►
From YouTube: CI WG demo: Southern Crossroads
Description
Date: 1/20/17
Presenter: Russ Clark
Institution: Georgia Tech U
South Big Data Hub
A
A
A
Today
we
operate
as
a
separate
of
501c3
organization,
so
people
ask
so
there's
these
three
terms:
right,
socks,
southern
light,
rail
in
georgia.
Tech's.
Let
me
try
to
demystify
that
the
name
we
all
go
by
socks.
That's
the
sort
of
the
name
are
most
known
by
it
and
what
we
do
we
use
SLR
or
somebody.
Braille
is
the
501c3
organization
under
which
Sox
operates
as
a
project.
A
A
We
are
affiliated
with
Georgia
Tech
as
an
organization.
Basically,
what
that
means
is:
there's
a
research
contract
from
sock
sock
as
a
separate
company
with
its
board
that
independent
of
Georgia
Tech,
but
Sox
has
a
very
few
staff.
If
any
most
of
the
time,
the
project
itself,
it's
is
operated
as
a
contract
with
Georgia
Tech.
So,
for
instance,
myself,
Kathy
Angelo,
the
Sox
executive
director
were
all
employees
of
Georgia
Tech
and
with
with
day.
B
A
Amortize,
the
combined
the
buying
power,
if
you
will
of
the
networking,
are
any
networking
community
to
negotiate
better
terms
with
providers
and
castes
who
is
our
executive
director
is
a
is
a
member
of
the
Executive
Board
to
the
quilt
as
well.
So
just
so
so
here's
the
answer-
and
why
is
rough
here
today?
The
sock,
Smith
print
looks
a
lot
like
the
Celtic
data
of
footprint
and,
of
course,
with
connectivity
and
support
to
institutions
throughout,
and
you
know,
even
if
you're
not
on
this
map,
there's
a
good
chance.
We're
collaborating
with
you
or
have
them.
A
A
A
So
that's
kind
of
the
overview
of
Sox
I
want
to
switch
gears
and
talk
a
little
bit
about
some
of
the
research
activities
that
Sox
is
directly
involved
in
and
working
with,
the
or
institutions,
and
on
on
funding
that
we've
received
directly
in
support
of
Sox
efforts.
The
first
is
routing
security,
then
Software,
Defined,
Networking
and
then
wrap
up
with
smart
cities
and
I
know:
there's
lots
of
interest
in
the
smart
cities
as
a
as
a
direction,
research
direction
and
in
funding
opportunities
there.
So
grouting
security
and,
let's
live
in
a
nutshell
right.
A
A
So
I
used
her
examples
here,
but
basically
this
is
about
taking
advantage
of
the
longer
prefix
matching
that
ipv4
wants
to
do
when
doing
routing
and
the
fact
that
in
this
case
they
a
an
attacker,
can
can
take
effectively.
Take
someone
offline
or
steal
the
traffic
or
or
what
have
you
simply
by
advertising
a
shorter
route
to
your
web
server,
and
so
you
know
not
a
new
problem
working
on
this
for
a
long
time.
A
The
solutions
du
jour,
if
you
will,
the
state
of
the
art,
is
a
rpki
or
research,
public,
key
infrastructure
and
BGP
SEC,
and
it's
really
a
two-step
two-pronged
approach
to
to
addressing
the
issue
and
and
RP
guides
the
precursor
right.
So
I
was
the
first
step.
If
you
will
is,
we
need
to
get
rpki
deployed,
and
then
we
can
can
work
on
BGP
SEC
and
the
efforts
of
the.
But
you
know
what
we're
about
today
really
is.
A
So
look
at
look
at
step,
one
done,
and
in
order
to
do
that
there
there
are
two
key
things:
the
need
to
happen.
First,
the
people
to
hold
an
address
in
this
case
would
say.
Universities
that
the
have
address
was
space
allocated
basically
need
to
take
the
legal
and
and
technical
step
of
asserting
their
ownership
of
their
addresses.
A
These
addresses
and
then
that
the
other
step
is
the
next
parallel
step
that
the
network
operators
need
to
actually
start.
Looking
for
this
stuff
in
you
know,
turn
it
on
in
their
routers
and
start
actually
checking
hey
is
the
the
person
advertising
this
route?
Are
they
the
valid
owner
of
this
realm
or
of
the
yes
space?
A
Those
are
ultimately
required,
but
these
things
really
are
two
different
things
and
really
what
we're
promoting
we're
trying
to
get
the
community
to
understand,
as
well
as
as
some
efforts
that
the
Internet
to
security
community
is
taking
now
really
is
the
you
know.
Both
of
these
things
are
doable.
We
understand
how
to
move
forward
and-
and
we
were
trying
to
work
with
the
community
and
get
this
going
so,
unfortunately,
nothing
can
put
out
challenges.
You
know
why
hasn't
this
already
been
done?
Well,
there's
some
some
legal
stuff
that
has
to
happen
for
address
holders.
A
Nsf
has
funded
us
and
Sox,
and
the
Sox
community
to
basically
work
through
this
and
try
to
get
some
some
movement
on
deployment
and
help
people
document
and
and
and
how
to
move
this
forward.
So
basically,
we
have
some
some
funding.
