►
Description
Keynote: SDI: Delivering the Efficiencies of the World’s Largest Clouds to Everyone - Jonathan H. Donaldson, Vice President, Data Center Group General Manager, Software Defined Infrastructure Group, Intel
Software Defined Infrastructure is now capable of delivering the efficiency and agility once available only to the world’s most advanced hyperscale data centers. Combined with the portability and density gains that containers and orchestration bring to applications, these advancements are a bellwether for widespread hybrid cloud adoption. In this talk, attendees will learn how Intel is accelerating the adoption of software defined cloud environments by making solutions easier to deploy and manage with Kubernetes, and how the CNCF Community Cluster is helping organizations of all sizes experience the many benefits of cloud native architecture at scale.
About Jonathan Donaldson
Jonathan H. Donaldson is Vice President in the Data Center Group and General Manager of the Software Defined Infrastructure Group at Intel Corporation. He leads the team responsible for defining and carrying out Intel's strategy for private, hybrid and public cloud automation. Donaldson joined Intel in 2013 from VCE, a joint venture formed in 2009 by Cisco Systems, EMC Corporation, Intel and VMWare.
Join us for KubeCon + CloudNativeCon in Barcelona May 20 - 23, Shanghai June 24 - 26, and San Diego November 18 - 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.
A
So
now,
for
something
completely
different,
all
right,
I,
don't
have
a
demo.
Unfortunately,
following
Kelsey
I,
wouldn't
attempt
to
do
a
demo
anyway,
because
I
can't
recover
nearly
as
well
as
you
can
when
it
fails
miserably.
So
you
know
in
full
disclosure.
This
is
normally
kind
of
a
sponsor
type
session
right
and
when
we
kind
of
sponsor
these
events-
and
normally
people
get
up
here
and
say,
hey
here's
how
you
can
you
know,
buy
my
stuff
or
invest
in
me
really.
A
A
So
without
further
ado,
we
kind
of
talked
about
things
and
software-defined
infrastructure,
it's
actually
in
the
title,
I
guess
on
the
previous
slide,
for
the
things
that
I
do
we
have
a
lot
of
customers
right,
probably
well,
not
probably,
but
we
know
more
customers
even
in
Google
does,
and
so
we
get
a
lot
of
feedback
from
them.
All
of
these
customers
saying
hey
here's
the
problems
that
we're
trying
to
solve
and
how
can
we
go
solve
that
and
so
the
way
to
go
solve
that
is
through
what
we
call
suffered
a
fine
infrastructure.
A
A
You
know
we
don't
like
a
downtime
very
much,
because
that
looks
bad
for
all
of
us,
and
so
can
we
get
to
you
know
the
mythical
you
know
569
of
performance,
always
you
know
uptime
type
of
concept,
so
we
kicked
off
what
was
called
the
cloud
for
all
initiative,
which
was
kind
of
the
umbrella
investment
that
we're
doing
in
things
like
cloud
native
compute
foundation,
the
oci
commodities
itself.
We
do
a
lot
of
work
around
OpenStack
we've
done
work
in
Apache
basis.
Who
knows
that?
A
We
have
about
four
to
six
thousand,
depending
on
how
you
count,
engineers
that
do
open
source
software
that
helped
drive
these
these
types
of
things
forward,
so
I'm,
just
the
kind
of
the
latest
representative
up
here
talking
about
these
things,
but
we
really
think
that
driving
this
type
of
innovation
right
having
these
types
of
communities
and
really
working
on
these
types
of
technologies
is
really
powerful
and
very
helpful
for
us
all.
So
it's
not
just
you
know
hey.
A
We
can
go,
do
some
marketing
events
and
we
can
sponsor
these
things,
one
of
the
biggest
things
when
we
started
talking
to
David
and
we
started
talking
to
Craig
when
he
was
there
at
Google
and
when
we
started
talking
to
the
cloud
native
compute
foundation,
or
so
it
was
like
hey.
What
can
we
actually
do,
materially?
That
will
help
push
Cooper
today's
forward
and
the
first
thing
we
got
is
well:
don't
don't
throw
more
developers
at
us
right
we've.
We've
got
lots
of
that
right.
A
We
actually
need
to
figure
out
kind
of
how
we
can
go.
You
know
wrangle
all
of
the
development
that's
coming
in
right.
How
do
we
go
make
sure
that
the
code
is
really
stable
and
that
the
projects
that
we
want
to
have
attached
to
these
types
of
cloud
native,
compute
deployment
methodologies
right
are
really
you
know
valuable
to
everyone,
and
so
we
said
well,
we
kind
of
have
hardware.
