►
From YouTube: IBM Blockchain Platform - Experiences in Creating Blockchain Solutions - Gari Singh, IBM
Description
Come hear from the CTO of the IBM Blockchain Platform about his real world experiences in creating several blockchain solutions. Learn how the two blockchain pioneers, IBM and FX Giant, CLS, have come together with 9 other financial institutions to create LedgerConnect, a set of blockchain based financial industry solutions.
https://hgf18.sched.com/event/IaLx/ibm-blockchain-platform-experiences-in-creating-blockchain-solutions-gari-singh-ibm
----
Hyperledger – Open Source Blockchain Technologies
Hyperledger is an open source community focused on developing a suite of stable frameworks, tools and libraries for enterprise-grade blockchain deployments.
It serves as a neutral home for various distributed ledger frameworks including Hyperledger Besu, Fabric, Sawtooth, Indy, as well as tools like Hyperledger Avalon, Cactus and libraries like Hyperledger Aries, Ursa. Learn more about Hyperledger projects: https://www.hyperledger.org/use
Case Studies: https://www.hyperledger.org/learn/case-studies
Training & Certification: https://www.hyperledger.org/learn/training
Tutorials: https://www.hyperledger.org/use/tutorials
Webinars: https://www.hyperledger.org/learn/webinars
Events: https://www.hyperledger.org/events
Vendor Directory: https://www.hyperledger.org/use/vendor-directory
Subscribe to the Hyperledger Newsletter: https://www.hyperledger.org/newsletter
Follow-us on Twitter @Hyperledger
Learn about Hyperledger Membership: https://www.hyperledger.org/about/join
#Hyperledger #HyperledgerGlobalForum2018 #Blockchain
A
So
so,
thanks
for
coming
and
I
guess
we'll
get
started
here.
So
I'm
gonna
talk
about
a
couple
of
things
today,
a
little
bit
just
about
what
we've
been
doing
from
an
IBM
perspective
on
blockchain
in
terms
of
our
work
with
sort
of
fabric,
what
we're
offering
from
a
platform
perspective
and
then
that'll
kind
of
lead
into
a
couple
of
customer
use
cases
and
some
kind
of
what
I
call
I
think
I
call
them
key.
Takeaways
from
you
know
those
engagements
right
things
that
we
actually
learned
I
think
we
have
35
minutes.
A
So
I'll
try
to
do
this
in
like
20
25
minutes,
so
we
can
at
least
take
some
take
some
time
for
some
questions
and
just
as
by
quick
way
of
introduction,
you
can
read
that
so
I'm
Gary
Singh
CTO
of
blockchain
at
IBM,
I'm,
actually
I,
was
just
laughing
with
our
no
I'm.
Actually,
ironically,
based
out
of
Boston,
which
were
in
that
room.
My
job
is
so.
A
So
you
know
I,
think
overall
I'm
not
sure.
Sometimes
people
I
think
people
associate
into
their
hyper
ledger.
There
was
hyper
ledger
fabric,
I,
think
people
know
IBM
work
on
hyper
ledger
fabric,
that's
great!
We
do
a
lot
of
work
in
open
source,
and
that
really
found
you
know
is
the
core
of
what
everything
we
do
on
the
blockchain
side
is
actually
based
on,
but
probably
in
trouble
for
saying
this
I
mean
we're.
Not
you
know,
I
think
we
do
have
a
strategy
of
not
just
everything
being
you
know.
A
We
believe
that
all
the
code
needed
to
be
an
open
source
right.
We
think
that
all
blockchain,
especially
if
you're
gonna,
have
trust
commitments
et
cetera
that
actually
be
done,
an
open
source
that
there
should
be
a
community
around
it,
that
we
want
to
have
a
governance
model
and
obviously
we
want
to
have
as
much
participation.
You
know
as
possible
from
many
companies,
and
you
know,
we've
got
a
number
of
big
companies
that
participate.
A
You
know
some
that
we
can
be
considered
on
the
business
side
competitors
and
if
it's
small
companies
as
well
all
contributions
are
always
welcome.
So
what
do
we
do
from
that
perspective?
So
we
obviously
leverage
and
engage
in
the
community
to
try
to
help
out
there.
You
saw
my
identities
in
there
by
the
way.
If
you
want
to
find
me
in
the
community,
my
name
is
master
singh
24
because
it
seemed
like
a
good
name.
A
So
I
thought
I
was
a
master.
My
number
in
college
is
24,
so
you
know
we
do
a
lot
of
work
out
there.
Now,
on
top
of
that,
what
we
really
do
is
we
looked
at?
How
do
we
actually
take
the
sort
of
open
source
right
and
add
what
I
call
sort
of
Enterprise
qualities
to
it
right
in
terms
of
how
to
run
it?
How
to
operate
it?
A
How
to
have
it
fully
support
it
right,
because
you
know
you
can
take,
you
can
take
open
source,
but
this
would
be
similar
to
you
know
how
many
people
use
you
know
Red
Hat
or
even
a
bun
to
do
distribution
versus
you
know.
I
know
mark
uses,
Red
Hat.
That's
why
I
mentioned
that
for
mark
versus
you
know
trying
to
build
your
own
Linux
distribution
right,
which
is
possible
right.
A
So
we
wanted
to
start
to
look
at
what
are
the
value
ads
that
we
can
build
above
and
beyond
the
actual
software
itself
right
and
leverage
all
the
capabilities
that
that
fabric
has
some
of
them
more
complex,
sometimes
than
people
might
like.
But
what
are
the
best
practices
around
that
and
we
kind
of
call
that
our
IBM
blockchain
platform-
and
that's
mainly
I'll
talk
about
that
in
the
next
few
slides
we're
also
involved
and
that's
more
on
our
product
side,
I
mean
just
for
some
people,
know
I.
Think,
there's
a
confusion
out
there.
A
We
don't
actually
modify
the
fabric
source
code
at
all
like
so
the
IBM
distribution
of
fabric
is
actually
the
fabric
code.
We
don't
even
maintain
an
internal
repository
of
it.
I
don't
allow
that
and
I
actually
took
people
who
do
it
actually
work
for
me.
So
I
can
measure
them
on
that,
because
we
wanted
to
make
sure
that
the
code
that
we
put
out
there
is
compatible
right.
We're
seeing
you've
probably
saw
announcements
from
AWS
you've
seen
you
know,
Oracle,
you
know,
Microsoft
has
it
in
their
cloud
as
well.
A
We
wanted
to
make
sure
that
if
people
do
use
the
technology
right
from
the
raw
fabric
level,
they
can
move
across
that
and
pick
the
plot.
