►
From YouTube: After Production: Prepare the Software Factory for Blockchain Main... Marco Alarcon & Jaime Pradenas
Description
After Production: Prepare the Software Factory for Blockchain Maintenance - Marco Alarcon & Jaime Pradenas, Bolsa de Santiago
Your Fabric project went live and is now on Production. After the celebration, it comes the not so exciting long term maintenance related to any IT system. At the Santiago Stock Exchange, we underwent a process of knowledge transfer from the R&D team toward the Software Factory to keep the Exchange Blockchain network running and evolving, with unforseen challenges that arise from the use of Blockchain technology. At the same time, the R&D team took the effort to plan the evolution of the Fabric network, with an aim to create a Blockchain Consortium and an explandable platform. This session expose how to plan your blockchain architecture for the future and what are the core activities required to keep your Blockchain network running in the long term, when at the same time it evolves with the incorporations of new business cases, requirements and participants. This sessions ends with the current plan of the Santiago Stock Exchange regarding the next Blockchain platform that is under development.
----
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 #HyperledgerGlobalForum2020 #Blockchain
A
All
right,
thank
you,
so
much
guys
for
being
here
today
with
us.
Today
we
are
going
to
share
with
you
all
our
experience
with
a
solution
on
production.
For
two
years.
My
name
is
Jaime
Briana's
I
am
senior
abroad
chain
protocol
manager
in
some
depth.
Change
with
me
is
Mark
caller
con
he's
the
head
of
blockchain
architecture,
he's
gonna.
Do
that
the
hard
part
of
this
presentation
so
I
like
to
start
briefly
with
a
small
introduction
about
our
country,
Chile,
and
also
this
Antioch
change.
A
beautiful
picture
comes
from
the
north
of
our
country.
A
A
We
were
the
first
country
in
South,
America
being
part
of
the
OECD,
and
we
also
have
the
highest
GDP
in
South
America.
Well,
Chile
has
been
recognized
as
a
very
nice
place
to
to
best
that's
why
we
have
those
records
they're
there
for
economic
of
freedom,
business,
environment
and
foreign
regime.
Also
other
facts
about
that
are,
for
example,
we
produce
a
lot
of
copper
initially
and
also
have
a
lot
of
little
reserves
about
our
products
for
exportation
or
wines
are
really
really
good.
A
Alright,
our
company
Santa
Exchange
is
a
127
years
old
company.
It's
a
list
at
the
stock
exchange.
So
if
you
want
to
get
a
stock,
you
can
go
to
the
market
and
get
one.
It's
really
easy.
We
trade
every
day,
3.5
billions
dollars,
mostly
in
fixed
income,
also
in
2018,
were
chosen
like
the
most
notable
stop
the
change
in
South
America.
A
B
Thank
thank
you
Jaime.
Well,
as
Jaime
said,
we
are
a
technological
company,
mainly
it's
very
strange,
because
normally
one
thinks
about
the
stock
exchange
as
a
more
a
traditional
way,
a
centralized
company.
We
are
also
a
centralized
company,
but
we
first
are
a
software
Factory
and
that's
why
we
are
talking
here
first.
What
I'm
going
to
do
is
a
summary
of
the
steps
that
we
underwent
to
get
our
blockchain
solution.
That
is
call
it
security
lending
repository
how
we
got
that
in
production.
It's
a
very
short
tail.
B
We
started
from
an
idea
created
a
minimal
Viable
Product,
and
then
we
release
it
that
not
easily
in
production.
Well,
this
is
after
what
happened
about
that.
When
everyone
is
happy,
we
open
the
champagne
bottle,
we
go
on
party
etc,
but
the
system
has
to
keep
going,
and
that
is
the
hardest
part
of
a
of
every
software
product.
So
for
that,
you
need
to
plan
how
to
put
it
to
plan,
how
you
will
scale
your
network
and
how
you
can
keep
it
going
as
smooth
as
you
can.
B
First,
we
had.
