bitpro가 원문에서 글자만 모은 내용이에요. 표·그림·수식은 빠지거나 깨질 수 있어요. 전체 4쪽 중 3쪽
t
two
possible
approaches.
Return
Value
:
The
foreign
call
to
the
precompiled
“system”
contract
could
return
the
chunks
specified
by
the
range
as
a
buffer.
23
Global
State
:
Once
the
precompiled
“system”
contract
has
been
invoked
the
chunks
become
accessible
via
a
global
that
is
exposed
to
all
subsequent
instructions
in
the
PD
transaction.
7.9.2
Calculating
Compute
Units
Because
the
execution
of
a
particular
smart
contract
function
may
take
one
code
path
or
another
depending
on
the
data
read
from
the
chunks,
calculating
compute
units
(CUs)
can
be
problematic.
There
are
a
few
possible
approaches
Simulate
With
Chunks
:
The
only
way
to
deterministically
simulate
the
CUs
required
to
complete
the
execution
of
the
instruction
is
to
have
the
unpacked
chunks
available
during
the
simulation.
This
would
require
the
simulating
node
to
retrieve
the
unpacked
chunks
during
the
simulation
request.
Simulate
Compute
Upper
Bound
:
In
this
case,
the
simulation
would
evaluate
all
code
paths
and
return
the
cost
of
the
most
compute-intensive
code
path.
This
way
the
user
always
pays
enough
gas
for
any
possible
computational
resources.
7.10
Programmable
data
roadmap
Blob
Data
-
MVP:
The
first
version
of
PD
transactions
will
expose
chunks
as
buffers
or
blob
data
to
the
contract
and
leave
the
interpretation
of
these
bytes
up
to
the
caller.
This
will
allow
PD
chunks
to
have
any
structure
or
format
the
caller
can
imagine.
Bundle
Format
v.1
-
Bundles:
Once
the
blob
chunks
are
working,
the
next
layer
of
functionality
will
be
a
DX
upgrade
that
allows
PD
contracts
to
load
a
bundle
and
data
items
from
the
chunks.
The
IrysVM
will
parse
the
chunks
in
the
range
specification
as
a
v1
bundle
format.
Bundle
Format
v.2
-
DataItems:
Once
the
v2
(merkelized)
bundle
format
exists,
a
DX
upgrade
will
allow
parsing
of
specific
data
items
from
a
larger
bundle
or
retrieve
smaller
nested
bundles
(or
their
data
items)
by
loading
only
the
chunks
that
store
the
specific
data
items
the
caller
is
interested
in.
Programmable
data
L2s:
In
the
future,
we
expect
users
will
develop
programmable
data
L2s
that
expand
the
compute
capacity
beyond
a
single-state
machine.
The
end
goal
here
is
a
shared
dataset
with
the
ability
for
anyone
to
spin
up
L2s
to
tap
into
data,
compute,
and
liquidity
resources.
8
Tokenomics
The
Irys
native
token,
IRYS,
underpins
the
economic
and
security
architecture
of
the
Irys
protocol.
Its
design
integrates
five
foundational
components
that
collectively
align
incentives
across
users,
miners,
24
and
the
broader
ecosystem.
Its
role
is
to
align
incentives
across
all
participants,
from
early
adopters
and
community
members
to
institutional
partners
and
long-term
investors.
8.1
Token
distribution
The
initial
supply
of
IRYS
is
10,000,000,000
tokens.
At
the
Token
Generation
Event
(TGE),
the
circulating
supply
will
be
approximately
2,000,000,000
tokens,
representing
20.0%
of
the
maximum
supply.
Figure
16:
Irys
Token
Allocation.
Figure
17:
Token
Distribution
Schedule
25
Ecosystem
(30%):
This
allocation
will
be
used
for
various
Irys
initiatives,
including
retaining
and
incentivizing
novel
decentralized
applications,
as
well
as
various
cross-chain
initiatives,
partnerships,
and
more.
The
allocation
will
be
held
by
a
secure,
multisig
wallet
in
collaboration
with
the
top
custodians
in
the
space.
Foundation
(9.9%):
This
is
used
to
further
initiatives
that
serve
to
widen
the
reach
of
Irys,
reducing
crypto’s
reliance
on
centralized
data
storage
and
fragmented
developer
experiences.
