►
Description
Join us for Kubernetes Forums Seoul, Sydney, Bengaluru and Delhi - learn more at kubecon.io
Don't miss KubeCon + CloudNativeCon 2020 events in Amsterdam March 30 - April 2, Shanghai July 28-30 and Boston November 17-20! Learn more at kubecon.io. The conference features presentations from developers and end users of Kubernetes, Prometheus, Envoy, and all of the other CNCF-hosted projects
How Linux Took Over the World of Finance - Christoph H Lameter, Jump Trading LLC
Linux has become the dominant operating system not only in high performance computing but also in the world of finance. Stock exchanges, Banks, Investment funds and "Quants" use Linux to implement complex trading strategies. This talk presents the history of how it happened and what key technology issues drove the adoption of Linux in the Industry and are driving it today. Competition in the industry leads to technology innovation and causes technology to be taken to extreme ends. Now the industry approaches the limitations of physics. The speed of light is limiting trading latencies. The hardware has been optimized to extremes but it is still running Linux. Calculations and deep learning dominate the landscape and the machines get ever intelligent. And its still all Linux.
https://sched.co/Nru5
A
Okay,
so
this
is
about
the
history,
how
Linux
came
to
be
used
in
the
world
of
finance
and
that's
kind
of
a
long
story
and
we're
not
talking
about
some
780th
year.
This
is
some
it's
of
course,
of
course
used
in
the
finance
industry,
but
these
are
specialized
arrangement
that
are
made
to
interface
with
exchanges.
A
A
It
consists
of
the
stock
market
where
they
use
Linux
to
match
the
trades,
and
there
are
special
Kwan's
of
their
investment
banks
that
remain
complex
trading
surgeries
and
try
to
make
money
out
of
the
customers
and
investment
banks
or
of
trading
on
the
stock
markets
like
the
quant.
Do
so
then,
there's
a
classic
stuff
like
customer
relation
management,
supply,
chain
management
and
all
the
other
things.
A
These
are
also
all
migrated
to
Linux
these
days
and
so
I'm
going
to
talk
a
bit
about
the
history,
why
these
decisions
were
made
and
how
the
industry
has
evolved,
and
so
we
have
a
variety
of
participants
and
these
in
the
area
of
Finance,
the
especial
of
the
stock
exchanges.
The
stock
exchanges
get
fees
from
the
companies
who
trade
on
the
stock
market
and
they
want
to
provide
an
equal
playing
field
for
all
the
investors
that
they
all
have
equal
access
and
equal
chance
to
buy
all
these
symbols.
A
So
their
interest
is
not
so
much
in
efficient
high-speed
trading,
but
is
in
fair
trading,
and
then
we
have
the
banks.
The
banks
do
trade
for
their
investment
portfolios
and
on
behalf
of
customers
and
their
main
entry
interest
is
to
get
the
best
price.
And
typically
the
banks
have
a
price
differential
between
what
the
customer
pays
and
what
they
pay
on
the
stock
market,
and
they,
the
interest,
is
to
have
the
most
profit
on
that
level.
And
today
the
corporate
giants
like
Google,
Facebook
and
so
on.
A
They
all
have
these
days
integrated
finance
operations
that
also
Dera
interact
with
the
exchanges
and
mostly
that
is
used
for
international
trading,
as
well
as
to
preserve
the
value
of
their
stocks
and
to
manage
the
wealth
that
they
are
accumulating
and
managing.
That
Wilbanks
is
something
that
I
wouldn't
be
doing.
They
wanted
and
they
want
to
have
their
own
departments
that
that
deal
with
this.
A
Then
we
have
small
investors
that
have
some
software
on
the
machines
at
home
and
they
interact
with
the
exchange
and
they
think
they
have
a
special
of
trading
orders,
but
usually
they
have
some
Manish
knowledge
where
they
can
make
money
of
these
things
and
their
interest
is
to
have
fast
access
to
the
Internet
and
be
competitive
with
the
big
guys,
and
there
are
professional
clients
quants.
