bitpro가 원문에서 글자만 모은 내용이에요. 표·그림·수식은 빠지거나 깨질 수 있어요. 전체 5쪽 중 1쪽
1
MultiversX
A Highly Scalable Public Blockchain via Adaptive State Sharding
and Secure Proof of Stake
[Technical whitepaper — release 2 — revision 2]
Updated: renaming to MultiversX
June 19, 2019 - The MultiversX Team
https://www.multiversx.com/
Abstract—The advent of secure public blockchains through
Bitcoin and later Ethereum, has brought forth a notable degree
of interest and capital influx, providing the premise for a
global wave of permissionless innovation. Despite lofty promises,
creating a decentralized, secure and scalable public blockchain
has proved to be a strenuous task.
This paper proposes MultiversX, a novel architecture which
goes beyond state of the art by introducing a genuine state
sharding scheme for practical scalability, eliminating energy
and computational waste while ensuring distributed fairness
through a Secure Proof of Stake (SPoS) consensus. Having a
strong focus on security, MultiversX’ network is built to ensure
resistance to known security problems like Sybil attack, Nothing
at Stake attack and others. In an ecosystem that strives for
interconnectivity, our solution for smart contracts offers an EVM
compliant engine to ensure interoperability by design.
Preliminary simulations and testnet results reflect that Mul-
tiversX exceeds Visa’s average throughput and achieves an im-
provement beyond three orders of magnitude or 1000x compared
to the existing viable approaches, while drastically reducing
the costs of bootstrapping and storage to ensure long term
sustainability.
I Introduction
1 General aspects
Cryptocurrency and smart contract platforms such as Bit-
coin and Ethereum have sparked considerable interest and
have become promising solutions for electronic payments,
decentralized applications and potential digital stores of value.
However, when compared to their centralized counterparts
in key metrics [1], the current state of affairs suggests that
present public blockchain iterations exhibit severe limitations,
particularly with respect to scalability, hindering their main-
stream adoption and delaying public use. In fact, it has
proved extremely challenging to deal with the current engi-
neering boundaries imposed by the trade-offs in the blockchain
trilemma paradigm [2]. Several solutions have been proposed,
but few of them have shown significant and viable results.
Thus, in order to solve the scalability problem, a complete
rethinking of public blockchain infrastructures was required.
2 Defining the challenges
Several challenges must be addressed properly in the pro-
cess of creating an innovative public blockchain solution
designed to scale:
• Full decentralization - Eliminating the need for any
trusted third party, hence removing any single point of
failure;
• Robust security - Allowing secure transactions and
preventing any attacks based on known attack vectors;
• High scalability - Enabling the network to achieve a
performance at least equal to the centralized counterpart,
as measured in TPS;
• Efficiency - Performing all network services with mini-
mal energy and computational requirements;
• Bootstrapping and storage enhancement - Ensuring a
competitive cost for data storage and synchronization;
• Cross-chain interoperability - Enforced by design, per-
mitting unlimited communication with external services.
Starting from the above challenges, we’ve created Multi-
versX as a complete rethinking of public blockchain infras-
tructure, specifically designed to be secure, efficient, scalable
and interoperable. MultiversX’ main contribution rests on two
cornerstone building blocks:
1) A genuine State Sharding approach: effectively parti-
tioning the blockchain and account state into multiple
shards, handled in parallel by different participating
validators;
2) Secure Proof of Stake consensus mechanism: an
improved variation of Proof of Stake (PoS) that ensures
long term security and distributed fairness, while elimi-
nating the need for energy intensive PoW algorithms.
3 Adaptive State Sharding
MultiversX proposes a dynamically adaptive sharding mech-
anism that enables shard computation and reorganizing based
on necessity and the number of active network nodes. The
reassignment of nodes in the shards at the beginning of
each epoch is progressive and nondeterministic, inducing no
temporary liveness penalties. Adaptive state sharding comes
with additional challenges compared to the static model. One
of the key-points resides in how shard-splitting and shard-
merging is done to prevent overall latency penalties introduced
by the synchronization/communication needs when the shard
number changes. Latency, in this case, is the communication
overhead required by nodes, in order to retrieve the new state,
once their shard address space assignment has been modified.
2
MultiversX proposes a solution for this problem below,
but first some notions have to be defined: users and nodes.