We
took
this
idea.
Blockchain
is
incoming
is
coming
and
we
saw
that
mostly
as
a
threat.
Why?
Because
blockchain
is
a
disruptive
technology,
is
a
decentralizing
technology
and
as
a
stock
in
stock
exchange.
It's
the
is
a
centralizing
entity,
so
we
took
the
best.
We
analyzed
it
in
conjunction
with
IBM
and
Surya,
a
series
of
use
cases
and
we
selected
short
sale
lending
as
the
best
use
case
to
implement
a
minimal
Viable
Product.
Why?
It's
that?
B
Because
it
doesn't
require
regulatory
changes
and
all
the
benefits
comes
mostly
from
optimization
in
the
current
processes,
so
we
then
created
our
Minimum
Viable
Product
instead.
Well,
there
is
a
lot
of
POC
out
there
that
stays
in
the
POC
stage
right
or
as
jasonkelly
from
IBM
once
tells
us
Pio
I
proof
of
incompetence,
okay!
Well,
this
is
not
a
POA.
It's
a
real
network
that
is
running
almost
from
two
years
and
the
core
of
the
network
is
the
Santiago
exchange.
B
We
are
the
administrator
of
the
network,
we
run
the
certificate
authority
service,
so
we
decide
who
gets
in
and
we'll
get
who
gets
out
each
participant
in
this
network,
mainly
brokerage
houses
and
the
regulator
can
own
their
own
peer
node.
We
have
two
kinds
of
node
passive,
active,
active
peer
nodes
that
can
read
and
write
and
passive
peer
nodes
that
only
need
to
consume
information.
B
This
is
a
market.
Why
a
network
it
interconnects
17
institution
in
Chile,
through
mainly
3
different
ways.
We
have
the
most
the
last
exciting
way
is
through
graphical
user
interface.
It's
for
the
brokers
that
don't
want
or
don't
have
the
expertise
to
do
deep
integration
with
their
system,
so
they
use
the
trading
a
stock
exchange
trading
system
and
it
looks
like
seen
a
Tesla
car
from
outside
it
looked
like
a
car.
B
Then
we
have
the
API
way
to
interconnect
to
rest
services
is
like
sitting
on
a
Tesla
car.
It
looks
nice,
but
it
still
looks
like
a
car,
but
it
allows
you
to
do
more
deep
integration,
and
then
we
have
the
tier
way
and
its
own
in
node
appear
node.
That
is
like
driving
the
car
right
now
you
see
that
it
drives
differently.
So
in
this
way
we
have
a
bank
that
has
their
own
peer
and
we
have
a
broker
and
a
bank
with
their
own
peer.
B
We
have
two
interconnected
in
two
entities
through
this
way
and
for
them
they
can
use.
The
information
like
is
their
own
database.
Okay,
for
the
first
version.
While
we
started
in
development
with
fabric
0-6
beta,
we
were,
we
would
like
to
be
tested
on,
and
the
first
version
in
production
was
with
fabric
1.1.
B
B
B
We
took
three
months
of
testing
with
all
the
market
participants.
We
had
to
do
some
minor
final
adjustment
before
going
live
and
mainly
is
Bay.
It
blocked
an
a
fabric
Network.
We
control
this
certificate
authorities.
We
can
connect
through
secure
channels,
mainly
VPN,
MPLS
or
fiber
channels.
Other
participants-
and
this
as
you
see
in
yellow,
are
the
legacy
system
that
are
integrated
with
the
current
fabric
network.
The
launch
date
was
June
2018
and
we
have
well
two
metrics
of
usage.
Master
service
agreements
is
the
enable
contracts
to
operate
in
short,
sells
currently
well.
B
For
this
year
we
have
39
contracts
done,
but
when
we
went
live,
we
almost
triplet
the
number
of
contracts
in
just
a
one
year
with
it
the
same
number
of
contracts
that
within
in
ten
years
before,
because
we
reduce
it
the
time
to
create
and
sign
a
contract
from
four
days
to
90
seconds
and
the
landing
operation
we
almost
a
is
not
doublet
is
tenfold.
