►
From YouTube: Lightning Talk: eBPF based UPF (User Plane Function) for 5G - Aapo Poutanen & Eric Lajoie
Description
Lightning Talk: eBPF based UPF (User Plane Function) for 5G - Aapo Poutanen, Cumucore & Eric Lajoie, Red Hat
Network Function Virtualization (NFV) has been used in the past years for efficient deployment of 4G and 5G network functions (NF) on the cloud. However, private industrial mobile networks have different requirements in terms of latency and connectivity. These requirements impact mainly the user plane. Thus, network functions handling the signaling can reside on the cloud but a new solution is required for managing the user plane on-premises to reduce latency and improve reliability. In this talk, we describe a novel design of the User Plane Function (UPF) based on eBPF framework available in Linux OS. Initial implementation and performance results of the eBPF based UPF will be shared during session.
A
Hey
everyone
thanks
for
joining
us
today.
My
name
is
eric
lejoy,
I'm
with
red
hat,
I'm
part
of
the
telco
vertical
within
emea,
even
though
I
have
customers
globally-
and
I
have
my
friend
oppo
here
with
me
today
from
one
of
our
partners
and
we're
going
to
be
talking
about
upf
with
eppf,
and
the
first
thing
I
wanted
to
show
you
guys
is
we're
short
on
time
for
q
a,
but
there
are
two
of
us.
So
if
you
open
that
link
at
the
very
top,
you
can
do
live
questions.
A
Now,
from
my
point
of
view,
in
the
red
hat
side,
we
are
a
key
platform
provider
for
a
lot
of
our
ecosystem,
so
you've
seen
comments
of
openshift.
Today,
kubernetes
there's
been
no
comments
of
open.
Sorry,
I'm
not
allowed
to
say
that
word
here,
but
you
have
seen
open
shift
mentioned
and
in
our
open
shift
we
act
as
a
thin
red
line.
A
So
my
one
minute
of
fame
here
today
is
to
show
you
that
if
every
single
one
of
you
guys
is
a
cnf
of
some
sort
in
a
5g
deployment,
you
could
consider
red
hat
being
this
aisle
down
the
center
that
you're
all
running
off
of
you
get
your
network
from
there.
You
get
your
storage,
you
get
everything
you
need
to
run
your
application
and
to
talk
to
each
other.
So
with
that,
I
want
to
hand
it
over
to
oppo
and
he
can
talk
about
what
he
and
kumacor
have
been
doing
on
top
of
openshift.
C
B
B
B
So
when
designing
the
upf,
the
goal
of
this
talk
is
that
you
can
all
make
your
own
ups
after
this
is
that
clearly
we
can
split
the
interfaces
into
two
there's:
the
control
plane
and
the
user
plan
pfcp
connection
low
packet
rate.
We
can
do
a
normal
user
plan.
Application
for
that
doesn't
require
anything
fancy,
so
here's
the
kumbukor
upf
user
plane
pieces.
B
B
B
The
classification
part
and
the
action
part
that
the
ups
does.
These
pdrs
are
this
packet
detection
rule
ies
that
are
received
from
the
n4
in
the
upf
and
then
the
upf
will
use
these
and
presidencies
and
all
kinds
of
ies
that
are
included
in
this
to
construct
or
detect
the
packet
or
not
detect
the
packet.
That's
the
decision
and
then
the
action
part.
B
And
how
the
kumquat
upf
does
this
is
it
will
construct
a
cache
of
all
the
ideas
that
are
received
and
then
construct
a
flow
and
their
flow
state
is
updated
due
with
every
update
so
like
in
obs,
for
example,
you
have
the
slow
path
and
the
fast
path
and
the
controller
can
install
flows
to
use
the
fast
path
for
packets
that
are
used
more
often.
So
this
is
the
kind
of
what
happens.
B
So
now
we
get
to
the
eppf
application
and
this
is
what
happens
to
an
uplink
packet
a
bit
simplified,
but
in
effect
the
decisions
are
the
same,
there's
very
little
to
do,
because
the
user
user
space
application
that
uses
the
n4
and
constructs
this
flow
diagram
has
already
done
all
this
work.
All
the
flows
are
constructed,
so
the
ebv
application
only
looks
up
and
executes
there's
a
tid
lookup
with
the
sa
qfi
fields
and
everything
packet
modification.
B
Either
there
is
a
gdp
popping
or
there
isn't
gdp
popping
and
then
xdp
redirection
xtp
is
used
for
maximum
packet
packet
rate,
but
you
could
also
use
some
other
hooks
for
eppf,
but
xdp
is
what
we
use
currently
about.
The
drivers.
Yeah
xdps,
xtp
red
hat,
has
implemented
all
the
kernel
backboards
that
are
required
for
vf.
For
example,
it
is
much
more
difficult
to
do
that
in
mainline
kernels,
but
with
red
hat
team
as
backing
us,
it's
easy
virt
io
also
we
use
that
hasn't
been
an
issue
but
other
other
ones.