Users are external actors and can be identified by an unique
account address; nodes are computers/devices in the Multi-
versX network that run our protocol. Notions like users, nodes,
addresses will be further described in chapter II.1 - Entities.
MultiversX solves this challenge by:
1) Dividing the account address space in shards, using a
binary tree which can be built with the sole requirement
of knowing the exact number of shards in a certain
epoch. Using this method, the accumulated latency is
reduced and the network liveness is improved in two
ways. First, thanks to the designed model, the dividing of
the account address space is predetermined by hierarchy.
Hence, there is no split overhead, meaning that one
shard breaks into two shards, each of them keeping
only one half of the previous address space in addition
to the associated state. Second, the latency is reduced
through the state redundancy mechanism, as the merge
is prepared by retaining the state in the sibling nodes.
2) Introducing a technique of balancing the nodes in each
shard, to achieve overall architecture equilibrium. This
technique ensures a balanced workload and reward for
each node in the network.
3) Designing a built-in mechanism for automatic transac-
tion routing in the corresponding shards, considerably
reduces latency as a result. The routing algorithm is de-
scribed in chapter IV .4 - MultiversX sharding approach.
4) In order to achieve considerable improvements with
respect to bootstrapping and storage, MultiversX makes
use of a shard pruning mechanism. This ensures sus-
tainability of our architecture even with a throughput of
tens of thousands of transactions per second (TPS).
4 Secure Proof of Stake (SPoS)
We introduce a Secure Proof of Stake consensus mecha-
nism, that expands on Algorand’s [3] idea of a random se-
lection mechanism, differentiating itself through the following
aspects:
1) MultiversX introduces an improvement which reduces
the latency allowing each node in the shard to determine
the members of the consensus group (block proposer and
validators) at the beginning of a round. This is possible
because the randomization factor r is stored in every
block and is created by the block proposer using a BLS
signature [4] on the previous r.
2) The block proposer is the validator in the consensus
group who’s hash of the public key and randomization
factor is the smallest. In contrast to Algorand’s [3] ap-
proach, where the random committee selection can take
up to 12 seconds, in MultiversX the time necessary for
random selection of the consensus group is considerably
reduced (estimated under 100 ms) excluding network
latency. Indeed, there is no communication requirement
for this random selection process, which enables Mul-
tiversX to have a newly and randomly selected group
that succeeds in committing a new block to the ledger
in each round. The tradeoff for this enhancement relies
on the premise that an adversary cannot adapt faster than
the round’s time frame and can choose not to propose
the block. A further improvement on the security of the
randomness source, would be the use of verifiable delay
functions (VDFs) in order to prevent any tampering
possibilities of the randomness source until it is too
late. Currently, the research in VDFs is still ongoing
- there only a few working (and poorly tested) VDF
implementations.
3) In addition to the stake factor generally used in PoS
architectures as a sole decision input, MultiversX refines
its consensus mechanism by adding an additional weight
factor called rating. The node’s probability to be selected
in the consensus group takes into consideration both
stake and rating. The rating of a block proposer is recal-
culated at the end of each epoch, except in cases where
slashing should occur, when the actual rating decrease
is done instantly, adding another layer of security by
promoting meritocracy.
4) A modified BLS multisignature scheme [5] with 2
communication rounds is used by the consensus group
for block signing
5) MultiversX considers formal verification for the critical
protocol implementations (e.g. SPoS consensus mecha-
nism) in order to validate the correctness of our algo-
rithms.
II Architecture Overview
1 Entities
There are two main entities in MultiversX: users and nodes.
Users, each holding a (finite) number of public / private
(Pk/sk) key pairs (e.g. in one or multiple wallet apps), use the
MultiversX network to deploy signed transactions for value
transfers or smart contracts’ execution. They can be identified
by one of their account addresses (derived from the public
key). The nodes are represented by the devices that form the
MultiversX network and can be passive or actively engaged
in processing tasks. Eligible validators are active participants
in MultiversX’ network. Specifically, they are responsible for
running consensus, adding blocks, maintaining the state and
being rewarded for their contribution. Each eligible validator
can be uniquely identified by a public key constructed through
Fig. 1: Relations between MultiversX entities
3
a derivation of the address that staked the necessary amount
and the node id. Relations between entities in the MultiversX
protocol are shown in Fig. 1.