A
A
Redhead
openshift
has
done
a
lot
of
performance,
tuning
and
scaling
for
communities
on
top
of
it
we're
doing
things
like
there's
a
Zika
virus
thing
going
on
with
boink
right
now
to
use
the
cluster
and
drive
up
the
performance,
so
we
can
kind
of
see
how
things
happen
at
scale
and
all
of
these
things
start
to
be
handed
back
into
the
community
right.
Usually
that
is
actually
one
of
the
major
criterias
for
being
on
the
cluster.
Right
is
hey,
it
has
to
be
open
source
and
you
have
to
contribute
back.
A
Those
contributions
back
into
the
community
can
either
be
write-ups
right.
Here's
what
we
found,
here's
the
caveat!
So
here's
how
we
did
it
or
they
can
be
bug,
fixes,
right,
PRS,
acceptor,
etc.
We've
got
11
in
the
queue
today
right.
It's
really
easy
to
get
access
to
the
cluster
itself.
You
just
create
an
issue
at
CN,
c,
f,
/,
cluster
right,
we're
bringing
we're
trying
to
come
to
you
to
gain
access
to
these
things.
A
The
other
important
thing
about
this
is,
is
you
know
in
the
next
500
nose
goes
online
and
and
q1,
so
our
total
investment
and
a
thousand
nodes-
these
are
big.
Big
beefy
machines
will
be
about
20
million
dollars
by
the
end
of
this
week,
and
so
the
idea
is,
is
we
want
you
to
be
able
to
leverage
these
systems
free
of
charge
right,
and
we
only
want
you
to
just
drive
innovation.
You
know,
drive
the
scale
drive.
A
The
performance
drive
the
things
that
people
need
to
operationalize
these
types
of
deployment
methodologies,
and
so
with
that
right,
the
other
thing
is,
you
know
we
who
here
operates
clouds
right?
Okay,
good,
so
you
know
it's
a
it's
hard
to
do
right
now.
We
are
learning
our
way
through
this
right.
As
a
team
we've
got
about
three
people
dedicated
to
this
and
we're
offering
something
slightly
different
than
most
people
do
on
this
system
is
bare
metal.
Complete
bare
metal
will
give
you
ipmi
access.
A
If
you
want
it
right
and
multi-tenancy,
I
pain
in
my
axis
is
very
difficult,
so
we're
working
through
all
the
bugs.
So
if
you
get
on
the
system,
you
have
a
problem.
Let
us
know
about
it,
preferably
not
up
on
stage
email
works.
Just
fine
and
we'd
like
to
see
you
guys
actually
come
in
here.
The
other
thing
that
we're
opening
up
I
think
Dan
announced
it
recently
is
applications
right.
A
So
who
else
here
in
the
early
days
right
started,
you
know
of
Linux
when
you
were
really
kind
of
starting
to
deploy
applications
right
when
you
had
to
figure
out
well,
how
do
I
do
DNS
on
Linux
how
to
do
dhcp
on
Linux
right?
You
went
to
some
of
the
things
like
source
for
it
or
not
sourceforge,
but
some
of
the
different
wiki
sites-
and
you
know
there
were
really
really
nice
write
ups
for
here's.
How
use
the
packages
you
pull
down?
Here's
the
configuration
files,
here's
what
everything
means!
A
A
How
do
we
get
more
people
driving
this
type
of
deployment
methodology
and
so
we're
opening
the
cluster
up
for
access
to
any
open
source
project
that
is
willing
to
come
in
and
container
eyes
are
applications
deploy
them
into
the
cluster
using
communities
and
any
of
the
other
projects
that
are
part
of
CNC
f
as
long
as
they
then
come
back
and
do
a
write-up
for
how
they
did
it?
What
they
found,
how
it
can
be
better
next
time
so
we
have
also
entered,
did
significantly
into
communities
volkan
significantly
as
a
start
right,
it's
getting
better.
A
A
So
again,
very
very
short:
I,
don't
have
a
whole
lot
of
things
to
demo
up
here
today,
you
can
get
on
our
cluster,
create
an
issue
if
you'd
like
it
like
to
be
on
there
we'd
love
to
have
you
and
with
that
I
just
want
to
thank
everyone
right.
This
is
amazing
to
see
when
Craig
and
myself
and
a
few
of
the
other
board,
members
of
the
CNC
f
board
kind
of
got
this
thing
kick-started.
We
had
no
idea
that
it
would
grow
this
fast
with
this
much
quality
right.
That
is,
that's
probably
the
biggest.