The so-called “impossible triangle” refers to the fact that in blockchain public chains, it is difficult to achieve both good “decentralization”, good system “security”, and high “transaction processing performance” at the same time. transaction processing performance. Some blockchain technology systems pay more attention to the fairness of decentralization, and the efficiency of transactions is relatively low; some blockchain technology systems pay more attention to efficiency, and the fairness of decentralization has to be sacrificed; while emphasizing fairness and efficiency, security cannot be ignored. The balance and equilibrium between these three has produced different directions of blockchain technology innovation.
The booming development of Ethernet ecology relies on his establishment of a huge developer-friendly community, which has attracted a large influx of developers for development and construction. Likewise, DFINITY is positioned as an Internet computer and wants more developers to come in to develop in a simpler way.
With its own underlying framework and blockchain-minded distributed applications, Ethernet and DFINITY allow developers to implement rich applications with a small amount of code, creating convenient and efficient applications by first attracting developers, and then using this to attract user participation.
Recently, Dominic, the founder of DFINITY, announced on social media that work on DFINITY’s network integration with ethereum will be next. If this goal is achieved, this will attract more developers to freely choose a lower cost, more efficient and more convenient environment for development, which is undoubtedly something that both DFINITY and the ethereum ecosystem would prefer to see. It can be seen that DFINITY is very similar to Ether, and DFINITY has touted itself as the “crazy sister” of Ether to illustrate this point, only more concerned about performance and and neural network governance model. So let’s take a look at some of the technical features of these “crazy sisters” from the “Blockchain Impossible Triangle”.
Decentralization and security issues (governance)
DFINITY implements a governance mechanism based on mobile democracy, compared to Ether’s “code is law”, they can both prevent hackers or nodes from doing evil, but Ether relies on core developers and technical means to restore some damaged parties to normal, and cannot use automated tools to stop or undo the evil, while DFINITY introduces a more automated on-chain governance mechanism called Network Neural System (NNS) to restore all parties to normal in the face of various evil doers.
DFINITY’s Neural Network System (NNS) can fully control all aspects of the network, for example, it can upgrade the protocols and software used by the node machines hosting the network; it can create new blockchain subnets to achieve capacity expansion; it can split subnets to equalize the network load; it can configure economic parameters, such as the ratio of ICP to cycles exchange; in extreme cases, it can even It can freeze/unfreeze/modify the software (smart contracts) in the network to protect the network, etc. Whether or not these actions are performed, or when and where they are performed, depends on the voting activity of the “neurons” created by the network participants. Anyone can lock up a certain amount of tokens to create “neurons” to submit proposals or vote on proposals submitted by other neurons, and allow “neurons” to choose different followers to follow for different kinds of proposals. This allows the collective wisdom and will of the community to govern the network. If the voting process can achieve true community consensus and the community does not break down when NNS exercises its power, DFINTY’s automated governance may represent an attractive alternative for certain types of applications. While NNS has powerful tools to address a wide range of governance issues, there is still no guarantee that these tools will be used to address application-specific issues; the proposal must pass. However, if there is a repair mechanism that everyone agrees to when the community builds consensus, and that mechanism is not subject to abuse, then this will be of great benefit.
To participate in the Ethernet network, one needs to run a specific type of software called node software or client. This node software enforces the rules of the Ethernet protocol. Each client is maintained by a different team, but they all follow the same Ethernet protocol specifications. The differences between them are mainly in some technical details that are important to developers, such as the programming language they use, the amount of developer support they have, or the open source license they use. The ethereum protocol specifications are high-level rules that coordinate things like how they propagate from end users to the mempool (where miners make transactions) and how miners propagate their blocks to the rest of the network after solving the cryptographic puzzle (aka proof of work). How they are encoded to the client is up to the team. The process of implementing software changes is very similar to the process of passing new laws in the real world. In the real world there are various stakeholders. Any one of these participants can submit a proposal on Github called an Ethernet Improvement Proposal (EIP). If a new EIP is included in a release, miners and other nodes will automatically adopt the changes the next time they upgrade their software. This is called a soft fork (soft fork). Miners and nodes can run different (old and new) versions of the software without causing any compatibility issues. The changes introduced are backward compatible. On the other hand, if the software changes made are incompatible with the current protocol, governance must schedule a hard fork or things will get messy. Ethercore developers have started referring to a hard fork as a “network upgrade”. During a hard fork, nodes that do not update their software are separated from the rest of the network because they no longer use the same “language” as the rest of the network. In other words, one group of miners follows a new set of rules, while another group of miners follows the old ones. This is why it is always important to schedule hard forks in advance to provide enough time for each node and miner to prepare. This is important because if a crypto exchange holding a large amount of user funds points its nodes to the wrong chain, it can cause a lot of confusion and disruption as users will be temporarily unable to get their funds back. Most hard forks are merely formal and are followed by all participants, but in rare cases they can be controversial and lead to ideological divisions.
