
you
are
getting
your
security
locks
and
at
some
point,
you're
saying
I,
want
that
all
the
data
that
comes
that
have
any
kind
of
credit
card
transaction
numbers
be
redacted
or
just
mask
it
before
it
hits
my
destination.
A
Yeah
just
go
to
the
UI
perform
three
for
clicks,
modify
the
data
test
it
and
then
you're
ready
to
go
elliptical
for
now
is
free
to
use.
You
just
can
get
started
it
deploys
inside
your
own
kubernetes
cluster.
You
guys
can
use
Helm,
you
can
use
Cube,
CTL
or
actually
also
you
can
try
out
our
new
Docker
desktop
extension.
So
if
you
have
your
own
Docker
desktop
running
in
your
laptop
and
you
want
to
manage,
login
automatically
just
go
ahead
in
select
station
and
you
will
be
ready
to
go
a
escalita.