bitpro가 원문에서 글자만 모은 내용이에요. 표·그림·수식은 빠지거나 깨질 수 있어요. 전체 6쪽 중 1쪽
Space
and
Time
(SXT
Chain)
The
blockchain
for
ZK-proven
data.
Scott
Dykstra
Jay
White,
PhD
Catherine
Daly
Ian
Joiner,
PhD
May
2025
v2.0
whitepaper
Abstract
Smart
contracts
lack
native
data
processing
capabilities,
and
securely
connecting
external
data
is
cumbersome.
Space
and
Time
(SXT
Chain)
addresses
these
challenges
by
introducing
Proof
of
SQL
,
a
breakthrough
zero-knowledge
(ZK)
circuit
that
provides
verifiable
SQL
queries
against
tamperproof
tables.
Delivered
as
a
decentralized
database
loaded
with
ready-to-go
indexed
blockchain
data
compatible
with
Proof
of
SQL,
SXT
Chain
is
ushering
in
a
new
era
of
data-rich
smart
contracts.
By
storing
only
cryptographic
commitments
onchain,
SXT
Chain
secures
an
unlimited
number
of
tables/ledgers
without
bloating
the
blockchain.
Developers
gain
the
ability
to
run
ZK-verified
queries
on
massive
datasets
from
major
blockchains
or
user-provided
offchain
data,
all
while
supplementing
the
limited
compute
power
of
smart
contracts.
This
powerful
combination
propels
the
creation
of
sophisticated,
cross-chain
dapps—ultimately
fulfilling
the
blockchain
vision
of
a
permissionless,
trustless,
and
data-rich
digital
economy.
1
Table
of
Contents
Abstract
1
Table
of
Contents
2
1
The
Problem:
Untapped
Potential
of
Web3
4
1.1
Smart
Contracts
Can’t
Access
Critical
Data
4
1.2
Current
ZK
Solutions
and
Their
Limitations
5
1.3
The
Need
for
an
“Onchain
Database”
5
2
The
Solution:
SXT
Chain
with
Proof
of
SQL
6
2.1
Coprocessing
Unlocked
for
Smart
Contracts
6
2.2
Query
Examples
for
ZK-proven
Onchain
Data
6
2.3
Design
Goals
8
2.4
Components
of
the
Network
9
2.4.1
Indexer
Nodes
9
2.4.2
Prover
Nodes
10
2.4.3
SXT
Chain
Validators
10
2.4.4
Target
Chain
and
Offchain
Components
11
2.5
Ingestion
Layer
with
Verifiable
Indexing
12
2.6
Decentralized
Database
Design
12
2.7
Infinite
Tables
with
Onchain
Verification
13
3
Design
of
the
Proof
of
SQL
Protocol
15
3.1
Overall
Architecture
15
3.1.1
Data
Ingestion
15
3.1.2
Query
Request
17
3.1.3
Proof
Protocol
Overview
17
3.2
Parsing
18
3.3
Query
Executor
and
Interactive
Protocol
Builder
18
3.3.1
Query
Execution
/
Witness
Generation
19
3.3.2
Commitment
Computation
/
Multiple
Rounds
21
3.3.3
Constraint
Building
21
3.4
Notation
21
3.5
Example
Subprotocols
22
3.5.1
Equality
Protocol
Builder
22
3.5.2
Group
By
Protocol
Builder
23
3.6
Relation
Proof
Protocol
(Sum-check)
24
3.7
Commitment
Evaluator
25
4
Use
Cases
for
the
Next-Generation
of
Web3
27
2
4.1
Building
a
Flexible
ZK-Rollup/L2
27
4.2
Secure
Bridges
and
Multichain
Data
Backends
27
4.3
Dapp
Backend
(Decentralized)
27
4.4
Data-Driven
Lending
27
4.5
Cross-Chain
Financial
Instruments
28
4.6
Gaming
Rewards
28
4.7
Social
Apps
28
4.8
Settlement
Systems
and
Third-Party
Auditing
29
4.9
Real
World
Assets
(RWAs)
29
4.10
Transaction
Security
and
Wallet
Whitelists
30
4.11
Liquidity
Pool
Rebalancing
Based
on
Market
Conditions
30
5
Economic
Model
for
SXT
Chain
30
5.1
SXT
Token
Utility
32
5.2
Security
Budget
Vault
32
5.3
Table
Owner
Rewards
Vault
33
5.4
Network
Fee
Structure
and
Data
Marketplace
34
6
Conclusion
34
Appendix
35
A
Data
that
Smart
Contracts
Can
Access
without
Space
and
Time
35
B
HTAP
to
Replace
Point
Solutions
36
C
Proof
of
SQL
“Bring
Your
Own
Database”
37
D
Leveraging
an
Append-Only
Database
as
a
Tamperproof
Offchain
Ledger
39
E
Space
and
Time
in
the
Market
40
E.1
Web3’s
evolution
as
a
digital
economy
powering
novel
apps
40
E.2
Proven
value
in
data
warehousing
41
E.3
New
compute
paradigm
removes
trust
requirements
41
References
43
3
1
The
Problem:
Untapped
Potential
of
Web3
1.1
Smart
Contracts
Can’t
Access
Critical
Data
Smart
contracts
underpin
major
decentralized
finance
(DeFi)
protocols
like
Uniswap,
Aave,
Compound,
and
Synthetix,
each
securing
billions
on
the
Ethereum
Virtual
Machine
(EVM).