The
number
of
landings
from
when,
when
we
went
a
live
with
the
system,
okay,.
B
First,
the
first
lesson
that
we
learned
when
we
have
this
net
were
running
for
so
much
time
is.
The
first
problem
is
how
you
add
new,
fresh
new
infrastructure
and
new
participants
to
your
current
network.
It's
very
painful!
If
you
do
this
by
hand,
we
have
a
identified
trading
nation
for
this
kind
of
scaling.
That
is
horizontal
scaling,
so
we
have.
B
So
we
have
first
different
layers
for
furrow
solution.
You
have
neat
hardware,
you
can
use
in
provider
or
on-premise
over
that
you
have
the
typical
network
infrastructure.
You
need
firewall,
you
need
security,
you
need
the
monitor.
You
need
to
monitor
your
wall.
Infrastructure
name
resolution
cover
that
you
have
your
container
layer
that
is
important
to
skill.
B
With
your
container
layer.
You
need
first,
a
new
script,
an
Orchestrator.
We
started
with
Dockers
worm
and
you
create
overlay
needles
and
over
that
you
have
your
fabric,
Network,
separated
in
channels,
and
you
create
specific
services
to
access
your
blocking
network
and
platform
services
for
all.
Your
business
need:
okay,
first,
how
you
do
the
internal
grout
or
the
internal
or
is
in
horizontal
skill.
B
What
we
did
is
moving
to
from
Dawkins
worm
to
Cuban
$8.00
worm
is
very
good
to
create
quickly
and
easily
dock
container
Orchestrator
and
network,
but
is
very
lacking
from
the
administrator
administration
perspective.
So
the
first
thing
that
we
have
to
do
is
switch
Orchestrator
and
currently
we
are
using
Cuban
aids
in
our
testing
environment
and
also
for
a
production.
B
Ok,
the
second
dimension
is
how
you
integrate
an
external
participant.
This
is
the
most
painful
point
that
we
identify
for
horizontal
growth
because
they
are,
they
have
their
own
infrastructure,
they
have
their
own
hardware
provider,
it
can
be
on-premise,
it
can
be
on
cloud,
they
have
their
own
network
and
container
infrastructure,
and
you
can
suggest
things,
but
you
cannot
mandate
so
how
you
do
this
first,
we'll
start
with
an
attack
install,
but
it's
not
easily
replicated.
B
How
you
can
easily
provide
also
an
external
growth?
Is
we
have
right
now,
a
customer
that
doesn't
want
to
provide
their
own
infrastructure?
So
I
don't
want
mention
voice,
but
they
came
to
us
and
they
say:
I
want
infrastructure
in
some
place
of
the
world
in
the
u.s.
exactly
precisely
and
I,
don't
want
to
create
any
infrastructure.
Ok,
we
as
a
stock
exchange
will
provide
you
a
node
as
a
service.
So
it's
easier
way
to
grow
your
infrastructure,
because
now
you
are
in
control
of
all
the
configuration
software
version,
etc.
B
Ok,
that
is
for
how
you
grow
your
network,
but
look
let's
look
inside
the
software
Factory
right
now:
ok
to
synthesize
the
effort
with
the
maintenance
team.
We
have
three
different
issues
that
has
to
be
solved.
First
is
training
the
research
and
development
team
that,
in
this
case,
I'm
in
charge
with
that
team
has
all
tech
knowledge.
We
are
we
hug.
We
concentrate
all
the
information,
because
is
the
team
that
created
the
Minimum
Viable
Product?
B
B
When
you
go
to
production,
your
solution
will
be
as
good
as
you
can
to
go
to
production.
You
have
a
schedules
to
meet,
so
some
sacrifice
will
be
done,
but
to
maintain
this
in
the
long-term,
you
have
to
do
a
lot
of
hardening
because
the
rest
of
your
solution,
reproduction-
has
a
higher
grade
for
standards
from
a
higher
standard
for
a
security.