Today
we
have
some
resources
available
to
you
know:
I
say
for
the
Sox
community,
but
I'll.
A
You
know
I'll
say
to
the
entire
south
big
data
hub
community,
whether
you're
a
Sox
member
or
not,
we're
we're
so
ready
and
able,
and
have
both
the
time
and
and
even
some
equipment
to
help
people
who
want
to
explore
this
and
and
deploy
it.
We
actually
did
a
sort
of
a
proof-of-concept
test
of
all
this
by
building
it
out
on
a
genie
on
the
genie
infrastructure
of
the
as
a
test
case,
stimulating
the
entire
bgp
deployment.
That's
what
that's
depicting,
but
anyway,
to
sort
of
wrap
up
the
rpki
part
of
this.
A
A
A
Our
science
driver
for
this
project
is
the
new
telescope
being
built
in
Chile
the
LSST,
and
you
know
when
we
talk
about
Big
Data.
This
is
going
to
be
some
big
data,
6.66
nap
gigabyte
images.
Basically,
they
need
to
be
transmitted
within
five
seconds
and
well
an
image
created
every
17
seconds
and
with
a
goal
of
getting
it
delivered
in
five
seconds.
A
They
don't
have
to
read
all
the
words:
it's
a
big
telescope
and
it's
going
to
be
down
here
and
we
got
to
figure
out
and
get
the
data
back
up
to
up.
You
know
so
all
the
researchers
around
the
world
that
want
to
interact
with
it
and
control
it,
not
just
it's
not
just
a
one-way
pipe,
but
people
want
to
be
able
to
control.
A
You
know
what
most
scientific
instrument
projects
we're
talking
about.
You
know
being
able
to
make
adjustments
in
real
time
so
that
you
don't
waste
your
time.
You
know
the
precious
time
that
you
have
using
the
instrument,
so
the
the
FBX
part
of
this
is
basically
building
a
international
software-defined
exchange
that
includes
these
three
sites
to
be
then
connected
to.
You
know
the
rest
of
the
world.
If
you
will,
but
you
know
from
a
research
side.
A
Well,
you
know
we're
looking
at
okay,
you
know
how
we
look,
how
we
leverage
at
the
end
across
multiple
management
domain,
so
the
you
know
to
us
at
the
end
of
the
day,
what
makes
this
offer
to
find
exchange
interesting
versus
other
Sdn
projects?
Is
we
have
to
deal
with
the
fact
that
this
isn't
one
network
operated
by
one
entity?
If
it's
international,
there's
multiple
parties,
multiple
players?
A
A
The
the
research
buzzword
of
2016
and
going
into
2017?
Of
course,
one
of
the
biggest
topics
is
around
the
smart
cities.
Work
and
we've
been
very
active
in
that,
not
just
in
Atlanta
and
in
the
region
but
internationally
as
well
and
participating
in
things
like
the
NIST,
smart
cities.
Global
City
teams
challenge,
as
well
as
the
u.s.
A
A
B
Are
you
working?
This
is
my
coherency.
Are
you
working
to
on
the
we?
Have
the
EXO
Genie
guys
downstairs,
who
are
running
the
racks
where
you
can
move
software
around
in
slices
I,
don't
know
if
I'm
describing
it
correctly,
but
that
idea
of
being
able
to
stand
up
you
know
of
computational
pieces
as
well
as
networks
to
create
sort
of
applications
on
demand.
Is
that
is
that
sort
of
playing
in
and
the
architecture
for
the
telescope,
yeah.
A
Absolutely
so
yeah
I
knew
there
were
folks
there
and
I.
Thank
you
for
the
reference,
because
I
didn't
know
how
much
you
guys
know
about
Genie
and
Genie
racks
and
and
the
extra
DD
project.
A
The
the
Genie
rack
effort
is
is
one
of
the
things
that's
influencing
what
we're
doing
in
the
Atlantic
waves,
the
the
telescope
project.
We
we
think
it's
going
to
be
critical,
that
there
are
not
just
you
know,
data,
switching
and
forwarding
infrastructure
at
the
exchange,
but
compute
and
and
some
amount
of
storage
as
well,
and
that
that's
a
part
of
the
story,
and
certainly
the
the
flexible
rack
resources
that
the
genie
the
genie
has
led.
The
effort
with
is
is
a
you
know,
an
important
part
of
that
story.
A
I
I
will
certainly
admit
that
from
a
so
that's
you
know,
this
is
a
that
part
of
it
is
not
nearly
as
well
developed
from
the
point
of
view
of
the
Atlantic
wave
project
is
a
five-year
project
and
we're
about
eighteen
months
into
it
and
we're
very
focused
on
if
you
will
be
getting
the
traffic
flows
in
you
know
as
part
of
it
right
but
I.
Don't
honestly,
I,
don't
see
how
this
works
ultimately
to
meet
the
needs
of
you
know.
A
I
can
do
the
flows
if
I've
only
got
one
customer
right,
but
the
reality
is
we're
going
to
have.
You
know
thousands
of
people
wanting
to
consume
various
and
work
with
this.
The
telescope
data,
as
well
as
you
know,
other
instruments
and
and
moving
that
computation,
but
also
being
smart
about
how
that
data
is
distributed.