B
B
So
when
you
deploy
this
in
the
cloud,
these
two
cases
are
what
are
most
common
for
me.
As
the
hardware
passthrough
super
simple,
you
have
a
hardware
device
that
is
for
the
upf,
follow
that
to
a
container
or
a
vm,
and
then
you
can
attach
the
xtp
application
to
that
one
in
the
virtualization
environment.
B
B
Also
when
the
bbf
applications
become
more
available,
you
can
chain
them
together
from
different
vendors.
If
you
want
some
feature
from
other
render,
you
can
use
that
and
like,
for
example,
our
upf
and
some
other
other
guys
tunneling
solutions
on
top
of
it
and
then,
as
we
are
like
a
small
networking
company,
we
can
implement
stuff.
You
want
so
just
tell
us
what
to
do
and
we'll
do
it.
B
C
One
two:
three:
okay,
perfect
works.
First
of
all,
thanks
for
a
very
good
presentation,
thank
you
enjoy
it.
I
have
two
questions
because
I
think
that
upf
in
eppf
is
a
holy
grail
of
the
industry.
The
question
is
how
you
are
solving
things
like
lawful
interception,
so
that
would
be
the
first
one.
Second,
one
how
you
are
doing
the
things
like
header
enrichment,
all
this
nasty,
dpi,
stuff,
etc
right,
because
the
training
itself
so
like
stripping
gtp,
headers
and
just
doing
a
three
two
and
six
is
pretty
much
essence
of
the
upf.
C
B
Yeah
I
can
answer
that
in
a
way
that
how
it's
done
and
how
I
would
like
to
see
it
done
how
it's
done
now
is
just
chaining,
vms
or
containers
together.
So
the
upf
will
then
forward
the
traffic
to
another
nf
which
will
do
this,
but
in
the
future.
How
I
want
to
see
it
done
is
to
do
it
in
xdp,
so
we
can
do
dynamic
flow.
Steering.
C
B
C
Is
about
the
future
because,
maybe
maybe
you
can
discuss
it
offline?
Is
it
really
feasible
to
put
such
a
complex
stuff
like
dpi
or
handling
enrichment
or
lawful
interception
in
a
small
sort
of
because
ebtf
programs
tend
to
be
should
be
actually
small
and
packaged
right?
The
question
is
if
a
huge
upf
is
something
which
can
really
fit
into
ebpf,
and
I
think
that
that
will
be
a
sort
of
event
discussion,
king
of
even
discussion,
because
I
think
that
this
is
one
of
the
critical
topics
in
the
industry
right
now,.
B
Yeah,
if
you
have
any
ideas
or
questions
there
is
my
contact
on
the
top.
Send
me
an
email.
We
can
discuss
it,
but
I
can
shortly
address
that,
like
ebpf
can
be
used
to
split
traffic,
it
can
be
used
to
steer
traffic
to
other
destinations
like
normally.
If
you
do
dbi,
you
have
some
kind
of
mirroring
somewhere,
so
you
can
use
bpf
for
that
and
then
you
can
program.
B
Do
the
exact
dpi
you
want
for
the
bpf
application
and
then
it
might
touch
only
very
small
amount
of
the
traffic.
So
that's
doable.
But
if
you
want
to
do
something
heavy,
then
of
course
you
I
think
you
need
another
environment
for
that,
so
you
need
steering
and
then
the
dpi,
that's
my
take
I
maybe
in
the
future.
It's
done
differently.
A
And
then
there's
the
platform
angle
right.
So
if
you
have
one
server
at
the
edge,
that's
acting
as
a
slice
for
your
5g
core,
let's
say
and
you
need
to
do
lawful,
intercept
there.
Are
you
really
going
to
do
that
at
the
cost
of
cpu
cores
in
that
small
little
limited
resource,
or
do
you
already
have
optical
taps
sitting
there?
So
it's,
I
think
in
my
view,
the
future
is
going
to
be
a
mix
of.
A
A
D
E
E
B
B
E
A
So
gargay
long
time,
I've
seen
that
name
in
the
past.
He
has
a
question:
how
do
we
deploy
an
ebbf
app
on
top
of
openshift?
So
today
it's
a
little
bit
of
a
work
around
and
hopefully
eventually
we'll
have,
I
would
say,
a
smoother
way
of
doing
it
with
the
kernel.
But
today
I
believe
you
you
were
directly
on
the
worker
nodes.
Doing.
B
Yeah,
but
currently
in
openshift
in
ovn,
this
cni
thing
as
long
as
you're
using
vf
as
the
virtualized
nic.
You
can
do
it.
There
used
to
be
this
limitation
in
the
linux
that
you
need
to
mount
an
xtp
application
on
both
ends
of
vf,
but
that
is
no
longer
the
case
if
you're
running
the
new
patch,
which
has
been
backported
for
red
hat,
some
some
version-
I
don't
remember
exactly,
but
we
can
get
back
to
you
if
you
want
to
know
that
yeah.