B
Okay,
well,
this
is
a
DevOps
structure
that
we
are
currently
working
together
with
IBM
is
helping
us
with
a
consultancy
services,
and
we
identified
that
the
main
effort
that
you
have
to
do
is
culture.
You
have
to
establish
a
new
culture.
You
need
to
have
all
your
maintenance
team,
your
developers,
designer
security
managers
and
product
owners
and
operator
team
on
this
new
culture
of
decentralized
application,
the
centralized
solution
and
it's
a
process
that
can
take
a
lot
of
time
and
is
not
easy.
B
B
B
Also,
since
you
created
a
dock,
API
is
Andres
services
for
your
solution.
You
have
to
train
about
them.
Also-
and
this
is
the
hardest
part-
is
training
them.
How
to
use
containers
and
orchestrators
normally
when
use
that
a
lot,
but
is
a
very
obscure
knowledge
and
how
to
create
or
maintain
chain
code,
since
they
will
see
that
as
a
traditional
program.
But
you
know
it's
not
a
traditional
is
a
state
machine
and
has
distributed
state
machines.
B
So
you
need
to
explain
how
to
create
a
distributed,
a
code
that
has
to
work
in
conjunction
and
finally,
the
development
and
maintenance
team
have
to
catch
up
on
the
development
tools,
a
integrated
development
environment,
a
continuous
in
Tunisian
system,
new
programming
languages,
etc
for
the
product
owners
instead
and
the
business
analyst.
You
also
have
to
train
in
about
a
blotching
core
concept:
hyper
ledger
fabric
concepts,
how
to
integrate
a
distributed,
ledger
technologies
to
the
business
process
and
how
to
design
solution
base
it
on
April
logic
fabric.
And
finally,
it's
good.
B
B
Next,
how
you
test
your
application?
Okay,
let's
move
on!
First,
you
need
that
your
development
and
maintenance
team
has
to
be
trained
in
how
to
do
distribute
the
testing
over
many
peers.
You
just
don't
have
to
test
on
your
local
machine.
Is
the
typical
development
ask
for
hey,
but
it
looks
and
manage
on
my
computer
right,
so
it
has
to
work
everywhere.
It's
this
is
not
for
a
blocking
solution.
You
have
to
test.
Also
you
in
your
test.
B
You
have
to
consider
also
external
organization
that
participate
in
your
network,
so
you
have
to
option
or
Mele
or
you
onboard
them,
on
your
test
phases
and
you're
on
your
distributed
test
or
you
simulate
them
on
your
test.
Normally,
we
use
option
two.
We
have
a
placeholder
block
attached
house
for
our
testings
and
finally,
you
have
to
train
how
you
build
package
and
deploy
the
chain
code
for
the
business
analyst
and
the
product
owner.
You
have
to
train
them,
how
to
do
the
functional
testing
for
distributed
application
and
for
the
operator
and
configuration
management
team.
B
Next
is
the
cultural
change
for
staging
and
production
environment
so
for
staging
and
I
will
be
talking
about
a
bit.
This
in
deep
later
is
how
to
execute
the
setup
of
the
staging
environment
and
how
to
test
your
solution
there
for
the
operator
and
configuration
management
team.
Also
I
will
be
explaining
this
a
bit
a
more
in
detail.
How
you
not
do
the
network
upgrades
and
the
network
expansions
network
expansion
I
mean
adding
new
organizations
or
new
peers
and
for
the
security
team.
B
Next
for
the
hardening
you
have
normally,
when
you
do
hardening
of
any
solution,
you
just
consider
the
hardware
or
the
virtualization
platform
until
the
operating
system,
operative
system
and
some
basic
services,
databases
directories,
a
web
servers,
etc.
But
we
saw
that
you
need
to
consider
a
lease
three
more
dimension
in
your
hair
lining.