Those
people
have
a
lot
of
money
and
they
have
sophisticated
operations.
A
Usually
they
have
equipment
that
is
located
at
the
stock
exchanges
and
they
want
to
extremely
low
low
latency
and
they
want
to
have
a
edge
about
everybody
else,
and
so
these
things,
that's
in
a
race
between
all
of
these
to
get
the
best
prices
and
typically
I.
Think
the
the
professional
Kwan's
win,
because
that's
why
they
make
the
money
and
they
and
typically
the
other
ones,
try
to
come
close
to
them,
but
they're
not
willing
to
make
the
investments
that
the
cons
do
to
do
to
get
that
low
latency.
So
where
does
this
happen?
A
The
most
important
exchanges
is
Chicago
and
New
York
London,
Frankfurt
and
Shanghai.
Now
it's
gets
bigger
and
bigger.
So
there's
not
a
lot
of
professional
operation
starting
up
here
and
a
lot
of
the
classic
entities
from
the
rest
of
the
world
is
trying
to
enter
China
now
and
also
do
these
trades
here.
A
So
how
it
does
actually,
this
is
actually
work
if
you
have
a
stock
exchange.
The
stock
exchange
has
a
matching
engine.
There
are
the
orders
of
the
buyers
and
sellers
flow
together
and
they
are
matched,
and
when
that
is
done,
a
trade
is
occurring
and
if
it
trade
has
occurred,
then
the
fact
that
the
vet
that
affected
the
fact
that
the
trade
has
occurred
and
has
occurred
as
well
as
there
may
be.
The
change
of
the
value
of
a
stock
symbol
is
spread
by
the
market
data
distribution
mechanism.
A
So
and
the
technology
sophistication
of
these
various
entities
is
different,
and
so
because
there
are
different
objectives
here
that
they
had
also
the
stock
exchange,
as
I
said
before,
since
we
are
small
groups,
if
you
don't
want
to,
if
you
don't
understand
something
or
have
a
question
just
speak
up,
it
may
be
the
best.
If
I
talk
to
first
tell
me
that
you.
A
So
what
can
technologies
are
being
used
here
and
you
didn't
have
intelligence
and
speed
for
intelligence?
You
can
use
something
like
kubernetes
or
an
HPC
cluster
to
do
data
processing,
and
then
you
need
to
at
the
exchange
you
if
you
connect
it
to
the
exchange
and
you
are
in
this
closed
loop.
You
want
to
react
as
fast
as
possible
because
you
want
to
have
an
advantage
and
catch
the
chance
of
making
it
trade
by
being
faster.
A
They
usually
have
very
high
capacities
in
the
fast
LAN
and
also
some
people
correlate
these
symbols
within
multiple
exchanges.
So
they
have
high
speed
white
area
connections
between
different
cities
where
they
can
get
information
faster
to
the
other
point
and
therefore
react
faster
if
it's
simple
changes,
for
example,
in
New
York
and
it
effects
will
affect
something
in
Chicago
and
so
what
they
also
do
is
they
want
to
write
very
efficient
code?
A
So
chip
needs
very
much
stripped
down
bare
metal,
it's
very
limited
and
just
focused
on
this
one
purpose
and
it's
extremely
optimized,
and
so
there's
a
lot
of
hot
rolling
going
on
in
that
area
and
a
lot
of
kernel
bypass.
There
are
interface
with
the
hardware
and
so
and
so
and
such
and
such
to
increase
the
speed
and
gain
an
advantage
on
the
other
participants
in
this
trading.
Loop.
A
A
We
had
ms-dos
first,
the
first
pcs
came
about
then
Novell
ball
created
the
first
local
area
network
and
then
son
and
proprietor
UNIX
is
showed
up
and
they
were
used
by
the
stock
exchanges,
as
well
as
by
the
traders
to
do
the
initial
trades,
which
were
very
slow
and
so
in
the
90s.