While
these
protocols
provide
trustless,
permissionless
asset
management,
they
lack
data-driven
sophistication—Aave,
for
instance,
can’t
differentiate
a
new
borrower
from
one
with
established
onchain
history.
In
traditional
finance
(TradFi),
options
derivatives
exceed
their
underlying
assets’
capital,
with
over
40
million
contracts
traded
daily.
Yet
these
options
are
largely
missing
onchain
because
trustless
deployment
requires
external
data
processing,
which
standard
smart
contracts
cannot
perform.
By
contrast,
onchain
perpetual
futures
see
an
annual
volume
of
$39
trillion,
thanks
to
simpler
logic
that
relies
only
on
token
price
feeds.
Smart
contracts
can
only
access
wallet
balances,
current
transaction
data,
basic
blockchain
metadata,
and
their
own
internal
state.
They
cannot
directly
query
the
historical
data
stored
by
blockchain
nodes.
Using
archive
nodes,
an
external
offchain
client
can
only
access
simple
datapoints
such
as:
●
The
state
of
the
entire
blockchain
at
any
given
time
in
history
(e.g.
who
is
the
first
owner
of
the
“Cryptopunk
#1”
NFT).
●
The
transactions—and
events
emitted
as
a
result
of
transactions—at
any
given
time
in
history
(e.g.
John’s
wallet
swapped
1000
USDC
to
0.5ETH
on
4/18/2023).
Offchain
indexing
solutions
can
parse
this
history
but
introduce
trust
assumptions;
such
solutions
are
typically
centralized
and
unverified.
Decentralized
oracle
networks
bridged
this
gap
by
providing
the
infrastructure
to
aggregate,
validate,
and
deliver
offchain
data
to
blockchains,
but
require
connectivity
to
external
systems
to
handle
complex
datasets,
SQL
execution,
rich
onchain
history,
etc.
As
multichain
protocols
spread,
many
rely
on
a
“hub”
contract
on
Ethereum
linked
to
“spoke”
contracts
on
other
chains.
Monitoring
spoke
activity
in
a
trustless
way
remains
difficult.
Similarly,
Web3
games
struggle
to
reward
in-game
actions
or
onchain
NFT
milestones
because
the
underlying
data
isn’t
directly
accessible
to
the
smart
contract.
Some
teams
resort
to
centralized
services,
undermining
the
trustless
ethos
of
Web3
and
highlighting
an
urgent
need
for
trustless
data
processing.
4
1.2
Current
ZK
Solutions
and
Their
Limitations
ZK
proofs
have
been
heralded
as
the
panacea
for
trustless
data
access,
offering
client
verification
by
allowing
data
to
be
confirmed
without
revealing
its
contents
to
the
client.
By
offloading
computation
offchain
and
verifying
the
results
onchain,
ZK
proofs
can
bolster
the
limited
data
access
and
compute
power
of
smart
contracts.
They
eliminate
the
need
for
cumbersome
offchain
consensus
by
letting
a
single
node
act
as
a
coprocessor
to
an
L1/L2
chain.
However,
current
ZK
solutions
are
still
maturing
and
face
several
key
challenges.
First,
ZK
proofs
lack
scalability
when
handling
large-scale
data.