Do
you
have
to
harden
the
notes
itself,
the
ledger
and
the
applications?
B
For
the
note,
for
example,
you
need
to
do
some
vulnerability
scan
because
they
are
pasted
on
the
notes
are
pasted,
since
our
container
are
based
on
a
base
container
that
can
be
scan
it.
You
have
to
do
a
design
review
of
your
network
topology
redundancy
test.
What
happens
if
a
node
goes
down
synchronization
test?
Well,
you
have
to
check
if
everything
goes
out
of
synchronization,
the
clock
is
cubed
for
every
node.
What
happened
you
also
could
do
some
tests
on
the
consensus
algorithm.
B
You
have
to
test
what
happens
if
in
for
me
know
you
had
you
have
to
test
if
you
have
information
leak,
because
since
the
smart
contract
are
programmed
by
humans,
the
humans
can
leak
some
information
and
they
cannot
foresee
that,
since
we
are
using
as
an
example,
we
are
using
channels
and
we
have
synchronization
between
them.
Maybe
some
information
doesn't
have
to
be
synchronized
between
some
channels.
You
have
to
check
the
functionality
of
the
chain
code.
B
You
have
to
do
review
pay
review
of
the
chain
code
to
check
if
we
secure
a,
not
a
dynamic,
a
code,
analysis
and
manual
code
analysis
over
the
chain
code
that
is
deployed.
Finally,
your
application
sees,
in
your
case
our
main,
a
merely
rest
services
base
it.
You
have
to
do
your
test
over
the
api's
and
also
you
can
apply
the
Ava's
top
ten
to
your
application.
B
B
Okay
moving
next,
your
solution
has
to
evolve
and
intake,
in
your
case,
integrate
new
markets.
So
we
added
technical
processes
to
our
solution
and
also
added
as
an
example.
New
markets
and
those
new
markets
propose
new
challenges.
We
added
in
January
2019
fixed
income
to
our
network
to
our
network
and
we
are
finalizing
the
test
for
adding
international
custodians
and
collaterals.
B
Okay,
so
for
the
technical
process
that
were
added,
we
added
business
checks
for
the
external
components
of
the
network.
We
added
more
detail,
eight
monitoring
and
alerting
system.
We
created
a
process
to
mirror
a
bit
to
do
mirroring
between
production
and
staging,
and
also
we
had
to
do
several
fabric
upgrades
version
applets,
and
we
are
doing
right
now
at
immigration
of
the
consensus
protocol,
also
from
Kafka
to
raft
okay.
Why
do
you
need
business
checks?
B
So
what
we
did
is
establish
a
automatic
check
for
some
business
processes
that
require
consistency
and
use
it,
a
single
tool
to
consolidate
the
checks
and
the
repair
actions
or
maintenance
team,
as
an
example,
is
profile
sent
in
the
use
of
Jenkins.
So
we
create
a
lot
of
automated
checks
on
our
Jenkins
Orchestrator,
and
for
that
you
need
to
plan
transaction
to
repair.
Repair
I
mean
repair
to
bring
to
the
correct
state.
B
Your
data
in
the
chain
code
as
an
example,
we
have
a
market
data
client
that
is
a
program
based
on
the
financial
exchange
protocol,
fixed
protocol
and
it's
data
feed
consumer
for
a
market
operation.
Normally
in
a
normal
day,
it
will
consume
short
sale,
traits
information
and
injects
to
the
fabric
network
through
specific
transaction,
but
this
program
can
fail
sometime
and
if
that
fails,
we
have.
We
have
a
consistency
check
that
is
executed
in
specific
hour
to
preemptively
preemptively
alert
that
there
is
a
inconsistency
and
the
alert
will
be
sent
by
email
in
this
case.
B
Next
we
have
that
you
need
to
have
good
monitoring
and
alert
system.