Also
the
ten
immiscible
were
created
is
a
specialized
dual
safe
computing
systems
that
are
still
being
used
today
in
some
esoteric
configurations
because
they
are
very
secure.
A
For
example,
they
created
a
message
API
for
this
multicast
distribution
for
the
market,
a
distribution,
that's
used
by
the
stock
market,
and
so
Microsoft
became
a
good
platform
to
use
for
training
for
trading
and
in
the
round
2000
the
Sun
systems
and
proprietary
UNIX.
This
became
very
big
in
the
trading
area
and
they
took
over.
The
next
was
used
at
the
fringes
for
some
side
processing,
but
nobody
retested
that
Lindos
could
do
the
job
and
there
were
some
fierce
discussions
and
disagreements
in
the
tech
department.
A
If
you
could
discussing
if
a
Windows
system
is
better
or
the
Linux
and
vice-versa
and
there's
histories
of
how
our
tech
department
being
fired,
because
then
owners
got
mad
at
them
and
horror
stories
from
that
area.
But
in
2007
2009
the
walls
were
won
by
Linux
and
we
saw
from
that
time
on.
We
saw
widespread
adoption
of
Linux
as
the
trading
OS
of
choice.
A
A
A
This
is
our
DMA
API,
which
allows
direct
memory
to
memory
transfer
between
different
systems
in
a
network
that
was
used
to
reduce
the
latency
s
and
to
avoid
the
high
latency
of
the
network
stack,
then
some
people
start
using
real-time
kernels
because
they
felt
they
were
more
efficient,
which
caused
some
discussion
because
actually
the
floor,
and
then
the
term
real
time
was
really
find.
And
now
we
have
a
different
notion
of
real
time
in
the
kernel.
It
actually
is
faster,
but
it's
not
really
the
academic
definition
of
real
time
anymore.
A
Then
they've
found
out
the
calcium
as
a
binaries
as
well,
so
special
compilers
learning
used
handcrafted
assembly
code
in
user
space,
and
then
there
was
a
discussion
about
the
methods
of
languages
to
use.
First,
there
was
a
bar
and
use
basic,
even
in
the
early
phases
and
the
universal
agreement
in
the
industry.
After
all
these
years
is
that
C++
is
the
best
one,
because
you
can
load
have
a
code
in
this,
and
most
of
the
code
fails
that
I
today
are
now
on
C++
and
the
industry.
There
were
numerous
Network
enhancements
being
made.
A
One
of
these
problems
of
the
industry
is
the
speed
of
light
problem.
You
want
to
be
fastest,
but
it's
a
physical
limitation.
Oh
how
fast
it
can't
communicate
information,
and
so
the
problem
is
here
that
you
cannot
really
go
faster
than
the
speed
of
light
and
so
there's
a
limit
to
where
you
can
go,
and
there
are
various
measures
to
get
you
closer
to
that
limit
and
the
closer
you
go,
the
more
specialized
hardware.
You
need,
the
more
sacrifices
you
may
make
in
terms
of
cost.
A
So
there's
a
this
exponential
curve
as
you
go
closer
to
it,
and
so
it's
always
a
depends
on
how
much
money
you
want
to
spend
to
add
user
latency.
It
takes
about
hundreds
of
milliseconds
to
send
a
signal
around
the
earth
that
is,
for
example,
the
problem.
You
can't
simultaneously
update
things
globally
and
in
today
the
trading
cycle
of
an
exchange
is
much
less
than
a
millisecond,
so
it
takes
hundreds
of
milliseconds
to
go
around
the
globe.
A
A
So
if
you
can't
have
an
advantage
prior
to
somebody
else,
you
can
make
a
lot
of
money,
and
so
that's
this
led
to
various
optimizations,
making
the
fiber
straight,
not
using
fiber
at
all,
because
five
hours
slow
slower
than
the
signal
over
the
air
and
there's
rumors
of
people
trying
to
use
blimps
to
bounce
the
signal
off
it.