You
know
pick
an
implementation,
a
platform
of
their
choice
right
in
that
case,
at
a
higher
level
and
I'll
talk
about
this.
Towards
the
back
end
of
my
presentation.
We
do
work
on
a
bunch
of
solution
so
primarily
from
a
solutions
that
we
actually
deliver
and
I'll
talk
a
little
bit
about
food
trust.
A
At
the
end,
we
actually
built
two
solutions
that
we,
where
we
are
actually
members
of
the
of
the
actual
consortiums
or
networks
themselves.
I'm
sure
everybody
has
seen
our
marketing
on
this,
so
cuz
you're,
coming
as
no
surprise
food
trust,
a
big
one
with
with
wall
and
again
I'll
talk
about
some
little
lessons
there.
We
also
do
a
network
called
now
call
trade
lands,
that's
a
global
trade
digitization
with
Maersk,
and
then
internally
we
actually
run
a
few
solutions
as
well.
So
we've
had
a
lot
of
experience
actually
on.
A
You
know
what
it
takes
to
actually
build
and
set
up
consortiums
as
well
as
actually
run
and
operate.
You
know,
networks
and
build
solutions
on
top
of
that
and
in
the
third
area,
is
to
try
to
grow
the
ecosystem
right
and
the
ecosystem
includes
people
in
the
community,
but
other
partners,
people
who
might
want
to
build
solutions,
add-ons
and
value,
adds
to
2
R
2
R
over
solutions.
A
So,
in
a
nutshell,
you
know
that's
our
fundamental
strategy,
all
driven
from
the
lowest
level,
though,
from
from
the
from
the
old
from
the
open
source
side
I'm
going
to
talk
mostly
just
from
the
hyper
ledger
side.
Today,
I'll
obviously
talked
mostly
about
fabric,
because
that's
sort
of
the
key
thing
in
our
in
our
platform.
We
do
actually
do
work
with
indie
as
well,
and
we've
been
doing
some
engagements.
A
We
don't
have
any
formal
product
offerings
around
that,
but
we
have
been
doing
some
engage
it's
based
on
the
indie
technology
in
the
identity
space
as
well
looking
to
see
how
we'll
sort
of
evolve
that
into
a
combination
of
potentially
the
blockchain
space,
but
also
just
the
decentralized
identity
space
as
well.
If
you
have
any
questions
on
that
feel
free
to
stop
by
the
booth
later,
we
can
talk
about
that.
I
won't
cover
that
in
these
slides,
so
look
I
mean
I.
A
Think
the
main
the
main
thing
I
just
wanted
to
say
on
this
slide,
there's
a
couple
of
new
things
that
are
happening
right,
I
think
we
do
from
an
IBM
perspective.
You
don't
have
to
use
everything
we
have,
but
we
do
offer
the
full
gamut
and
range
of
services.
As
I
said,
we
offer
the
you
know
the
open-source
side.
A
We
offer
the
product
platform
side
running
in
the
cloud
I'm
going
to
talk
about
our
other
distributions
that
we
have
as
well,
and
we
do
have
consulting
services
and
any
in
solution
platforms
as
well,
and
you
can,
you
know,
pick
to
work
with
us
on
all
that.
We
do
believe
that
that
actually
makes
us.
You
know
very
unique
actually
in
this
space
or
our
focus
on
the
actual.
What
we
call
our
IBM
blockchain
platform
right
has
really
been.
How
do
we
make
it
easy?
A
For
you
know,
one
thing
is
that
there's
people
love
to
look
at
this
sort
of,
like
you
know,
technology
I,
think
right,
I
think
we're
somewhat
past
the
stage.
Maybe
I
mean
it's
funny:
I'm,
a
low-level
Regina
sky,
I,
love
talking
about
low-level
consensus
protocols
and
all
that
fun
stuff.
At
the
end
of
the
day
right,
you
have
to
come
up
with.
What's
your
value
proposition
and
use
gains
that
you're
trying
to
do
what
problem
when
you're?
Actually
trying
to
solve
right,
and
can
you
prove
out
this
blockchain
viable
for
that?
A
Is
it
feasible
for
that
right?
So
you
know,
but
we
wanted
to
try
to
make
happen.
Was
yes
in
the
open
source
can
learn
about
the
technology.
Read
the
code
to
your
heart's
content,
if
you
want
to,
but
now
how
am
I
going
to
make
it
easy
for
you?
If
you
want
to
build
a
trade
finance
solution
or
you
actually
just
want
to
build,
you
know
some
type
whatever
offering
of
maybe
a
food
trust
or
a
supply
chain
offering
right.
How
do
we
make
it
easy
to
develop
on
that?
A
The
other
piece
is
once
you've
got
past
that
point.
One
of
the
one
of
the
hardest
points
is
actually.
How
do
you
get
this
thing
up
and
running
I
mean
you
heard
about
trust.
You
know
earlier
this
morning,
but
and
apparently
you
know,
I
gathered
from
that
conversation
that
only
public
blockchains
are
interesting,
but
we
probably
don't
need
them
and
permission
block
chains.
Aren't
that
interesting.
But
you
know
you:
can
you
can
do
it
other
ways,
but
the
permit
that
the
interesting
thing
on
there
was
the
trust
right.
A
So
how
are
we
going
to
set
these
things
up
right,
I
mean
if
you're.
If
you
know
about
fabric-
and
you
know
about
the
number
of
things
right-
we
have
it's
a
permission
network.
We
have
this
concepts
of
organizations,
you
divide
up.
Here's
you
divide
up
the
trust
model,
the
consensus
model
who's
going
to
provide
ordering.
What
are
the
rules?
You
know
on
channels
and
while
all
the
technical
plumbing
is
there,
how
do
we
make
that,
like
simple,
too
easier
to
do
right
and
what
are
the
common
policies
that
people
want
to
do
right?
A
Who
has
to
approve
that
we
add,
like
a
new
person
to
the
network
right
and
how
do
those
tools
match
with
might
actually
be,
as
Bruce
kind
of
alluded
to
actual
documents
right?
There
are
consortiums
that
have
legal
entities
that
have
legal
documents
that
actually
govern
the
processes
by
which
new
people
will
be
added.
Who
will
write
the
code?
Who
will
be
responsible
for
that
right?
So
how
do
we
have
a
set
of
tools
that
also
can
enforce
and
map
around
those
policies?
A
And
the
last
part
is
really
a
provider
service
around
helping
to
kind
of
run
pieces
of
this
network
for
people
and
again
you
can
I
look
at
this
kind
of
what
part
of
the
value
stack.
Do
you
want
to
focus
on
clearly,
like
I
said
we
put
the
thing
on
an
open
source.
