|
Scalability on TransactionsBog'liq Huaqun Xingjie - A Survey on blockchain Technology and its Security - 2022 MarchScalability on Transactions. In Table I, the maximum
TPS is from 27 of Bitcoin to EOS of 3,996. PoW is capable of
processing anywhere between 10 and 27 TPS worldwide.
Ethereum 2.0 will upgrade and switch to the more efficient
protocol PoS to make Ethereum more scalable, and will
support 1000s TPS [131]. A few delegates in EOS that uses
DPoS consensus algorithm have the right to vote and validate
blocks, and hence EOS is more centralized and is easier for
some delegates to combine together to launch 51% attacks.
The communication cost in PBFT quickly grows up if the
number of nodes increases, and hence it is suitable for private
setup without large number of nodes, but with many
transactions. Currently Hyperledger Fabric based on PBFT
achieves about 3,500 TPS. Hyperledger Sawtooth based on
PoET achieves 2,300 TPS.
In 2019, Perun was proposed as off-chain payment
channel system instead of on-chain transactions to increase
TPS [132], and a sidechain construction was provided for PoS
sidechain systems to enable the scalability [133]. In 2020, Yu
et al. proposed O
HIE
as a permissionless protocol to improve
the transaction throughput to 4-10Mbps [134]. Currently,
Ethereum and Bitcoin process only about 5KB or 10 TPS on
average. So O
HIE
can achieve 8,000 - 20,000 TPS. On the other
hand, Visa's payment network can process over 65,000 TPS
stated in August 2017 [135]. Thus, the scalability of
Blockchain in terms of TPS in real distribution environment
is still an outstanding challenge.
Scalability on Chain Data Sharing. The block sizes for
Bitcoin, Bitcoin Cash and Ethereum are 1MB, between 8 MB
and 32 MB, and under 60KB respectively. IBM Blockchain
supply chain solutions [136] and VeChain [137] record the
shared data on the Blockchain which limit the scalability of
their solution. A large number of stakeholders may be
involved in, and the data that need to be shared among the
stakeholders could be massive and not limited to logistic data.
As more stakeholders join and the shared data grows, the on-
chain data sharing system will be in danger of scalability
issues.
To increase the scalability and also take advantage of
Blockchain technology, the data can be shared on an off-chain
dedicated channel and the link or even proof of the data
sharing can be recorded in the Blockchain for tracing and
auditing. Off-chain data sharing solutions require inter-
company channels which increase the company’s burden for
building and maintaining these channels. In addition, these
solutions cannot guarantee the integrity of the data shared by
a company. For example, Company A can tamper the original
data to make the data meet Company B’s specific requirement
and then share the data with Company B. To decrease the
burden of the involved companies, the data can be positioned
and shared on a cloud platform. We have proposed one
technology in this area of a Blockchain-based access control
and data sharing framework for supply chains, which can be
referred to our patent filed document [138].
B. Securer Software Codes
From Table VIII, we can know that almost every year
attacks on software code and smart contract have happened.
Security is a non-negotiable aspect for any asset related
software. Smart contract security is high requirement because
smart contract deals with the valuable information, e.g.,
cryptocurrencies, token, and other digital assets. The
transactions built with smart contract are irreversible, and
software codes of smart contract are very difficult to be
modified or patched if a bug is discovered [139]. A few
constraints on smart contracts are in place to secure the
Blockchain environment from attackers. Additional to the
accounts and transactions being immutable and secured
through the cryptographically hashed chains, for example,
EOS faces the challenge to secure the smart contract execution
to withstand malicious attacks [140]. In 2020, there is a
research work of Flash Boys 2.0 continuing to show the risks
of smart contract that the arbitrage bots and miner extractable
value of transaction-ordering dependencies in smart contracts
pose a realistic threat to Ethereum [141]. According to [139],
there is very hard to assurance the security of smart contract
code, and hence guarantee security of smart contracts is one
of outstanding challenges for Blockchain.
C. Audit, Zero Trust & Anomaly Detection
|
| |