Some
people
have
been
trying
to
use
the
moon
to
bonds
of
weird
things.
Some
people
discussing
neutrinos
to
shoot
through
the
earth
a
detective
on
the
other
side.
A
A
So
and
we
have
a
transition
here
from
manual
to
automated
trading
manual,
takes
hours
to
arrange-
and
now
we
have
many
seconds
inside
like
that,
but
this
is
probably
too
much
so
and
too
slow.
I
have
not
have
much
time
left.
So
the
how
why
was
in
linux
used
for
this?
Their
suspect
GP
depends
whoever
gets
at
first.
So
you
must
be
able
to
control
your
check
if
you
can't
control
your
check
and
use
third-party
tools
and
they
make
a
change
that
will
increase
the
latency.
A
A
A
So
that
also-
and
then
we
have
two
problems
with
the
highlight
in
the
window.
That
looks
darker
that
caused
the
move
to
Linux,
because
there
was
a
lot
of
people
analyze
the
latency
of
trading,
and
they
found
that
they
had
high
lightness
that
were
saying
they're.
Also
phone
high
latency
is
in
the
Linux
network
stack,
but
these
guys
got
active
and
contributed
to
who
the
Linux
Network
stack
changed
it.
A
So
suddenly
the
network
was
better
and
Windows
didn't
change,
so
that
kind
of
dynamic
caused
a
pretty
energetic
move
to
linear
and
Windows
Microsoft
rarely
updated
that
so
after
some
time
frame,
the
differentiation
became
a
factor
of
10
also,
and
it
was
not
possible
anymore
for
anybody
any
reasonable
person
to
use
that.
So
the
customer
didn't
hear
of
Linux
one
out
and
the
other
thing
is
most
Internet
services.
A
And
then
we
get
this
adoption
of
the
exchanges
of
Linux
and
that's
also
them
cost
a
lot
of
the
hang
or
left
rough
companies
are
still
hanging
on
to
convert
the
system
pseudonyms
because
they
saw
ok,
the
exchanges
are
doing
it
and
they
feel
it
safe
and
so
I
can
now
also
use
my
Linux
systems
and
one
of
the
last
competitors
out.
There
was
some
Solaris
and
some
of
the
exchanges
held
out
to
the
very
end
for
this,
and
but
after
Oracle
was
bought
son.
A
That
was
a
sign
for
the
in
ok,
they
lost,
they
are
not
properly
and
all
that
would
be
going
away.
And
at
that
point
we
saw
the
very
fast
adoption
of
Linux
for
the
rest
of
it,
and
these
son
systems
were
kicked
out
very
fast
because
everybody
felt
they
had
a
liability
on
their
hand
and
they
felt
that
that
system
enano
stable
and
could
be
subverted
by
any
change.
That
Orca
would
do
any
minute
now.
So
the
control
issue
was
there
with
Linux.
They
had
control
over
the
source
code.
A
They
had
reliable
operation,
there's
nobody
who
could
take
it
over
and
mess
around
with
it
and
there's
still
some
minor
uses
of
ax
and
some
BSD
is
in
the
industry
in
some
niche
corners,
I
think
and
that
may
not
be
going
away,
but
they,
the
trust
that
people
have
of
IBM
is
much
higher
than
they
have
of
son
and
BSD
is
also
open.
Source
and
people
also
have
no
control
issues
with
their
because
he
has
a
source
code
as
well.
So
this
remains.
A
And
then
there
are
various
real-time
versions
of
both
that
I
use
mainly
by
the
exchanges.
Some
have
been
using
Gen
2
because
with
Gen
you
can
recompile
your
binaries
and
you
can
make
them
smaller.