The
documentation
has
gotten
better
about
how
to
run
it.
You
can
choose
to
run
it
your
so
right,
but,
as
you
start
becoming
part
of
a
big
network,
it
becomes
part
of
your
business.
A
Do
you
want
to
focus
more
value
on
like
the
operational
side
of
this
particular
piece
or
on
building
out
the
value
of
the
network,
the
smart
contracts
etc?
So
we
chose
to
focus
on
those
on
those
pieces,
and
you
know
that's
our
basic.
You
know
premise
across
this
was
to
you
know,
reduce
the
risk.
Make
project
successful,
hopefully
make
things
quicker
for
you.
We
do
believe
that
we've,
you
know,
especially
taking
the
fact
of
that
we
use
the
stuff
ourselves.
A
A
What
do
you
do
about
backing
up
the
identities,
making
any
kind
of
enterprise
ready
and,
of
course,
as
I
said,
if
you
choose
that
you
don't
like
us
anymore,
we
want
you
to
be
able
to
move
right,
see
wherever
you
want
to
or,
however
you
may
want
to
do
your
implementation
right
and
that's
that's
the
nature
of
staying
open
and
using
the
open
source
technology,
so
I
just
wanted
to
cover.
You
know
quickly
through
what
we
actually
kind
of
all
for
today
and
I'll.
Preface
this
by
saying
you
know.
A
Originally,
we
started
out
cloud
first,
so
we
started
out.
Basically
you
know
our
distribution
of
the
fabric,
the
IBM
blockchain
platform
was
essentially
a
cloud,
a
cloud
first
solution.
You
can
say
cloud
native
cloud
first
and
it
was
where
IBM.
So
it
was
IBM
cloud
in
the
beginning.
That
made
a
lot
of
sense,
because
you
know
a
lot
of
times.
Networks
were
more,
you
know,
pilot
networks,
or
you
had
a
single
entity
that
was
kind
of
like
the
backbone
of
the
network.
A
You
know
Morris
coming
up
with
the
idea
and
inviting
people,
and
they
were
probably
even
bringing
in
players
that
didn't
want
to
like
they
didn't
even
know.
Sometimes
that
blockchain
was
under
the
covers
and
wanted
to
sort
of
operate
it
and
operating
it
all
in
one
place.
You
know
made
sense
just
from
that
perspective,
especially
from
a
risk.
You
know
minimizing
risk
perspective,
so
we
have.
A
You
know
two
of
our
own
cloud
plans
that
I'll
talk
a
little
bit
about,
but
we've
also
now
you
know
sort
of
moved
for
the
model,
as
people
have
been
advancing
that
you
know,
we
should
really
run
this
stuff
like
it.
The
software
itself
right
fabric
software
runs
anywhere,
so
some
people
will
falsely
tell
you
in
there
claims
their
company.
You
know
that
we
only
run
on
the
IBM
mainframe
and
it
only
runs
on
the
I'm,
a
mainframe.
A
Well,
that's
absolutely
untrue,
but
the
software
will
run
anywhere
if
people
do
run
it
anywhere
and
we
have
multiple
cloud
providers,
but
we
wanted
to
also
be
able
to
expand
and
extend
I
think
our
expertise
into
those
platforms
as
well.
So
we've
started
to
introduce
offerings
that
you
can
now
run
yourself.
You
can
buy.
You
know
our
typical
software
distribution
model
on
top
of
our
kubernetes
distribution.
We've
also
done
some
work
to
move
our
peer
into
AWS
and
we
also
kind
of
have
an
offering
at
the
end
of
the
day.
A
If
you
start
to
feel
yourself
in
trouble
and
need
you
know
some
level
of
support
and
you
choose
to
just
you
know,
download
and
use
our.
We
have
these
images
that
are
basically
just
non-licensed
images.
We
do
offer
support
on
those,
so
we
started
out
kind
of
with
a
lifecycle
of
in
the
what
we
call
it.
Our
starter
plan-
and
you
know
for
I,
don't
know
how
many
people
are
again
how
many
people
are
actually
familiar
with
fabric?
A
Ok,
so
pretty
good,
so
I'll.
You
know,
I
know
I
suppose
more
business
side,
but
at
the
end
of
the
day,
some
of
the
big
things
you
wanted
to
get
started
on
starter
plan.
Right
are
two
things
right.
You
wanted
to
be
able
to
quickly
deploy
your
smart
contracts
and
logic
and
apps
right.
You
want
to
be
able
to
make
that
because
that's
the
value
that
you're
showing
the
line
of
business
and
then
number
two
you
might
not
have
all
your
partners
lined
up
right.
Your
consortium,
you
know
it's,
you
know.
A
I
know
it's
hard
to
even
get
agreements
in
there
and
then
it's
actually
pretty
hard
to
actually
get
people
to.
You
know
sign
up
to
any
group
to
groups
to
agree
to
do
anything
together
right
just
from
timing,
resource
etc,
but
you
might
want
to
actually
prove
out
like
what
would
this
look
like?
If
I
have
you
know,
seven
banks
in
the
consortium
right?
How
does
this
all
work?
So
one
of
the
things
that
we
actually
did
in
there
was
basically
to
allow
you
to
quickly
spin
up
like
a
multi
organization
simulation
right.
A
So
you
can
just
say:
here's
my
ten
banks,
here's
what
it
would
look
like
and
kind
of
show
that
value.
It
also
allows
you
to
go,
bring
in
and
show
your.
You
know
if
you're
running
a
pilot
to
actually
go
show
one
of
these
partners
right
that
hey
here's,
what
would
actually
look
like
you
would
actually
have
this
node,
etc.
A
So
that
was
really
what
we
what
we
went
through
on
this
sort
of
starter
plan
and
we've
actually
had
very
you-
know
very
good
adoption
on
that,
both
in
the
sort
of
POC
development
and
even
pilot
phase,
the
next.
The
next
phase
was:
what
do
we
have
from
a
sort
of
production
perspective
right
and
that
you
can
eyes
can
read
this
as
much
as
you
know,
I
can.
But
this
was
essentially
how
do
we
have
production
cap?
You
know
how
do
we
provide
all
the
resilient?
A
Take
all
the
operational
responsibilities
out
of
your
hands
to
provide
the
resilience
that
you
need
right
if
you're
moving
this
thing
to
go
run,
you
know
sudo,
somewhat
mission,
critical
application
right.
How
do
you
make
sure
that
you
know
the
data
is
still
there
that
your
nodes
are
up
and
running
that
data
is
backed
up
and
even
still
in
the
early
days
here,
even
things
like?
How
do
you
restore
data?
