A blockchain testing company claims to have found something quite shocking: EOS EOS, a blockchain protocol that was worth $ 4 billion a few months ago, may not actually be a blockchain.
In a new experiment, the Whiteblock benchmarking firm came to the conclusion that the EOS signal (and its RAM market) is essentially a cloud computing service – and it is built on a totally aggregated reasoning. Therefore, it lacks some of the most fundamental aspects of the blockchain, such as the unchanged.
"With practical testing and experiments in a controlled laboratory environment, this research provides a thorough and objective model [EOS’] design, performance, and finances to present a report on the blockchain community, "says the paper, which Whiteblock shared directly with Hard Fork.
Whiteblock created a copy of EOS to destroy
Unlike trademark blocks that allow anyone to contribute to the network, EOS chooses who will process transactions (block makers) through a complicated voting process called the Delegated Proof-of-Stake.
During these elections, each EOS lesson equals one vote, which means that those holding a lot of EOS have more of a majority in who controls the network.
So, to run this experiment, Whiteblock ran a copy of EOS, which claims to work in the same way as the real thing.
"It performs the same software The Whiteblock-enabled block producers perform the same functions that a block producer would produce in the main net," said Jacques Cole, head of Whiteblock's technology at Hard Fork. "We provide nodes in a controlled test environment, set up network conditions between these nodes to mimic real performance and automate their process and actions to observe their behavior and measure their performance in a deterministic way."
Whiteblock initially began testing the copy of the EOS Chain Block in September. The tests were carried out in an isolated environment and continued for two months.
The company describes EOS as more than a network providing promises for "blackbox" computing resources for user access. Even more frustrating, he says that the entire EOS system is based on an incomplete, centralized model.
"EOS is not a blockchain, rather a distributed homogeneous database management system, a clear distinction that their transactions are not cryptographically validated," says Whiteblock. "EOS block producers are highly concentrated and users can access the network only through block producers as intermediaries. Block producers are a single failure point for the whole system."
EOS is not so fast and can be controlled by a cartel
A large piece of paper is dedicated to demonstrating that there is no proper protocol to prevent collusion among block producers to maintain their role as block producers with minimal protection against bad cartel makers to crush the entire network .
Therefore, the report shows that EOS suffers from consensual failures without Byzantine Inability to Fail (BFT), leaving the network open for control by rogue, consensual members.
For a blockchain to have a BFT, the network must be able to withstand system failures resulting from situations associated with a mathematical puzzle called Byzantine General Problem. " If it can not, bad actors are theoretically able to process false transactions, so BFT is related to the reliability of a blockchain.
"Conceptually, it is impossible for EOS to apply the Byzantine fault tolerance. A real BFT system would not be prone to cartels that are formed in the system, [but] […] partnerships are easily formed on EOS, eliminating any attempt to claim the BFT. "
In particular, the researchers note that the primary threat to EOS's integrity is Sybil attack, which includes bad actors that force other network contributors not to be able to process transactions by creating false identities and use them to launch spam and DDoS attacks.
"This is in fact a great vulnerability in the system, as false users are basically able to create malicious bills much faster than block producers are able to reach a consensus [on which accounts to exclude], "Warns Whiteblock." This further demonstrates the high level of concentration in the EOS network and the enormous power these block manufacturers have. "
The report then states that block producers do not process transactions based on a consensus algorithm, but they confirm transactions "mechanically" without officially checking the validity of transactions.
For this reason, Whiteblock's reference data revealed that the volume of transactions that can be processed by EOS is significantly lower than originally claimed in marketing materials, never exceeding 250 transactions per second (TPS) , even with optimal settings such as zero delay and packet loss.
It should be noted that other testers have previously been compared to the speed of the EOS network. The general belief is that the current maximum EOS performance is about 4,000 TPS.
EOS white paper indicates that it is entirely possible that EOS will scale one day to process millions of transactions per second.