Furthermore, the network is divided into smaller units called
shards. An eligible validator is assigned to a shard based on
an algorithm that keeps the nodes evenly distributed across
shards, depending on the tree level. Each shard contains a
randomly selected consensus group. Any block proposer is
responsible to aggregate transactions into a new block. The
validators are responsible to either reject, or approve the
proposed block, thereby validating it and committing it to the
blockchain.
2 Intrinsic token
MultiversX grants access to the usage of its network through
intrinsic utility token called eGold, in short EGLD. All
costs for processing transactions, running smart contracts and
rewards for various contributions to the network will be paid in
EGLD. References to fees, payments or balances are assumed
to be in EGLD.
3 Threat model
MultiversX assumes a byzantine adversarial model, where
at least 23n+1 of the eligible nodes in a shard are honest. The
protocol permits the existence of adversaries that have stake or
good rating, delay or send conflicting messages, compromise
other nodes, have bugs or collude among themselves, but as
long as 23n+1 of the eligible validators in a shard are honest/not
compromised, the protocol can achieve consensus.
The protocol assumes highly adaptive adversaries, which
however cannot adapt faster than a round’s timeframe. The
computational power of an adversary is bounded, therefore
the cryptographic assumptions granted by the security level of
the chosen primitives hold firmly within the complexity class
of problems solvable by a Turing machine in polynomial time.
The network of honest nodes is assumed to form a well
connected graph and the propagation of their messages is done
in a bounded time ∆.
Attack vectors’ prevention
1) Sybil attacks: mitigated through the stake locking when
joining the network. This way the generation of new
identities has a cost equal to the minimum stake;
2) Nothing at stake: removed through the need of multiple
signatures, not just from proposer, and the stake slashing.
The reward per block compared to the stake locked will
discourage such behavior;
3) Long range attacks: mitigated by our pruning mech-
anism, the use of a randomly selected consensus group
every round (and not just a single proposer) and stake
locking. On top of all these, our pBFT consensus algo-
rithm ensures finality;
4) DDoS attacks: the consensus group is randomly sam-
pled every round (few seconds); the small time frame
making DDoS almost impossible.
Other attack vectors we have taken into consideration are:
shard takeover attack, transaction censorship, double spend,
bribery attacks, etc.
4 Chronology
In MultiversX’ network, the timeline is split into epochs and
rounds. The epochs have a fixed duration, set to one day (can
be modified as the architecture evolves), at the end of which
the shards reorganization and pruning is triggered. The epochs
are further divided into rounds, lasting for a fixed timeframe.
A new consensus group is randomly selected per shard in each
round, that can commit a maximum of one block in the shard’s
ledger.
New validators can join the network by locking their stake,
as presented in chapter V .2 - Secure Proof of Stake. They are
added to the unassigned node pool in the current epoch e, are
assigned to the waiting list of a shard at the beginning of epoch
e + 1, but can only become eligible validators to participate
in consensus and get rewarded in the next epoch e + 2.
The timeline aspects are further detailed in section IX.1.
III Related Work
MultiversX was designed upon and inspired by the ideas
from Ethereum [6], Omniledger [7], Zilliqa [8], Algorand [3]
and ChainSpace [9]. Our architecture goes beyond state of
the art and can be seen as an augmentation of the existing
models, improving the performance while focusing to achieve
a better nash equilibrium state between security, scalability
and decentralization.
1 Ethereum
Much of Ethereum’s [6] success can be attributed to the
introduction of its decentralized applications layer through
EVM [10], Solidity [11] and Web3j [12]. While Dapps have
been one of the core features of ethereum, scalability has
proved a pressing limitation. Considerable research has been
put into solving this problem, however results have been
negligible up to this point. Still, few promising improvements
are being proposed: Casper [13] prepares an update that will
replace the current Proof of Work (PoW) consensus with a
Proof of Stake (PoS), while Plasma based side-chains and
sharding are expected to become available in the near future,
alleviating Ethereum’s scalability problem at least partially
[14].
Compared to Ethereum, MultiversX eliminates both energy
and computational waste from PoW algorithms by imple-
menting a SPoS consensus while using transaction processing
parallelism through sharding.