You
know
if
you
chose
to
have
like
a
checkpoint,
because
somebody,
you
know,
messed
up
something
in
the
application
right.
You
know.
A
How
do
we
do
that,
so
those
are
really
the
things
that
we
kind
of
put
together
in
the
enterprise
plan,
and
we
essentially
had
those
these
two
plans.
You
know
through
the
middle
of
this
year,
but
as
we
started
to
evolve,
I'm
gonna
skip
that
one
as
we
started
to
you
know,
evolve
later,
one
of
the
one
of
the
things
that
that
came
up
or
hey
I've
started
my
consortium,
my
consortium
right
and
food
trust
might
be
a
good
example
of
this.
But
hey.
A
We
want
to
bring
on
XYZ
company,
but
XYZ
company
has
a
contract
with
Azure
or
AWS
right
or
or
Oracle
right.
How
do
we
allow
those
guys
to
kind
of
connect
up
their
peer
join
the
network,
but
still
how
could
you
actually
support
that?
Or
how
can
you
you
know,
make
part
of
that
in
there?
So
in
our
kind
of
our
first
phase,
just
by
really
we
just
kind
of
picked
it
because
it
was
probably
the
most
demanded
at
the
time
we
actually
started
to
implement
the
ability
to
deploy
the
first
stage
of
our
stuff.
A
We
just
deploy
the
peer
out
there
today,
but
kind
of
fully
managed
peer
supported
by
IBM
running
on
AWS.
Moving
forward
we're
going
to
move,
you
know
we'll
move
the
entire
stack
to
AWS
I
know:
AWS
has
their
own
AWS
management
service.
Ours
is
better,
but
no,
but
in
it
is,
but
but
we
do
realize
that
AWS
has
a
lot
of
other
compelling
services,
etc
right
that
people
may
use
they
have
that
contract.
A
So
how
do
we
provide
that
same
level
of
support
all
of
our
tooling
etc
right
to
actually
support
nodes
running
in
multiple
cloud?
And
how
do
you
get?
You
know
an
aggregate
view
right
that
you
could
actually
be
a
node
running
in
AWS
and
the
ordering
service
is
running
somewhere
else,
and
you
want
to
have
the
management
tool
to
govern
and
create
channels
and
participate
in
that
in
our
in
our
environments.
Today
we
require
you
to
be
all
in
IBM
cloud
in
our
environments
coming
up
through
one
Q
and
two
Q.
A
This
is
the
first
foray
into
that
will
will
allow
kind
of
multi.
Basically,
nodes
can
be
anywhere
they
can
be
run
by.
They
can
either
be
ours
license
from
us
and
managed
by
us.
They
could
actually
be
other
people's
nodes
and
our
tooling
will
seamlessly
make
that
network
all
work
together.
The
AWS
peer
is
kind
of
our
first
unifor
way
into
that
on
the
on
the
side
of
distributing
it
to
you
run
it
wherever
you
want.
You
know
within
your
own
within
your
own
data
center
or
you
just
you
know,
want
to
run
something.
A
You
know,
however,
you
want
to
deploy
it
in
whatever
environment.
Maybe
we
also
wanted
to
have
a
distribution
there.
We
kind
of
believe
that,
at
least
for
people
who
want
to
try
to
set
up
multiple
components,
Doosan
boxes,
etc
and
I
think
this
is
true
on
the
open
source
side
and
I
mean,
if
you
ask
mark
from
you,
know,
Red
Hat
and
other
companies.
You
know
kubernetes
seems
to
be
the
the
best
orchestra
at
you
know.
A
The
predominant
I
would
say
it's
the
best,
but
I'll
just
say
to
be
fair:
the
predominant
orchestration
platform
out
there,
even
docker
right
is
actually
supporting
kubernetes.
So
we
have
a
IBM
distribution
of
kubernetes
called
IBM
cloud
private.
So
we
support
a
distribution
that
combines
basically
our
supported
kubernetes
with
running
all
of
fabric.
A
You
know
as
kubernetes
service
and
eks
amazon's,
kubernetes
service
and
GCE
as
well
right,
so
when
basically
move
our
model
to
have
native
support
for
basically
any
kubernetes
you'll
be
able
to
have
all
of
our
tooling
you've
seen
that
it
has
the
ability
to
employ
peers,
configure
orderers
configure
cas.
So
all
that
will
actually
work
will
just
seamlessly
deploy
our
tool
into
the
kubernetes
cluster
and
be
able
to
work
in
any
kubernetes,
so
that's
kind
of
where
we
started
to
look
at
this
whole.
A
This
whole
model
and,
as
I
said,
if
somebody
chose
to
run,
you
know
their
own,
their
own
version
of
it
or
the
peers.
They'd
still
be
able
to
take,
for
example,
our
management
tools,
etc
and
and
register
either
other
people's
peers
or
even
their.
You
know,
open
source
or
supported
by
through
some
other
vendor
within
their.
So
that
gives
us
this
vision.
Right
of
you
know,
blockchain
at
its
best
right.
It's
not
one
vendor
controlling
the
software.
It's
not
one
vendor
controlling
the
cloud.
It's
not
one
organization
called
controlling
everything.
A
And
all
of
this
stuff
was
really
I
mean
this
all
makes
sense
right
and
we
could
have
started
out
and
I
think
it's
now
that
we're
seeing
that
the
markets
actually
ready
for
that
right,
we're
seeing
more
and
more
of
those
times
where
the
main
set
of
people
have
gotten
started.
The
foundation
is
in
and
now
they're
looking
to
add
the
new.
You
know
the
latest
and
greatest
member
or
whoever
maybe
to
their
network,
and
they
bring
these
other
requirements
to
you.
So
that's
really,
you
know
our
core
our
core
direction.
A
So
again,
just
a
if
you
didn't
listen
to
anything
I
said
to
sum
it
all
up.
You
know
we
also
have
other
there's
other
requirements
that
are
out
there
right
and
and
not
every
and
believe
it
or
not.
Not.
You
know,
even
across
all
the
clouds
in
the
world
right,
AWS
IBM,
as
your
Google,
there
are
still
places
in
the
world
where
there's
not
a
cloud.
A
I
know
it's
very
hard
to
believe,
but
there
actually
are
places
that
on
a
cloud
and
you
do
need
to
have
the
ability
to
deploy
you
know
into
those
into
those
infrastructures,
and
you
might
want
to
support
a
thing.
So
we
want
to
make
sure
that
we,
you
know,
have
to
give
you
that
fully
supported
model.
We
want
to
make
sure
that
you
don't
have
to
just
pick
us.