Thus,
you
have
the
benefit
of
caching
effects
and
the
binaries
will
run
faster
if
you
just
remove
other
features
that
you
don't
need
and
so
that
this
was
observed
by
various
exchange.
The
various
companies
and
Gen
2
is
very,
is
a
favor
of
the
industry
and
then
there's
Ubuntu
and
Debian.
A
The
reason
for
that
is
to
evade
the
commercial
entities
like
Greta
and
Susie,
because
at
that
point
you
interact
with
the
last
company
and
the
you
at
the
whims
of
the
large
company
right
with
Ubuntu
and
Debian.
You
have
a
much
wider
availability
of
all
source
code
and
you
have
their
own
influence
over
the
release
of
it.
If
you
can
get
hold
of
a
maintainer
or
if
you
hire
a
maintainer,
so
Ubuntu
and
Debian
is
prevalent,
especially
in
embedded
systems.
A
Debian
is
very
prevalent
and
maybe
bought
by
a
very
small
system.
That's
dedicated
to
a
certain
function.
Debian
is
mainly
used,
so
the
challenges
for
them
from
finance
open
source
completely
contributes
area.
Everybody
thinks
that
they
have
the
secret
sauce
and
it's
very
valuable
and
they're
not
giving
it
away.
So
if
they
hack
the
Linux
kernel,
they
will
first
make
sure
that
they
can
keep
it
advantage.
They
don't
want
to
publish
it.
A
The
technique
that
I
hope
develop
is
ok,
let's
hold
it,
but
if
you
see
the
competitors
trying
to
make
a
move
to
change
it
and
O's
kernel
on
that
level,
you
dump
your
stuff
in
there
and
you
show
it's
better
and
you
kill
the
change
that
the
competitor
wanted
to
do,
because
now
he
needs
to
change
the
software.
No,
you
don't
need
to
change
your
mind,
so
the
involvement
in
option
development
is
clearly
matures
and
most
of
the
developers
are
constantly
seeing
regressions
of
various
kernel.
Covenants.
A
A
A
B
A
B
A
The
bet
was
factually
incorrect.
He
thinks
that
somebody
front
one
ran
him
know
they
are
very
intentional
on
systems
on
the
network
that
observe
all
trading
participants
and
he
tried
it
every
day
at
ten
o'clock,
and
this
is
new
that,
after
a
couple
of
times
that
he
always
straight
to
ten
o'clock
and
it
prepared
the
market
for
him
so
that
he
had
the
most
most
expensive
thing
at
that
point.
A
So
he
got
paranoid
about
a
neural
network
that
observed
him
and
he
was
suspecting
that
people
had
advantage
of
him,
but
I,
don't
think
he
got
the
thing
right.
He
it
was
an
investment
banker.
An
investment
banker
think
they
can
do
what
the
traders,
what
the
quants
do,
but
the
crit
funds
know
much
more
than
they
do,
and
so
this
was
written
by
banker
and
he
had
some
limited
knowledge
on
what
was
going
on
the
beckysan
soliders
exchange.
A
C
Yeah,
you
just
mentioned
that,
and
the
financial
company
ID
people
stick
to
the
old,
colonel,
so
I'm
just
wondering
how
they
saw
their
security
issues,
because
you
know
most
of
latest
colonel
has
the
many
important
security
patches.
So
if
they
just
stick
to
the
load,
colonel
and
you,
you
know
until
they
back
port
the
secret
purchased
by
themselves.
No,
they.
A
Don't
care
about
that
because
the
swelling
segment
I
showed
here
is
it
in
the
beginning
there.
This
whole
cycle
is
in
a
private
world
where
the
public
cannot
access.
This
is
highly
regulated
and
these
guys
paid
out
of
money
and
for
this-
and
none
of
this
is
accessible
from
the
internet,
so
the
exposure
to
risk
is
very
low
and
at
that
this
boundary
the
exchange,
as
well
as
the
trade
of
the
protocol.
Everything
that's
going
on.