DFINITY enables fast block generation through its threshold relay technology, which uses a threshold signature to enable a group of verifiers in a peer-to-peer relay network to quickly agree. A threshold signature is a group signature that can only be constituted by a combination of some threshold signatures of the members and therefore represents a cryptographic proof of the agreement of these members. In this case, it is more expensive for the culprit to attack because the members of the threshold signature group are different for each block and are randomly selected, so the attacker must have enough interest to control enough signers to influence the consensus. The larger the group, the more difficult and costly this control becomes. Its security depends on the use of unpredictable and tamper-proof random number selection to verify group members, which is achieved by threshold group BLS signatures. Group signatures from previous block times are used to randomly select the verification group for the current block, then generate group signatures to be used in the next block time, and so on.
DFINITY uses a secure distributed secret key generation (DKG) protocol that enables group members to interactively generate group signature secret keys. The security of this protocol has not been proven in networks containing adversaries, and while DFINITY claims that the threshold relay consensus process can tolerate the high failure rate of the DKG protocol and continue to work, there are some challenges in optimizing and designing secret key generation.
For both individuals and organizations, launching a successful attack on the Bitcoin or Ether PoW blockchain or taking control of the entire chain would require a huge capital cost. When Ether transitioned to the PoS (Proof of Stake) mechanism during the Serenity phase, users could earn additional ETH rewards by verifying blocks after becoming verifiers by pledging 32 ETH. In the PoS mechanism, the cost of attacking the Ethernet network will be tied to the cost of the ETH spent. Unlike the energy-consuming mining method used in the PoW mechanism, validators in the PoS mechanism will “pledge” ETH and will lose some or all of the pledged ETH for attempting to commit fraud (validating invalid blocks). the greater the number of validators in the Ethernet 2.0 network that have pledged ETH, the more secure the network will be. The more secure the network is, the more ETH an attacker will need to buy to launch an attack. Moreover, such an attack will likely result in a rapid increase in the price of ETH, thus making it more costly for the attacker.
In short, DFINITY and Ether are both using their signature algorithms and consensus mechanisms to increase the cost of doing evil in the network, so that the attacker can gain much less than the cost of doing evil.
Speaking of security, DFINITY says that DFINITY allows nodes in the network to quickly agree to generate new blocks quickly and regularly through threshold relay technology and the BLS cryptographic algorithm, and DFINITY claims that the use of Probabilistic Slot Protocol may increase the number of transactions on each block by a factor of 25 (compared to Ether compared to Ether). This protocol divides each block time into priority slots that identify the blockchain generator. The higher the slot priority, the higher the block weight. Thus in the case of multiple blocks generated within a given block time, the network selects the valid chain with the largest cumulative weight. The random value generated within the previous block time allows all honest nodes to autonomously agree on the slot priority and block weight, and therefore becomes the valid chain with the largest weight. Chains consisting of blocks are only valid after passing the group threshold signature notarization, and the notarization of each block time quickly eliminates the chain with the lighter weight, making the transaction confirmation time only 2 blocks.
As Ethernet 2.0 enters the beacon chain stage, its fragmented network, though not carrying transaction data, has already taken a key step in speeding up and scaling up, and its consensus mechanism POW will also be transformed into POS after the merger execution of 2.0. At present, POW is still used as the underlying consensus of Ethernet’s main chain, and its lack of efficiency and scalability is one of the factors for the launch of Ethernet 2.0. Under the POW mechanism, its Under the POW mechanism, the node absentee can only generate blocks step by step, in the process, the absentee needs to store and synchronize more and more data, and with the hot DEFI project, the surge of transaction volume makes the network cost and congestion greatly increase. Eth 2.0 will eventually be divided into 1,024 slices, which it hopes should theoretically increase network throughput by > 1,000 times. But there is an upper limit to scaling through such slices.
DFINITY’s vision is to scale the network “almost infinitely” through an architecture that divides consensus, authentication and storage into different layers. The consensus layer only determines the outgoing blocks, not the transaction data. Unlike Ether where each absentee needs to keep all the data, DFINITY’s storage layer is divided into multiple chains, each of which is responsible for handling transactions that update the state of a slice. The validation layer is responsible for combining the hashes of all the slices into a Merkle tree-like structure, storing the hashes of the root state on the chains. Although this architecture provides a solid theoretical basis for scaling, there are still some significant issues to be solved in practice, such as transactions affecting the state of multiple slices.
There are countless projects that have been named “competitor of Ether”, which proves that Ether’s technology development and ecological construction are the best in the blockchain industry. It is a relationship of interaction and mutual improvement. For developers and users alike, providing more options is pleasing to the entire community.
Posted by:CoinYuppie，Reprinted with attribution to:https://coinyuppie.com/dfinity-and-ether-from-the-impossible-triangle/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.