A
You
can
still
get
support
from
us
if
you
want
to
work
with
us,
but
we
want
to
make
sure
that
we
can
have
all
those
networks
in
there
and
that's
really
going
to
start
to
help
expand
Network
right,
because
you
don't
want
to
have
a
thing
to
say:
hey
our
network
requires
you
to
run
on
IBM
I.
Don't
like
IBM
right.
We
wanted
to
do
the
network
that
people
shouldn't
know
right.
They
should
just
know
that
I've
got
my
peers.
A
Here's
my
consortium,
there's
a
network
of
things,
do
I
want
to
join
this
network
because
I
need
to
be
part
of
the
supply
chain
or
am
I
going
to
join
this
network,
because
Walmart
has
mandated
that
I
must,
you
know,
provide
freshness
for
greens
in
food
trust
right.
So
how
do
you
break
down
those
barriers
by
allowing
deployment
anywhere
low
passing
starts
out,
but
this
was
just
trying
to
show
that
I
think
we're
you
know
doing
doing
pretty
well
across
that
we
there
will
be
a
presentation
later.
We
don't.
A
You
know
IBM
as
of
right
now,
but
we
you
know
we
can
consider
where
we
release
this
based
on
all
the
work
we've
done
with
writing
smart
contracts
and
making
this
easier
for
people
to
get
started.
We
have
just
recently
released
a
new
development
tool
as
a
plugin
to
vs
code
again
to
really
aid
and
quickly.
You
know
speed
up
development.
A
You
know
as
well
right
and
we
see
those
as
kind
of
the
focus
areas
that
that
we
can
work
on
moving
forward,
all
right,
so
that's
kind
of
where
the
platform's
evolve
to
what
we
have
today
and
where
we're
going
and
a
lot
of
that,
as
I
said,
you
know,
comes
from
our
experience.
Working
with
existing
customers.
I
wanted
to
just
run
quickly.
You
know
I
do
want
to
leave
a
little
bit
of
time
for
questions,
for
a
couple
of
you
know,
use
cases
and
bring
out
some
key
points.
A
I
don't
expect
everybody
to
read
all
these
slides,
but
you've
probably
seen
announcements
out
there.
So
an
early
member
of
a
hyper
ledger
was
CLS
CLS.
Basically,
you
can
read
all
about
CLS,
but
at
the
end
of
the
day
they
do
massive
compression
for
multilateral
netting
for
cross-border
currency
settlement
right
one
of
the
things
that
they
do
it
for
like
60
currencies.
Some
number
like
that,
one
of
the
things
that
they
wanted
to
start
look
you
know
looking
at
was:
how
do
we
can
we
more
see
it?
A
Can
we
more
easily
add
other
currencies
and
at
least
provide
some
of
our
netting
services?
We
already
have
a
consortium
of
banks.
How
can
we
add
those
and
just
add
those
into
the
system?
And
since
we
need
to
do
some
new
development
anyway
distributed
Ledger's,
look
interesting
for
this
technology.
So
why
not?
Let's
pick
you
know
blockchain
for
this
right
and
they
were
very
interested
in
they
had
looked
at.
A
So
it's
you
know
the
business
relationships
are
in
place.
The
use
case
was
kind
of
in
place
right,
so
I
made
it
a
little
bit
you
didn't.
We
didn't
have
like
that.
We
were
over
that
hurdle.
You
know
once
we
started
the
other.
The
other
thing
that
was
in
here
was
you
know
initially
the
promise
of
the
use
case
right.
It
was
a
combination
right
because
I
think
people
were
incentives.
A
You
have
the
novelty
of
blockchain
and
to
be
relevant
by
saying
they
were
doing
something
with
blockchain,
so
that
was
good,
but
but
the
use
case
itself
is
actually
what
kind
of
drove
the
project
forward.
More
so
than
just
saying,
hey,
you
know,
blockchain
is
the
answer.
What's
the
problem
right,
they
really
pitched
more.
What
problem
where
we
actually
are?
We
actually
suits
solving.
A
The
other
thing
you
know
was
no
I,
believe
everybody
can
look
and
do
things
on
their
own
I.
Think
that
I
think
that's
you
know
great
CLS
took
again
took
an
interesting
approach.
Was
we're
gonna
join
hyper
ledger?
We
know
that
we're
going
to
actually
be
allowed,
basically
to
contribute
code,
and
we
don't
want
to
be
the
people
like
building
and
maintaining
and
learning
the
core
plumbing,
because
we
want
to
build
a
netting
right,
but
they
chose
to
you
want
to
have
influence.
A
A
A
Generally
speaking,
if
the
roadmap
has
the
features
on
the
alignment
that
you
want
it
to
be,
you
can
align
your
development
and
delivery
to
that
and,
in
the
end,
those
two
actually
staying
the
course
the
bottom
line
is.
Technology
will
usually
end
up
delivering
before
the
business
is
actually
able
to
put
the
thing
up
for
non-technical
reasons
and
I
think
a
lot
of
times.
People
make
the
mistake,
thinking
that
they
have
to
solve
every
problem
up
front,
that
they
won't
start
a
project
if
a
features
missing-
or
we
want
to.
A
A
We
needed
some
privacy
or
separation,
but
they
still
inherently
proved
out
that
the
stuff
could
work
right
and
that
was
a
big
part
of
it,
and
then
they
knew
as
these
features
came
along
and
they
were
influencing
him
because
they
were
able
to
talk
to
the
community
that
we
were
able
to
keep
the
technology
moving
along
as
they
needed
to
meet
their
implementation,
and
that's
a
really
important
point,
because
look
we've
actually
made
the
mistake
ourselves
right.
We
have
a
customer,
you
know
or
like
any
company,
it's
out
there
and
they
say
hey.
A
We
need
XYZ
and
we
do
XYZ,
even
though
XYZ
was
planned
in
order
like
three
months
out,
we
double
down
and
try
to
get
it
done.
You
know
tomorrow,
right
and
that
usually
just
throws
like
the
whole
balance
off,
so
I
think
you
know
being
able
to
align
that
and
it
for
the
technical
folks
out
there
in
line
of
business
right.
This
is
the
what
you
have
to
figure
out
how
to
meet
innovation
and
execution
right,
because
in
pure
innovation
you
don't
have
to
deliver
anything
right.
A
You
deliver
a
demo,
you
don't
have
production
right
in
production
and
for
production
level
systems
right
you're
supposed
to
deliver,
and
you
have
this
credit.
Well,
you
can't
base
your
decision
on
what
you're
gonna
deliver
in
twelve
months
or,
however
long
your
development
cycle
is,
unlike
the
tech
today,
you
should
be
constantly
evolving.