2 Omniledger
Omniledger [7] proposes a novel scale-out distributed ledger
that preserves long term security under permission-less op-
eration. It ensures security and correctness by using a bias-
resistant public-randomness protocol for choosing large, statis-
tically representative shards that process transactions. To com-
mit transactions atomically across shards, Omniledger intro-
duces Atomix, an efficient cross-shard commit protocol. The
concept is a two-phase client-driven ”lock/unlock” protocol
that ensures that nodes can either fully commit a transaction
across shards, or obtain ”rejection proofs” to abort and unlock
4
the state affected by partially completed transactions. Om-
niledger also optimizes performance via parallel intra-shard
transaction processing, ledger pruning via collectively-signed
state blocks, and low-latency ”trust-but-verify” validation for
low-value transactions. The consensus used in Omniledger is a
BFT variation, named ByzCoinX, that increases performance
and robustness against DoS attacks.
Compared to Omniledger, MultiversX has an adaptive ap-
proach on state sharding, a faster random selection of the
consensus group and an improved security by replacing the
validators’ set after every round (a few seconds) not after every
epoch (1 day).
3 Zilliqa
Zilliqa [8] is the first transaction-sharding architecture that
allows the mining network to process transactions in parallel
and reach a high throughput by dividing the mining network
into shards. Specifically, its design allows a higher transaction
rate as more nodes are joining the network. The key is
to ensure that shards process different transactions, with no
overlaps and therefore no double-spending. Zilliqa uses pBFT
[15] for consensus and PoW to establish identities and prevent
Sybil attacks.
Compared to Zilliqa, MultiversX pushes the limits of shard-
ing by using not only transaction sharding but also state shard-
ing. MultiversX completely eliminates the PoW mechanism
and uses SPoS for consensus. Both architectures are building
their own smart contract engine, but MultiversX aims not
only for EVM compliance, so that SC written for Ethereum
will run seamlessly on our VM, but also aims to achieve
interoperability between blockchains.
4 Algorand
Algorand [3] proposes a public ledger that keeps the con-
venience and efficiency of centralized systems, without the
inefficiencies and weaknesses of current decentralized imple-
mentations. The leader and the set of verifiers are randomly
chosen, based on their signature applied to the last block’s
quantity value. The selections are immune to manipulations
and unpredictable until the last moment. The consensus relies
on a novel message-passing Byzantine Agreement that enables
the community and the protocol to evolve without hard forks.
Compared to Algorand, MultiversX doesn’t have a single
blockchain, instead it increases transaction’s throughput using
sharding. MultiversX also improves on Algorand’s idea of ran-
dom selection by reducing the selection time of the consensus
group from over 12 seconds to less than a second, but assumes
that the adversaries cannot adapt within a round.
5 Chainspace
Chainspace [9] is a distributed ledger platform for high
integrity and transparent processing of transactions. It uses
language agnostic and privacy-friendly smart contracts for
extensibility. The sharded architecture allows a linearly scal-
able transaction processing throughput using S-BAC, a novel
distributed atomic commit protocol that guarantees consistency
and offers high auditability. Privacy features are implemented
through modern zero knowledge techniques, while the consen-
sus is ensured by BFT.
Compared to Chainspace, where the TPS decreases with
each node added in a shard, MultiversX’ approach is not
influenced by the number of nodes in a shard, because the con-
sensus group has a fixed size. A strong point for Chainspace
is the approach for language agnostic smart contracts, while
MultiversX focuses on building an abstraction layer for EVM
compliance. Both projects use different approaches for state
sharding to enhance performance. However, MultiversX goes
a step further by anticipating the blockchain size problem in
high throughput architectures and uses an efficient pruning
mechanism. Moreover, MultiversX exhibits a higher resistance
to sudden changes in node population and malicious shard
takeover by introducing shard redundancy, a new feature for
sharded blockchains.
IV Scalability via Adaptive State Sharding
1 Why sharding
Sharding was first used in databases and is a method for dis-
tributing data across multiple machines. This scaling technique
can be used in blockchains to partition states and transaction
processing, so that each node would process only a fraction of
all transactions in parallel with other nodes. As long as there
is a sufficient number of nodes verifying each transaction so
that the system maintains high reliability and securi프로젝트로 돌아가기