This
$IRYS
allocation
will
be
used
to
fund
further
development,
risk
assessments,
audits
and
more.
Core
Team
and
Advisors
(18.8%):
This
portion
of
the
$IRYS
allocation
(18.8%)
represents
the
distribution
to
the
core
contributors
to
the
Irys
Layer
1
Protocol,
as
well
as
advisors
who
have
worked
to
bring
the
L1
protocol
to
market.
All
core
contributors
are
locked
on
a
1-year
cliff,
with
3-year
linear
monthly
vesting
thereafter.
No
core
contributor
tokens
are
unlocked
prior
to
the
1-year
cliff,
and
no
staking
rewards
will
be
unlocked
in
advance
of
vesting
schedules.
Airdrop
and
Future
Incentives
(8%):
This
allocation
is
designed
to
recognize
meaningful
contributions
that
helped
bring
Irys
to
market.
A
portion
of
this
allocation
is
reserved
for
future
incentive
programs
designed
to
attract
aligned
contributors,
developers,
and
users
over
time.
This
ensures
that
Irys
continues
to
grow
with
the
right
people,
not
just
the
loudest.
Liquidity,
Market
Stability
and
Launch
Partnerships
(8%):
This
allocation
will
be
deployed
to
market
makers
and
launch
partners
to
establish
robust
liquidity
profiles
from
day
one.
Our
goal
is
to
create
an
efficient
trading
environment
that
supports
organic
volume
growth
and
prevents
manipulation.
By
anchoring
this
allocation
to
strategic
partners
and
a
mix
of
CEX
and
DEX
infrastructure,
we
ensure
Irys
is
accessible,
liquid,
and
trade-ready,
without
compromising
long-term
sustainability.
Investors
(25.3%):
This
represents
token
rights
obtained
by
investors
backing
the
Irys
protocol’s
development,
to
bootstrap
the
protocol.
All
investors
are
locked
on
a
1-year
cliff,
with
3-year
linear
monthly
vesting
thereafter.
No
investor
tokens
are
unlocked
prior
to
the
1-year
cliff,
and
no
staking
rewards
will
be
unlocked
in
advance
of
vesting
schedules.
8.2
Token
utility
The
token
will
have
all
of
the
following
utilities
at
launch:
Native
Asset:
$IRYS
serves
as
the
native
utility
token
of
the
Irys
network,
functioning
as
the
primary
medium
for
payments,
collateral,
and
settlement
across
all
protocol-level
operations.
Fees:
All
network
actions—including
data
uploads,
programmable
data
executions,
and
contract
interactions—are
denominated
in
$IRYS.
Unlike
traditional
datachains,
Irys
employs
a
USD-pegged
26
pricing
model
for
both
temporary
and
permanent
data
storage,
recalibrated
annually
to
reflect
real-world
storage
costs.
This
ensures
stable,
predictable
economics
for
users
and
developers
independent
of
token
volatility.
Security:
Token
emissions
and
rewards
are
allocated
to
validators
and
miners
who
maintain
consensus
integrity
and
data
availability.
This
incentivization
structure
provides
economic
protection
against
spam,
denial-of-service
attacks,
and
other
adversarial
behaviors,
ensuring
that
network
reliability
scales
with
participation.
Endowment:
A
portion
of
$IRYS
is
directed
into
a
protocol-level
endowment
that
underwrites
miners’
long-term
storage
commitments.
The
endowment
functions
as
a
self-sustaining
reserve,
covering
future
storage
liabilities
and
reinforcing
the
permanence
guarantees
of
the
network’s
data
layer.
Staking:
Miners
and
validators
must
stake
$IRYS
tokens
as
bonded
collateral
to
participate
in
consensus
and
data
verification.
This
mechanism
enforces
accountability—malicious
or
negligent
behavior
results
in
slashing—and
ensures
that
economic
security
is
directly
tied
to
the
integrity
of
the
network.
Delegation
mechanisms
allow
token
holders
to
contribute
to
network
security
indirectly,
earning
rewards
proportional
to
their
delegated
stake.