That
doesn't
mean
that
you
couldn't
run
on
fabric
1:1,
for
example,
but
why
would
I
before
releasing
fabric
one
of
the
Google
versions,
every
quarter
and
I'm
not
delivering
for
six
quarters?
A
Why
would
you
six
quarters
later
still
be
running
on
fabric
one
more
as
an
example
right,
so
you
have
to
have
this
adjustment
of
like
technical
innovation
aligned
with
the
business
right
so
that
you
can
stay
aligned
and
keep
going
and
stay
ahead.
So
those
are
some
of
the
key
lessons
that
we
actually
learned
on
that
project.
A
We
did
another
project
with
secure
key,
which
was
basically
you
know
an
attribute
sharing
trusted
identity,
attribute,
sharing
type
type
network
with
a
bunch
of
banks
in
seven
banks
in
Canada,
and
the
basic
use
case
compelling
use
case
for
them
right
was
hey
if
I've
already
been
KY
KY
seed
I'll
make
up
a
word,
but
KY
seed
by
TD,
Bank
and
I
want
to
go
open
up.
You
know,
Rogers,
you
know
telephone
account
or
vice
versa.
A
Do
we
really
have
to
go
through
all
that
information,
or
can
I
leverage
my
TD
information
that
they're
willing
to
share
that
this
guy
really
lives
at
this
address?
He
has
this
account
with
Rogers
right
and
vice
versa,
right
or
if
I
have
a
TD
account
and
I
want
to
open
up
an
orange
account
or
an
RBC
account.
A
You
know
for
some
other
type
of
month
financial
instrument,
why
you
know
go
through
the
process
again,
so
this
whole
idea
was
about
how
do
I
share
that
information
right,
set
up
the
rules
for
sharing
that
information
and
and
then
also
make
sure
that
you
didn't
actually
want
to
know
where
that
information
from
right.
So
the
person
you
requested,
the
information
and
shared
it
RBC
shouldn't
know
that
it
came
from
TD,
but
they
just
know
that
it
came
from
somebody
who
was
allowed
to
provide
at
the
station
for
the
address.
A
So
again
the
compelling
use
case
now,
so
they
had
a
use
case.
The
technical
pieces
were
tough
right,
but
fundamentally
biggest
problem
here,
big
prop
bigger
problem.
Here
then
the
technical
delivery-
and
this
goes
back
to
my
last
point-
was
it
takes
a
long
time
to
get
all
these
folks
to
agree
on
all
the
rules?
What
they're
going
to
do
it
goes
to
their
systems,
people
their
security
people.
Somebody
else
comes
up
with
some
new
requirement.
These
these
things
take
a
long
time
right.
The
other
nice
thing
on
the
security
Pro
is
security.
A
Was
that
hey
there
were
features
that
they
wanted
in
fabric
right.
So
if
you
want,
if
you
are
a
tech
shop-
and
you
want
stuff
in
there-
the
fact
that
the
codes
open-source,
we
said-
hey
we'll
help
design
it,
we
actually
actually
put
some
people
on
it,
but
hey
you
guys
can
contribute
those
features
right,
don't
fork
the
code,
don't
create
your
custom
stuff
just
contribute
the
features
to
open-source
right
which
again
benefits
the
community
and
yourselves.
A
The
these
were
the
other
thing
here
was
once
they
got
past
all
the
legal
work
and
all
the
stuff
that
was
in
there.
It
really
was
hey
I,
don't
for
those
of
you
know,
fabric,
it's
like
hey
I,
don't
want
like
security
or
IBM
to
be
the
one
who
sets
up
all
the
channels
right.
I,
don't
want
to
be
that
you
know
we
don't
trust
those
guys
to
do
it.
We
want
to
like,
have
hands
on
keyboard
and
go.
Do
it
right,
so
they
actually
went
and
use
all
they
had
like
a
formal
process.
A
They
said
when
the
CA
was
generated.
They
actually
go
in,
create
their
org.
They
literally
go
and
create
their
channels.
I
want
to
have
my
channel
with
these
three
banks.
Those
banks
create
it,
they
do
the
whole
process
themselves
right.
They
actually
work
through
everything.
It's
all
you
know
under
their
control,
so
actually
having
that
sort
of
governance,
that
kind
of
met.
A
What
was
written
down
and
seeing
it
actually
made
people
feel
a
lot
bit
mean
it
worked
and
it
made
people
feel
you
know
that
they
were
actually
part
of
the
whole
process,
and
then
this
was
an
example
where
I
think
we
made
some
mistakes
in
trying
to
accelerate
various
things
right.
In
the
end,
we
found
that
if
we
had
more
stayed
on
the
course
of
where
the
roadmap
was
going,
where
the
IBM
blockchain
platform
was
going
and
the
alignment
of
the
the
issues
up
top,
you
know
things
would
have
rolled.
A
So
again,
a
bunch
of
banks
got
together.
You
have
a
compelling
need
to
sort
of
you
know,
share.
You
know,
credit
and
information.
You
know
across
the
board.
This
was
an
interesting
one
as
well.
It
again
had
the
same
kind
of
issues,
because
a
new
entity
needed
to
be
created.
You
know
from
these
banks
had
to
have
a
governance
policy.
There's
actually
people
in
here
who
actually
worked
on
the
governance
stuff,
like
one
poor
guy
who's
sitting
back.
A
There
literally
would
just
be
in
meetings
with
these
guys
about
who's,
going
to
be
the
person
who
issues
the
certificates
and
agreeing
on
the
fact
that
you
know
in
the
end
they
agreed
I'm
like
cert
Europe
was
gonna,
be
the
people
who
I
think
was
who
issue
all
the
certificates
for
all
the
all
the
client
applications
that
were
running
right,
but
just
getting
all
that
stuff
lined
up
right.
You
know
takes
a
lot
of
time,
so
that
was
you
know
very
difficult
right
and
then
making
sure
that
the
tech
met
with
it.
A
A
Right
because,
fundamentally,
right
you're,
not
it's
not
the
same
thing
as
like
no
offense
to
Bitcoin,
but
it's
not
the
same
thing
as
like
Bitcoin,
which
is
just
like
I,
send
a
transaction
right,
though
it's
pretty
easy
to
build
a
new
client
they're
right
here,
you're
now
building
a
whole
thing
on
porting
workflow,
so
you
have
an
entire
application.
Suite
that's
built
on
top
of
this
right.
A
That,
actually,
is
you
know,
75%
of
the
time
cost
money
and
effort
right
is
actually
on
that
compared
to
the
actual
parts
of
writing
the
smart
contracts
on
the
blockchain,
so
we
started
to
also
kind
of
you
know
notice
that
as
well
right
that
people,
don't
you
know,
fully
plan
ahead.