Proof
generation
often
takes
minutes,
making
ZK
proofs
impractical
for
complex
queries
against,
for
example,
multi-year
transaction
histories.
Second,
the
ecosystem
is
highly
fragmented,
with
some
projects
aimed
at
privacy,
others
at
scaling,
and
still
others
at
specific
use
cases
like
rollups.
Tying
these
disparate
solutions
together
is
time-consuming,
costly,
and
prone
to
error.
Third,
the
developer
experience
remains
unfamiliar
and
demanding,
often
requiring
new
languages
(e.g.,
Cairo)
and
specialized
hardware
deployments
(e.g.,
zkEVM
rollups).
Finally,
existing
ZK
solutions
do
not
address
the
missing
query
layer
in
Web3,
leaving
decentralized
applications
without
a
truly
trustless
way
to
process
and
analyze
large
datasets.
1.3
The
Need
for
an
“Onchain
Database”
In
traditional
SaaS
applications,
business
logic
follows
a
straightforward
cycle
of
querying
data,
executing
actions
based
on
the
query
results,
and
updating
system
state.
This
pattern
is
evident
across
industries:
lenders
check
a
borrower’s
creditworthiness,
decide
terms,
and
update
loan
states;
e-commerce
platforms
query
product
availability,
manage
purchases,
and
adjust
inventories;
travel
services
query
accommodations
and
flights
before
finalizing
bookings;
and
social
networks
query
a
user’s
connections
to
display
and
rank
content.
However,
in
Web3,
while
blockchains
and
smart
contracts
provide
state
management
and
basic
logic,
they
lack
a
dedicated
query
layer
which
limits
data
processing
and
onchain
computation
for
more
advanced
decentralized
applications.
Space
and
Time
addresses
this
gap
by
serving
as
a
“query
coprocessor,”
delivering
robust
offchain
data
processing
alongside
major
blockchains.
Over
the
past
15
years,
tech
conglomerates
have
accumulated
and
monetized
user
data
within
proprietary
cloud
warehouses.
This
centralized
control
conflicts
with
the
Web3
vision
of
a
decentralized
internet,
where
individuals
manage
their
own
information,
maintain
privacy
via
wallet-based
identities,
and
participate
in
transparent
governance.
A
community-operated
database
model
can
help
reclaim
data
sovereignty,
as
end
users
collectively
determine
how
data
is
accessed,
used,
and
monetized.
Through
wallet-based
public
key
infrastructure
(PKI),
individuals
can
sign
off
on
how
their
interactions
are
recorded,
shared,
or
deleted.
5
Yet,
existing
blockchains
primarily
function
as
single-ledger
databases
with
no
native
query
engine,
resulting
in
state
bloat
and
limited
data
capabilities.
Achieving
the
data
flexibility
required
by
modern
decentralized
apps
calls
for
a
more
expansive
approach—one
that
supports
near-infinite
tables/ledgers,
each
dedicated
to
a
specific
crypto
app
domain,
while
preserving
the
trustless
security
model
of
blockchains.
2
The
Solution:
SXT
Chain
with
Proof
of
SQL
2.1
Coprocessing
Unlocked
for
Smart
Contracts
SXT
Chain
addresses
the
need
for
a
query
coprocessor
for
smart
contracts
by
providing
a
tamperproof,
community-operated
decentralized
database
that
indexes
and
verifies
data
from
major
blockchains
or
user-owned
tables
using
ZK
proofs.
Community-operated
validators
store
threshold-signed
cryptographic
commitments
(one
per
table)
instead
of
raw
data.
Validators
and
archive
nodes
locally
pin
raw
data
or
act
as
comprehensive
repositories
to
ensure
redundancy
and
availability.
Indexed
data
from
major
blockchains
(e.g.,
Ethereum,
Bitcoin,
ZKsync,
Sui,
Avalanche,
Base)
is
validated
and
summarized
into
onchain
cryptographic
commitments
(e.g.,
KZG,
Nova,
Dory),
which
act
as
tamperproof
digital
fingerprints
of
the
underlying
tables.
When
data
insertions
are
submitted
via
DDL/DML
transactions,
the
network
updates
the
affected
table's
cryptographic
commitment
to
reflect
its
new
state.
Proof
of
SQL
is
a
groundbreaking
ZK
proof
protocol
that
ensures
the
accuracy
of
SQL
query
results
and
the
integrity
of
underlying
data
tables
stored
on
SXT
Chain
using
cryptographic
commitments
specifically
to
ensure
underlying
data
tables
are
tamperproof
as
part
of
the
mechanism
(described
in
a
later
section
of
this
paper).