8.3
Token
emissions
Each
block,
new
IRYS
tokens
are
minted
as
block
rewards
to
incentivize
miners
for
their
contributions
to
network
security
and
data
storage.
Irys
adopts
a
predictable
inflation
schedule
modeled
on
a
traditional
decay
curve
to
ensure
long-term
sustainability.
The
initial
annualized
inflation
rate
is
approximately
2%,
halving
every
four
years
until
it
reaches
a
terminal
rate
of
0.25%.
This
gradual
reduction
aligns
token
issuance
with
network
maturity;
front-loading
incentives
to
bootstrap
storage
capacity
and
decentralization,
then
transitioning
toward
a
deflationary
equilibrium
as
protocol
usage
and
fee-based
rewards
become
the
dominant
economic
drivers.
8.4
Burn
mechanism
The
$IRYS
token
economy
is
designed
to
exhibit
strong
deflationary
dynamics
early
in
the
network’s
lifecycle,
gradually
transitioning
toward
equilibrium
as
usage
demand
offsets
the
decay
of
issuance.
Inflationary
rewards
distributed
through
block
subsidies
are
programmed
to
decline
over
time,
while
multiple
burn
and
sink
mechanisms
continuously
remove
tokens
from
circulation,
ultimately
creating
a
net
deflationary
effect
as
network
activity
scales.
The
primary
source
of
this
deflation
arises
from
long-term
storage
fees.
Payments
made
for
data
stored
in
extended-duration
or
permanent
ledgers
(e.g.,
greater
than
two
weeks)
are
allocated
to
the
protocol’s
27
endowment
contract,
a
reserve
designed
to
cover
miners’
future
storage
obligations.
Because
these
endowment
tokens
are
effectively
locked
indefinitely
to
preserve
the
network’s
data
permanence
guarantees,
they
function
as
a
one-way
economic
sink,
removing
the
corresponding
$IRYS
supply
from
active
circulation.
This
mechanism
ensures
that
as
the
total
volume
of
stored
data
grows,
so
too
does
the
deflationary
pull
exerted
by
the
endowment.
A
secondary
deflationary
vector
is
introduced
through
the
execution
fee
burn
mechanism.
For
every
programmable
data
or
smart
contract
transaction
executed
on
IrysVM,
50%
of
the
associated
execution
fees
are
permanently
burned,
while
the
remaining
portion
compensates
the
block
producer.
This
design
links
token
supply
contraction
directly
to
network
utilization,
meaning
that
as
onchain
activity,
data
composability,
and
programmable
workloads
increase,
a
proportional
share
of
tokens
will
be
destroyed.
Together,
these
two
mechanisms,
endowment
locking
and
execution
fee
burning,
establish
a
feedback
loop
between
protocol
usage
and
token
scarcity.
In
early
phases,
declining
block
rewards
amplify
deflationary
pressure,
while
in
later
stages,
sustained
demand
for
storage
and
computation
ensures
a
continuously
contracting
circulating
supply.
The
result
is
an
economic
system
where
long-term
participation,
data
growth,
and
application
activity
all
reinforce
$IRYS’s
purchasing
power
and
sustainability.
9
Fees
9.1
Minimum
Fee
Parameter
Irys
implements
a
minimum
fee
to
mitigate
network
spam
and
to
ensure
that
tx
fees
can
be
easily
denominated
with
the
atomic
units.
The
minimum
fee
is
$0.001
(1/10th
of
a
cent)
as
determined
by
Irys’
price
approximation
mechanism.
The
same
minimum
fee
is
paid
to
the
provider
of
ingress-proofs
for
publishing
permanent
data.