The
idea
here
is
to
cover
at
least
the
four
matrix
of
golden
rule
for
any
system:
latency,
the
traffic
rate,
the
error
rate
and
the
saturation
saturation
as
a
understood
as
the
number
of
CPU
claims
that
are
not
met
by
the
orchestrator
or
also
you
can
use
a
CPU
starvation
as
a
replacement
or
as
I
like
to
understand.
That
is,
please
give
me
more
course.
B
Okay,
for,
since
we
went
in
production,
we
identified
the
critical
metric
for
fabric
for
the
fabric
network.
In
our
production,
environment
is
the
couchdb
saturation
if
the
CouchDB
is
start
with
a
cpu
starve
it.
We
know
that
we
have
serious
problem
and
for
the
external
component
is
the
APA
error
rate
and
the
a
latency
that
will
signal
immediately
to
us
that
we
have
a
problem
also,
and
for
that
you
have
to
establish
real-time
alert
mechanism
by
email,
instant
messaging,
SMS,
etc.
B
Okay
and
well
in
this
diagram,
this
is
the
streamlined
version
of
our
solution.
We
are
using
this
for
a
new
development
that
we
are
creating
over
a
fabric
network.
It's
a
very
generic
way
to
understand
what
are
we
doing
and
see
if
you
have
the
end
user
system
with
the
graphical
user
interface,
an
external
system,
you
have
the
communication
layer
with
the
firewall
and
load
balancer,
the
external
app
access
of
the
applications
with
the
business
API
and
apps,
and
the
authentication
authorization
service.
B
You
have
the
internal
access
layer
for
your
application,
with
the
integration
API
and
also
with
a
smart
contract.
Query
Kiche
that
we
have
created
to
optimize
some
queries
that
we
there
are
very
slow.
You
have
the
common
service
API,
for
example,
reports
API,
mailing
skelos
tasks,
legacy
access.
Finally,
we
have
two
generic
api's,
a
client
block,
a
blockchain
client
and
a
blockchain
administrator
API,
and
we
have
our
blockchain
runtime.
That
is
mainly
based
on
IP
layer
fabric.
B
So
we
have
created
an
environment
mirroring
process
because
we
have
to
test
any
new
release
over
the
same
production
hardware
or
the
same
production
cluster,
so
how
we
can
test
production
without
injecting
a
data
into
production.
We
create
a
clone
of
our
production
environment
over
production,
but
we
only
clone.
First,
we
create
a
clone
of
the
fabric,
Network
an
exact
clone
using
the
orchestrator,
and
we
create
a
copy
of
the
integration
api's
to
access
the
clone
network,
so
the
other
all
the
system
doesn't
know
that
is
accessing
a
staging
environment.
Using
the
production
entry
points.
B
Okay
moving
next,
how
would
we
have
done
upgrades
to
the
fabric
network?
You
have
you
can
upgrade
your
node.
You
can
update
your
client
SDK
your
chain
code,
SDK,
or
also
the
jingle
itself,
the
consensus
protocol,
and
also
the
application
and
services
that
are
satellites
that
are
working
with
your
solution.
B
Okay,
first
for
the
fabric
network
upgrade
we
have
identified
that
it
affects
mainly
just
the
order
and
the
peer
node.
It
doesn't
affect
the
rest
of
the
solution,
because
it's
from
the
upgrade
is
the
easiest
one.
You
have
just
to
upgrade
the
image
container
version
and
you
are
done
unless
you
want
to
also
upgrade
the
capabilities
of
the
network.
B
Next
is
the
client
SDK
for
appraising
the
client
SDK.
You
will
affect
the
peer
nodes,
your
a
access,
API
and
also
your
integration
API
and
as
a
bonus,
we
also
saw
that
every
time
we
are
upgrading
the
client
SDK,
we
are
also
affecting
the
smart
contre.
It
is
my
contact
query.
Kajam
next
is
the
chain
code
SDK.
It
will
affect
directly
the
smart
contract.