The
network's
Prover
nodes,
running
the
Proof
of
SQL
protocol,
handle
both
query
processing
and
proof
generation.
The
protocol
targets
online
latencies
while
proving
computations
over
entire
chain
histories
an
order
of
magnitude
faster
than
state-of-the
art
zkVMs
and
coprocessors.
SQL
proving
times
are
currently
benchmarked
at
under
a
second
for
analytic
queries
on
million-row
tables
using
a
single
NVIDIA
GPU.
2.2
Query
Examples
for
ZK-proven
Onchain
Data
Ultimately,
Space
and
Time
allows
smart
contracts
to
“ask
ZK-proven
questions”
regarding
activity
on
their
own
chain,
other
chains,
or
offchain.
For
example,
a
smart
contract
can
request
the
SQL
equivalent
of:
6
●
Loans
paid
off:
“Show
me
the
volume
of
loans
already
paid
off
by
this
wallet
attempting
a
borrow
transaction”
(in
order
to
offer
a
discounted
loan
rate
to
a
loyal
borrower
who
has
paid
off
loans
onchain
already).
●
Liquidity
Pools
TVL:
“Show
me
all
liquidity
pools
with
a
TVL
greater
than
$1M
that
were
deployed
at
least
one
month
ago.”
●
Liquidity
Pools
Collateral:
“Sum
up
the
total
collateral
available
right
now
in
all
liquidity
pools
our
protocol
operates
across
the
following
four
chains…”
●
DEX
Loyalty
Program:
“Given
that
wallet
‘xyz’
is
currently
attempting
a
trade,
show
me
the
number
of
prior
trades
this
wallet
has
already
made
with
DEX
liquidity
pool
address
‘xyz’”
(in
order
for
the
DEX
contract
can
apply
a
discount
or
reward
to
the
current
trade).
●
LP
position
management
(active/dynamic):
“Alert
me
when
the
delta
between
the
average
price
of
the
LINK
and
ETH
has
deviated
more
than
10%”
(for
Uniswap
v4
LP
rebalancing
via
hooks).
●
Avg.
Lending
Rates
Onchain:
“Show
me
the
volume-weighted
average
lending
rate
for
USDC
on
Aave,
Maker,
and
Compound
right
now.”
●
Wallets
with
Token
Balance
Criteria:
“Show
me
all
wallets
that
have
a
balance
>
$1000
of
LINK
tokens
and
have
interacted
with
at
least
one
Chainlink
smart
contract.”
●
Bridge
Transactions:
“Show
me
all
transactions
across
‘xyz’
bridges
that
moved
at
least
$1M
of
tokens
per
wallet
from
Chain
A
to
Chain
B.”
●
Gamer
Achievements:
“Show
me
all
gamer
wallets
that
have
at
least
2
hours
of
playtime
in-game,
have
minted
our
NFT,
and
played
with
‘xyz’
weapon.”
●
CeFi
Options
Markets:
“Give
me
the
at-the-money
implied
volatility
of
Bitcoin
call
options
expiring
end-of-year
averaged
(volume-weighted)
across
Deribit
and
Binance.”
●
Gas
Oracles:
“Show
me
the
average
gas
used
across
all
transactions
on
‘xyz’
bridge
over
the
last
hour.”
●
Token
Price
Oracles:
“Show
me
the
volume-weighted
average
price
of
LINK
and
wrapped
LINK
tokens
traded
across
the
hundreds
of
liquidity
pools
that
swap
LINK:
Uniswap,
PancakeSwap,
Sushiswap,
Trader
Joe,
Quickswap,
etc.”
●
Trustless
ETFs/Indexes:
“Create
a
volume-weighted
index
of
the
following
10
tokens
calculated
from
price
oracle
SQL
calculations
and
deliver
it
to
my
contract
as
a
single
price
index.”
●
Airdrop
Criteria:
“Roll
up
the
wallet
transaction
histories
of
all
wallets
that
meet
the
following
criteria
for
my
airdrop…”
●
Governance
Engagement
Score:
“Roll
up
engagement
scores
for
wallet
activity
that
meets
the
following
criteria…
where
volume
transacted
in
‘xyz’
token,
onchain
governance
voting
participation,
and
past
NFT
ownership
will
contribute
to
rewards/discounts
associated
with
a
wallet
engagement
score.”