9.2
Term
Fees
The
pricing
model
for
term
storage
determines
the
cost
of
providing
storage
for
data,
10
replicas
for
n
epochs
Pricing
Parameter
Value
Annualized
Cost
of
operating
16TB
HDD
$44.00
Number
of
Replicas
10
Calculation
Value
28
Daily
Cost
per
TB
$0.0075
Daily
Cost
of
16TB
HDD
$0.12
Total
Fee
Per
Epoch
Storage
Price
(TB)
$0.0753
Total
Fee
Per
Epoch
Storage
Price
(GB)
$0.00007358
Epoch
Fee
Calculator
Data
Size
(TB)
Total
Fee
Per
Epoch
Storage
Price
1
Epoch
1
$0.0753
5
Epochs
1
$0.3767
An
additional
5%
fee
is
added
for
inclusion
in
the
block
(scales
with
the
size
of
transaction
data)
1TB
of
Term
Data
in
the
Submit
ledger
(
5
epochs
)
term_fee
=
term_cost
+
5%
term_fee
=
$0.3767
+
5%
=
0.3955
->
$0.40
1GB
of
Term
Data
in
the
Submit
ledger
(
5
epochs
)
term_fee
=
$0.00039
As
$0.00039
is
below
the
network
minimum
fee
of
$0.01
the
term_fee
becomes:
term_fee
=
$0.01
Note
:
if
repacking
term
partitions
after
they
expire
represents
an
ongoing
expense
to
miners,
this
cost
will
be
quantified
and
included
in
the
term
data
pricing.
9.3
Perm
Fees
The
pricing
model
for
permanent
data
has
some
additional
factors
to
account
for.
Because
the
users
are
paying
for
centuries
of
storage
upfront
the
model
has
to
account
for
declines
in
the
physical
costs
of
storage
(due
to
technological
gains)
over
that
time
period.
Irys
chooses
an
extremely
safe
1%
annualized
decline
in
the
cost
of
storage
as
a
factor
for
pricing
permanent
data.
(Observed
declines
in
storage
costs
over
the
last
50
years
have
been
>
25%
year
on
year.)
Pricing
Parameter
Values
Annualized
Cost
of
operating
16TB
HDD
$44.00
29
Safe
annual
decline
in
cost
of
storage
(decay
rate)
1.00%
Number
of
Replicas
10
Years
of
storage
paid
for
200
Cost
Per
TB
$2,381.54
Cost
Per
GB
$2.33
Because
permanent
data
must
first
pass
through
the
submit
ledger
(term
data)
on
its
way
to
the
publish
ledger,
the
fee
includes
the
cost
of
submit
ledger
storage
as
well.
Perm
data
requires
10
ingress-proofs,
ingress-proofs
are
the
same
as
the
5%
immediate
reward
for
including
the
transaction
in
a
block.
(scales
with
data
size,
shares
the
minimum
$0.01
fee
floor).
1TB
of
Permanent
Data
perm_fee
=
term_fee
+
(ingress_fee
*
10)
+
perm_cost
perm_fee
=
$0.40
+
($0.018835
*
10)
+
$2,381.56
->
$2,382.14
1GB
of
Permanent
Data
perm_fee
=
term_fee
+
(ingress_fee
*
10)
+
perm_cost
perm_fee
=
$0.01
+
($0.01
*
10)
+
$2.33
->
$2.44
If
a
user
fails
to
upload
data
during
the
submit
ledger
term
duration
or
the
network
fails
to
achieve
the
required
number
of
ingress-proofs,
the
user’s
ingress_fee’s
and
perm_cost
are
refunded
when
the
submit
ledger
transaction
expires
at
the
end
of
5
epochs
(the
submit
ledger
term
duration)
9.4
Consensus
Pricing
Mechanism
The
process
of
promoting
data
from
the
submit
ledger
to
the
permanent
ledger
involves
multiple
phases,
resulting
in
a
staged
payment
model
for
permanent
data.
All
transactions,
whether
intended
for
permanent
(perm)
or
temporary
(term)
data,
initially
enter
the
submit
ledger.
The
payment
process
for
term
data
is
consistent
across
all
transactions,
while
permanent
data
incurs
additional
payments
to
incentivize
the
complete
publishing
process.
30
9.5
Term
Data
Payment
Distribution
1.
User
posts
a
transaction,
including
the
term_fee.
2.
Block
producer
transaction
inclusion:
a.
Block
producer
includes
the
transaction
in
a
block.
b.
Block
producer's
balance
increases
by
5%
of
the
term_fee.
c.
Remaining
95%
of
term_fee
is
added
to
the
treasury
(tracked
in
block
headers).
3.
The
user
uploads
data
chunks
associated
with
their
transaction.
4.
Miners
assigned
to
store
chunks
gossip
them
amongst
themselves.
5.
Term
ledger
expiration
payout:
a.