"During real-world testing 50 [milliseconds] the delay of cash routing and packet loss of 0.01 percent, performance dropped below 50 TPS, placing the system in close proximity to the performance in Ethereum, "Whiteblock says.
Bitcoin is capable of processing up to 7 TPS and Ethereum can process around 20.
Whiteblock says EOS does not use encryption
According to Cole, EOS stores all transaction-related data in a table type designed by EOS leader, Dan Larimer, called Chainbase.
When the EOS network confirms transactions, Whiteblock argues that block producers simply refer to new transaction data against this table instead of confirming their legitimacy with cryptography.
The company says that EOS transactions only happen because the block producers update the data stored in the subject Chainbase, rather than cryptographically verified changes to the status of the underlying blockchain, as is the case with Ethereum.
"All of these actions work in an environment where there is no cryptographic validation of contracts and transactions," the survey said. "EOS is basically the same as a central cloud computing architecture [client/server] without the fundamental components of a blockchain or peer-to-peer network. "
Having participants in the network validating transactions by checking a specific table has consequences. Not only is it unusual for encryption, but it provides programmers with virtually countless undos, meaning EOS transactions can be reversed by those who have access (such as block producers).
Indeed, there have been cases of EOS reverse transactions and frozen accounts.
"The ability to uncover the state-related story (or anybody else about it) is a concept that comes in direct contradiction to the real definition of what can be considered as a blockchain, characterized by unchanging data," he concluded Cole.
Apart from that, differently
Hard Fork has reached many EOS block producers for comments. At the time of printing, most representatives refused to comment, noting that they await the publication of the full report.
A source, an EOS dApp developer, said Whiteblock's interpretation of how EOS validated the transactions was particularly "odd."
The source explains this Chainbase is just a method of storing real-time information, such as folders or files. In this case, EOS stores transaction data in one Chainbase, which is placed in a central position (similar to a Bitcoin main hub).
Blocked producers then validate transactions using cryptography, writing the confirmed transactions in a blockchain as part of the process.
Chains are designed to optimize EOS performance by enabling EOS to store data in RAM, which is then used to quickly generate blocks.
This suggests that EOS actually uses cryptography, as opposed to Whiteblock's conclusions.
A look behind the curtain
It should be noted that ConsenSys, commissioned by Whiteblock, is invest heavily in the Ethereum blockhouse ecosystem, the main competitor of EOS. She describes herself as a "production production studio" that focuses on the development of Ethereum-powered platforms.
And if all this is not quite strange – EOS was actually burden in the Ethereum chain, starting as a token ERC-20.
But after the Block.one launches the EOS network in June to become its own mine blockchain, many Ethereum brands have made the change – Everipedia is one, managed by Larry Singer, co-founder of Wikipedia.
We also have to say that EOS is no stranger to confrontation. The launch of the mainnet was nightmarish, the whole process lasting more than a week, with block producers not agreeing if EOS was ready to live on its own.
Since EOS is fully operational, Block.one has paid over $ 400,000 to independent security researchers to detect critical code imperfections that were not set up in time for the launch.
In any case, Consensys states that it will use Whiteblock's findings to develop comprehensive reports that it will provide to its partners, including Ledger Capital, Bo Shen, Enterprise Ethereum Alliance, Microsoft, and Google. Academic institutions such as MITUSC and the Duke are committed to contributing resources to complete the survey.
Furthermore– strangely, one of the listed partners, Bo Shen, is in fact co-founder of Bitshares, an autonomous blockchain organization developed by Dan Larimer and released in 2014.
The research has consistently highlighted the architectural similarities between Dan Larimer's Bitshares system and EOS – particularly the ability to restore the situation and make changes to the genetics block.
Whiteblock has allowed Hard Fork to publish the study, which includes a detailed representation of its methodology along with the detailed test results. Full exposure can be accessed from this link,
For those who remain convinced, the company also says it will pass the EOS benchmarking test in November.
Very funny, you can even RSVP here.
Publication 1 November 2018 – 13:00 UTC