7
●
Governance
Qualification:
“Find
all
wallets
that
have
participated
in
governance
votes
and
rank
them
with
a
cross-section
of
the
amount
of
‘xyz’
token
held
and
how
long
it’s
been
held.”
●
Weather
(Onchain
Insurance
Payouts):
“Show
me
the
average
wind
speeds
across
all
reported
weather
stations
in
Miami,
Florida
today.”
●
Github
Activity
(Open
Source
Contributions):
“Rank
developer
wallets
by
their
total
code
contributions
through
ingested
GitHub
activity
data.”
●
Oracle
Activity:
“Show
me
total
volume
of
oracle
requests
over
the
last
hour
that
did
not
complete
the
request
within
two
blocks.”
●
Pricing
Blockspace:
“Determine
blockspace
‘market
rates’
by
averaging
transaction
gas
fees
along
with
block
header
metadata.”
●
NFT
Floor
Prices:
“Look
at
NFT
trading
information
onchain
and
find
the
current
average
floor
price
for
collection
‘xyz.’”
●
Perp
Funding
Rates:
“Calculate
the
implied
volatility
of
both
ETH
and
BTC
using
onchain
perpetual
futures
funding
rates
as
a
leading
indicator.”
●
Ethereum
Staking
Yields:
“Show
me
the
risk-minimized
average
returns
on
Ethereum
liquid
staking
via
Lido
over
the
past
hour.”
2.3
Design
Goals
The
Space
and
Time
solution
was
designed
with
the
following
goals
around
data
processing:
1.
End-to-End
Trustless
:
Leverage
a
ZK
approach
for
proving
rather
than
building
yet
another
consensus-based,
‘trust-minimized’
approach
with
12
to
30
nodes.
This
removes
reliance
on
a
manually
configured
set
of
permissioned
nodes
while
offering
improved
scalability.
2.
Verifiable
Onchain:
Ensure
ZK
proofs
generated
offchain
are
verifiable
within
the
computational
capabilities
of
the
EVM,
rather
than
only
verifying
offchain
using
a
third-party
service
such
as
an
oracle
or
relayer.
Thus,
a
smart
contract
on
any
major
chain
(consensus-based
or
ZK-rollup
chains)
can
verify
the
proof
and
relay
the
verified
query
result
to
a
client
contract
which
made
the
request.
3.
Familiar,
Easy
Developer
Experience
:
Provide
a
UX
common
in
Web2
(SQL)
so
developers
don’t
have
to
learn
any
extensive
new
frameworks
or
languages.
4.
Support
for
Arbitrary
Computations
:
Facilitate
common
data
processing
jobs
that
require
aggregations,
sorts,
filters,
arithmetic,
joins,
etc.,
as
well
as
Turing-complete
arbitrary
computations
using
SQL.
5.
Low-Latency
at
100
Gigabyte-Scale
:
Deliver
verified
query
results
back
to
a
client
smart
contract
within
seconds—not
minutes—using
a
more
scalable
approach
than
8
redundant
computations
proven
with
consensus.
Must
support
queries
over
the
entire
chain
state
(often
multiple
terabytes
of
data
per
chain
since
genesis,
broken
up
into
sub-tables
that
often
exceed
100
gigabytes
each).
6.
Comprehensive
Chain
Data
Coverage
:
Persist
entire
copies
of
each
major
chain
(including
all
events,
transactions,
blocks,
logs,
balances,
token
price
changes,
etc.)
to
facilitate
cross-chain
queries
against
current
and
historical
activity.
Given
the
limitations
of
arbitrary
ZK
proof
circuits
and
the
potential
strengths
of
a
circuit
constructed
around
a
SQL
parser,
we
considered
the
following:
1.
Optimized
Circuit
Design:
We
narrowed
our
initial
focus
to
only
SQL
operations
in
order
to
create
highly
optimized
ZK
circuits
tailored
for
data
processing.
This
level
of
specificity
allows
for
streamlined
verification
and
reduced
computational
overhead.
2.
Broad
Applicability:
SQL
is
ubiquitous—a
widely
understood
and
universally
accepted
language
for
data
operations
that
operates
at
the
heart
of
both
complex
financial
systems
and
basic
web
applicat프로젝트로 돌아가기