B
And
the
consensus,
the
consensus
effect
mainly
the
peer
nodes,
the
order
in
notes
and
your
eye
access
API,
but
also
reads:
backup
and
a
shutdown,
partial
or
total
shutdown
of
your
network
dependently.
It
depends
on
how
you're
doing
this
it's
and
from
the
business
aspect
is
the
change
that
a
doesn't
give
you
a
business
value.
B
How
you
can
justify
this.
This
upgrade
is
very
hard,
but
you
have
to
do
that
and
finally,
the
application
and
services
you
will
mainly
affect
the
or
your
a
services,
API
integration
and
your
business
apps,
but
also
it
might
require
all
the
previous
upgrade
if
needed,
because
maybe
maybe
you
are
using
new
network
capabilities
for
your
api's
and
also
you
have
to
do
the
consensus
to
upgrade
not
operate,
etc.
B
A
Thank
You
Marcos,
so
the
question
was
what
is
next
now,
so
we
decide
to
work
together
with
those
companies
to
create
an
new
ecosystem
to
run
different
applications
base
it
in
hyper
layer
fabric
GTD
is
a
telecom
company
with
a
lot
of
infrastructure
that
they
will
change.
I
already
talked
about
that
and
DCB
is
the
Chilean.
A
So
we
will
be
working
on
this
during
this
year
and
next
year,
and
this
is
a
representation
of
what
we
want
to
do
now.
So
in
the
bottom,
you
can
see
all
the
things
that
we
have
to
create
to
make
this
possible
every
category
here
has
a
huge
backlog
of
stuff
that
we
have
to
to
develop,
and
we
include
that
the
governance,
because
we
know
that
is
one
of
the
most
important
things
that
we
have
to
create
here
then,
in
the
center
you
can
see
a
solution
providers
layer.
A
We
will
provide
an
SDK
note,
help
desk,
belong
solutions,
etc,
and
you
can
see
spinning
different
networks
by
sitting
in
D,
for
example,
hyper
little
fabric
and
in
the
top
we
will
have
different
apps
running
and
they
will
be
used
by
different
companies
from
different
national
market
and
the
first
application
that
we
will
have.
There
obviously
will
be
the
one
that
market
talk
about
securities
lending,
so
a
road
map,
basically
is.
We
will
develop
and
develop
for
two
years
using
scram,
and
now
we
are
talking
with
different
solution
providers-
different
members
general
members.
A
So
if
you
guys
have
any
app
that
you
think
that
could
work
in
our
country
doctor
with
us
later
and
well,
we're
planning
to
you
have
also
a
hackathon
on
October
the
in
this
year
to
test
or
SDK.
So
that's
it.
Thank
you
so
much
all
of
you
and
if
you
have
any
question
or
something
talk
with
us,
thank
you.
B
Will
be
you
asked
what
a
kind
of
underlying
infrastructure
we
deployed?
Our
solution?
Ok,
well,
I,
wanted
I
didn't
want
to
talk
about
that,
because
we
can
talk
for
hours
about
that.
But
if
you
want
I
can
explain
you
later,
but
we
did
this
originally
in
an
I
prayed,
a
solution
on
premise
and
in
cloud.
So
we
wanted
to
test
the
technology
and
see
if
we,
a
cool,
create
an
ivory
network
based
on
cloud
providers
and
around
on-premise
infrastructure,
but
is
from
the
orchestrator
perspective,
is
just
a
local
network.
B
B
B
B
B
B
B
Well,
for
the
staging
environment
is
very
strange
for
any
other
industry,
but
we
have
to
test
everything
every
new
functionality
in
production,
so
for
a
traditional
system
that
I
used
databases
is
very
easy.
You
can
do
a
backup
and
restore
you
do
a
backup.
You
do
all
your
tests
and
you
just
apply
the
Euro
restore,
but
for
blotching
how
you
delete
transactions.
You
cannot
do
that.
So
that's
why
we
created
this
clone
system
that
creates
a
clone
of
the
production
network
over
the
production
environment
and
we
copy
everything
from
the
production
environment.