You
know
thinking
that
hey
it's
the
blockchain
piece,
I
just
need
blotching
expertise.
No,
actually
you
need
some
of
that
too,
the
smart
contracts,
but
then
you
need
your
same
standard,
typical
people
that
know
how
to
build,
distributed,
applications
and
all
the
integration
points.
A
So
that
was
that
was
the
keys
there
and
then
you
know
food
trust.
You
know
the
last
one.
This
was
like,
you
know
very
I,
think
interesting.
You
know
learning
from
from
our
perspective,
especially
on
being
a
more
on
the
product
side,
because
we
were
actually
you
can
call
it.
You
know
I
hate,
to
use
these
terms,
but
I'm
just
going
to
well
you're,
partially
one
of
the
founders
of
the
network
of
food
trust
with
Walmart
right.
So
it
was
so
in
our
model.
We
you
know
with
Walmart
is
actually
that
we
are.
A
You
know,
however,
the
network
itself
makes
money
right,
that's
how
we
IBM
make
money.
So
we
are,
we
do
run
some
of
the
nodes,
etc.
But
it's
not
a
you
know:
supplier
or
provider
model
right,
we're
actually
part
of
the
overall
network
itself
right.
So
we
of
course
have
to
get
the
you
know,
subject
matter
expertise
on
food
trust
coming
from
Walmart.
You
know
building
out
the
actual
you
network
itself,
but
now
we
actually
have
to
go
and
work
with
partners
etc
to
bring
in
to
actually
get
people.
A
The
only
way
that
we
kind
of
make
money
is
when
the
network
gets
more
people
and
more
people
start
paying
for
the
data,
etc,
etc.
Right
which
becomes
it
gives
us
an
interesting
perspective
on
how
the
sort
of
business
model
you
know
technical
part,
wasn't
the
hard
part
per
se
right.
It's
really
the
business
model
that
becomes,
you
know,
really
tricky
there,
and
then,
of
course,
you
know
it's
a
great
example
of
having
you
know
both
a
strong
player
right,
Walmart's,
a
very
there's.
A
lot
Walmart
has
a
lot
of
suppliers
right,
so
they
can.
A
You
know
on
their
mandate,
so
that's
one
of
those
ones
where,
if
you're
kind
of
the
you
know
a
compelling
event
coupled
with
you
know,
somebody
who
is
I,
don't
know
what
the
term
is,
but
I'll
just
use
like
power
player.
You
know
you
know
big
anchor
in
this
stuff,
who
has
that
influential
influential
partner
can
actually
again
push
forward
on
the
solution,
assuming
that
blockchain
wasn't
right
technology,
which
it
kind
of
what
is
here
right,
because
you
want
to
make
sure
that
nobody
can.
A
You
want
have
multiple
copies
of
the
records
out
there
right
so
that
nobody
can
come
back
and
actually
tamper
with
one
right.
They
actually
have
to
go
out
and
modify
multiple
of
them
right,
because
the
big
problem
in
the
chain
today
is
right.
It's
document
oriented
in
people
forge
documents
right
and
they
do
it
after
the
fact
not
just
ahead
of
that
fact.
So
that
was
the
I
think
that
was
the
that's
been.
A
A
So,
like
the
question
was
about
being
allowed,
if
you
were
running
a
peer
somewhere
else
in
another
cloud
and
connecting
it
to
you,
you
know
some
networks
or
other
things
that
might
be
running
from
with
an
IBM
perspective,
so
yeah
I'm
from
a
technical
perspective,
it's
all
possible
today,
I
think
from
from
the
offering
perspective
of
actually
having
like
more
integrated
into
the
tooling
and
consoles.
That's
what
we
have
coming
forward
over
the
over
over
the
next
quarter,
but
yeah,
and
so
we
planned
actually
I
think
there's
a
session
later
from
the
Oracle.
A
We
actually
did
some
Interop
work
to
show
what
would
actually
happen
if
you
deploy
so
I've
connected
up
here,
for
example,
in
the
IBM
blockchain
platform,
is
actually
joined.
I
mean
I
had
to
use
the
fabric
API,
but
I
was
able
to
get
enough
information
from
the
Oracle
cloud
to
actually
have
it
join.
For
example,
in
ordering
service
running
you
know,
in
the
Oracle
network,
I
was
able
to
take
like
the
AWS
P
or
a
product
that
we
have
and
have
it
registered
with
the
CA
there.
So
we've
been
working
on.
C
So
I
was
gonna.
Ask
a
similar
question.
I
asked
in
the
last
session,
which
was
around
you
know
like
in
food
trust.
You've
got
so
many
competitors.
How?
How
did
you
architect
the
solution
to
kind
of
off
escape
the
the
volume
that
maybe
vendor
a
is
having
with
with
Walmart
and
without
having
you
know
a
proliferation,
maybe
of
a
lot
of
channels?
Yeah.
A
A
Can
you
know,
obtain
it
if
they
need
to,
and
you
have
that
full
history,
so
we
started
out
currently
more
with
like
channels
all
the
data
itself
is
encrypted,
and
then
we
provide
access
control
on
that.
But
that,
like
you,
said,
that's
not
going
to
fully
scale
as
you
try
to
build
the
whole
massive
chain,
but
for
the
tier
1
and
tier
2
suppliers
that
part
works
well
and
it
makes
sense
to
probably
keep
those
out
as
channels
anyway,
because
those
are
the
kind
of
supply
supply
lines
that
would
go.
A
D
A
So
the
question:
oh
yeah,
you
guys
you're
in
the
mic,
so
yeah
I
mean
I.
Think
there's
I
mean
there's
a
number
of
things
that
you
know
affect.
You
know
performance
right.
So
there's
you
know
as
especially
as
you
start
to
spread
things
out
right.
You
know
you're.
Obviously,
that's
where
latency
can
come.
You
know
more
into
the
mix
right,
but
on
the
overall
you
know,
I
think
throughput
side
of
things
right.
You
know,
we've
been
working,
I,
think
purely
on
a
sort
of
fabric
side
right
to
be
able
to.
A
How
do
we
mor
parallelize
operations?
How
do
we
make
them?
You
know
how
do
we
make
some
things
that
were
traditionally
sequential?
Where
how
can
we
make
certain
things
you
know
run
in
parallel
and
be
non
sequential,
I
think
the
other
case
is
right.
Now
we
have
a
kind
of
a
generic
processing
model,
so
you
know
you
can
kind
of
the
chain
code
model
that
we
have
today
allows
you
to
run
kind
of
generic
smart
contracts.