When
the
transaction
expires
from
the
submit
ledger
(when
the
partitions
containing
its
chunks
are
reset
at
an
epoch
boundary),
each
miner
is
paid
their
portion
(term_fee
/
10)
for
all
assigned
chunks
expiring
in
their
partition.
b.
For
a
full
16TB
partition,
this
payout
is
approximately
$0.60
per
miner.
c.
Miners
continue
to
earn
full
inflation/block
rewards
from
any
blocks
they
produce
while
mining
these
partitions.
9.6
Additional
Incentives
This
payment
structure
creates
additional
incentives
for
miners
to
participate
in
term
ledgers:
●
Miners
receive
a
payout
when
data
expires
from
their
partitions.
●
Because
miners
must
re-pack
the
partitions
after
expiration,
this
additional
fee
encourages
ongoing
participation
and
maintenance
of
the
network.
9.7
Permanent
Data
Payment
Distribution
Users
pay
the
following
fees
for
permanent
data
storage:
term_fee:
Standard
fee
for
term
storage
perm_fee:
Fee
for
permanent
storage
5%
of
term_fee
for
block
inclusion
5%
of
term_fee
for
each
ingress-proof
9.8
Fee
Distribution
term_fee:
Processed
identically
to
regular
term
data
transactions.
31
Block
Inclusion
Fee:
5%
of
term_fee
paid
immediately
to
the
block
producer
including
the
transaction.
Ingress-Proof
Fees:
5%
of
term_fee
for
each
ingress-proof
provided.
Perm_fee:
Prepaid
amount
covering
200
years
x
10
replicas
with
1%
annual
decline
in
storage
costs.
Added
to
the
treasury.
10
Submit
Ledger
Expiry
(Epoch
Boundary)
Processing
Refund
Scenario
If
transaction
data
was
never
uploaded:
Ingress-proof
fees
and
perm_fee
are
refunded
to
the
uploader.
Promotion
Scenario
If
data
was
promoted
to
permanent
storage:
Protocol
inspects
all
permanent
transactions
with
ingress-proofs.
Pays
out
the
ingress-provers.
10.1
Epoch
Boundary
Payment
Distribution
Tasks
For
each
expiring
submit
ledger
transaction:
1.
Inspect
the
transaction
to
determine
if
it
was
intended
for
the
publish
ledger.
2.
If
intended
for
publish
ledger,
check
if
it
arrived:
a.
If
Published:
Reward
ingress-proof
submitters
with
their
5%
rewards.
b.
If
not
Published:
Refund
perm_fee
and
ingress-proof
fees
to
the
address
that
posted
the
tx.
3.
Tabulate
the
amount
of
data
posted
to
the
expiring
partition.
4.
Pay
each
partition
owner
the
term_fee
for
storing
that
amount
of
data.
11
Roadmap
11.1
Scaling
programmable
data
Programmable
data,
at
its
core,
represents
a
new
computational
paradigm
where
datasets
are
no
longer
passive
records
but
active,
composable
building
blocks
for
onchain
applications.
It
establishes
a
universal
data
layer:
an
openly
accessible
substrate
where
developers,
users,
and
protocols
can
read,
write,
and
execute
logic
directly
against
verifiable,
onchain
data.
Unlike
traditional
blockchains
that
32
separate
storage
from
execution,
programmable
data
unifies
both,
creating
a
shared
environment
where
information,
compute,
and
value
flow
seamlessly
across
applications.
A
critical
evolution
of
this
model
will
come
through
Programmable
Data
Layer-2s
(PD
L2s):
scaling
environments
that
extend
IrysVM’s
compute
capacity
while
maintaining
trustless
interoperability
with
Irys’s
global
dataset.
PD
L2s
function
as
specialized
execution
domains,
enabling
parallel
computation,
lower
latency,
and
domain-specific
optimizations
without
fragmenting
state
or
data
availability.
They
are
natively
anchored
to
Irys’s
data
layer
through
verifiable
proofs,
ensuring
that
all
reads,
writes,
and
computations
performed
off-chain
remain
cryptographically
linked
to
canonical
data
on
the
base
layer.
Within
this
architecture,
the
Irys
dataset
bifu프로젝트로 돌아가기