A
You
can
kind
of
do
anything
you
want
so,
for
example,
work
that
we're
doing
specifically
on,
like
fungible
tokens,
we're
actually
building
a
purpose-built
transaction
model
for
flexible
tokens
within
fabric
and
integrate
those
within
trained
code.
So
it's
kind
of
both
approaches
right
to
look
at.
If
there's
like
specific
use
cases
that
we
know
and
common
things,
we
can
actually
build
more
optimistic.
We
can
build
a
more
optimistic
code
path
anyway,
because
we
have
custom
transaction
processing
that
we
can
do
on
the
other
side.
Mostly
most
of
this
stuff
is
illuminating
contention.
A
You
know
within
the
system,
it's
you
know
itself
right
and
figuring
out
how
to
parallel
lies.
You
know
the
throughput
in
operations
and
then
we,
you
know
we
have
a
lot
of
those.
So
we
have
that
we
do
a
lot
of
that
work.
On
the
you
know,
open
source
side,
there's
some
work
on
the
copper
and
other
projects
out
there
and
we've
been
doing
a
lot
of
that
work.
You
know
internally,
as
well
from
from
an
IBM
perspective,
looking
to
push
that
back
in
I.
A
E
A
A
So
these
three,
so
we
trade
used
it
security,
didn't
so
only
only
we
trade
actually
used
composer
I
mean
at
the
time
composer
when
we
started
CLS
composer
didn't
exist
and
security
guys
are
really
smart,
so
they
just
write
code
themselves
and
and
actually
I'll
most
of
the
code
was
a
go
and
Java
was
written
for
for
food
trust,
yeah.
So
I
didn't
mention
composer.
That's
why
I
kind
of
snuck
in
the
vs
code
thing.
So
what
are
we're?
A
A
Those
as
more
on
a
higher
level
API
on
the
client
SDKs
will
still
have
the
load
for
you
guys.
Don't
worry
for
the
people
who
love
to
write
low-level
plumbing
that'll
still
be
there
for
you,
you
can
you
can
figure
out
what
peers
you
want
to
send
things
to.
We
won't
take
that
away,
but
well
right
at
that
higher
level
you
know
into
there
and
then
a
chain
code
side,
it's
basically
more
of
an
inversion
right.
A
Where
now
we
write
those
become
libraries
that
you
can
use
within
core
chain
code,
so
you
can
choose
to
mix
and
match
right.
So
if
you
choose
to
write
a
higher-level
library
and
you
need
to
drill
down
to
a
low-level
function,
you
can
so
that's
the
the
model
going
forward,
we're
doing
node
first,
only
because
most
of
this
stuff,
the
composer
model
was
more
node
and
node
I
think
we
have
pretty
good.
You
know
use
cases
in
one
for,
as
we
refine
that
we'll
move
that
model
into
the
into
the
other
languages
as
well
and.
A
E
A
So
it's
really
it's
really.
Where
are
where
the
stuff
distributed
by
IBM
would
be
so
yeah
I
mean.
Are
we
we
have
this
stuff
running
on
like
so
there's
an
IBM,
you
can
get
support
and
licensed
product
from
IBM
for
like
AWS
today,
IBM
cloud
private
runs
on
AWS
and
Azure.
Today,
we'll
do
aks
and
AWS
will
be
aks
and
eks
will
be
doing
in
probably
2q
Google
I.
Don't
know
when
will
when
will
do,
but
we
could
I
think
you
know
it
just
comes
down
to
whoever
asked
for
that's
for
one
but
I
think.
A
But
the
real
thing
comes
down
to
the
key
point
in
there
is
the
fact
that
we
go
basically
kubernetes
native
and
a
second
we
go
kubernetes
native,
all
I
really
have
to
do
is
deploy
all
week.
Sorry,
all
we
really
have
to
do
is
the
our
our
tool,
basically
our
kind
of
deployment
tool
into
kubernetes,
and
then
we
will
have
to
have
a
permission
to
basically
you
know
in
whatever
namespace
it's
in
to
run
the
coop
api's
because
well
basically
have
operators
under
the
covers
and
then
it'll
be
able
to
provision
into
anything.
A
So
essentially,
once
we
actually
get
that
done,
then
it
just
becomes
basically
a
testing
case
right
I
mean
because
literally
we'll
we'll
just
deploy
the
same
thing
into
native
coop
and
use
the
coop
ap
is
so
the
only
much
small
difference
will
probably
be
two
things
they'll
be.
If
there's
almost
everybody
gives
us
access
to
coop
control,
some
might
make
us
do
it
as
an
opera.
A
You
know,
might
make
us
use
pure
deploying
stuff
as
an
operator
versus
helm,
charts
if
they're
allowed
and
then
the
other
will
be
mapping
what
type
of
persistent
volumes
we
actually
purchase.
Some
volume
claims
and
storage
and
want
to
use
and
making
sure
that
support
it.
But
from
from
that
perspective,
I
anticipate
that
once
we
get
the
work
done
in
one
queue
we'll
be
able
to
roll
on
a
niku
platform.
That's.
A
A
A
A
I
mean
I
yeah,
I,
think
I
mean
yeah.
Me
said
we're
at
one
point,
but
I
think
that
more
is
gonna.
Move
to
yeah
I
mean
it
doesn't
mean
you
don't
want
to
run.
If
you
can
run
everything
in
a
single
cloud
in
the
same
place,
you
might
as
well
just
run
a
centralized
database
and
it
probably
faster
and
more
efficient
by
the
way
as
a
side
saying
if
you
can
use
a
centralized
beta
database
and
solve
your
problem,
I
highly
recommend
it
but
yeah.
A
Just
because
I'm
not
you
know,
build
block
chains,
but
you
know
that's
the
solution
in
there,
but
yeah
I,
think
I
mean
more
and
more
that's
what
we're
seeing
right.
People,
people
have
their
own
preferences
and
choices
and
also
for
the
integrity
of
the
overall
network.
Right,
you
know,
being
able
to
I
think
as
apps
become
more
more
mission-critical
right.
People
are
going
to
have
more
concerns
that
if
it
requires
like
five
peers
to
you
know,
come
to
consensus
on
something
right.
A
You
don't
and
you
need
to
have
or
even
was
like
three
out
of
five
right.
How
do
I
make
sure
that
those
are
available
so
I
mean
people
should
probably
I
mean?
If
you
really
wanted
to
be
paranoid
about
it,
you
should
actually
go
and
figure
out
who
the
ISP
providers
are,
where
all
the
peering
is
blah
blah
blah
right
and
actually
start.
Looking
at
the
overall
like
low-level
networking
stuff
to
pick,
you
know
which
data
centers
and
locations
you
should
actually
be
in,
but
yeah,
that's
that's
the
direction
that
we
see.
Thanks.