Talent Square Competitors –


n
nThis article is the second article in the “Tower of the Pacific Competitors” series, which explores the challenges of the etherfront project in the field of intelligent contract platforms. Today we take a look at DFINITY.n
nDFINITY is an intelligent platform that is compatible with the Etherbus, implementing some of the ideas for handling block chain performance, expansion and management. Although DFINITY belongs to the category of “etherfoot killer”, the project is more inclined to cooperate, not confront, with the attitude of the tower. DFINITY provides funding and assistance to the APF project and provides technology for its use. DFINITY advertised himself as the “crazy sisters” of the ether square, to show that they are very similar to the ether square, but more concerned about the performance and nerve center governance model.n
nThese specialities make DFINITY start to expand and strengthen the EVM ecosystem, providing applications with different platforms. However, if DFINITY can successfully create an EVM-compliant smart contract platform – with higher transaction throughput, shorter confirmation time and a governance mechanism that can resolve public disputes without leading to community fragmentation, it will become a new deployment The choice of the application, and with the expansion of the network effect, will attract more existing applications. Of course, the challenge of DFINITY will be to meet these promises while meeting the security needs of the public chain (many of which are at risk).n
nThe DFINITY project is currently in the stage between prototyping and production; there is no announcement of a network or block chain to deploy and test existing smart contracts. Although the official white paper has not been released, but DFINITY has published a lot of technical information, so that we can understand the project and the bridge between the similarities and differences between. Then we have to take a closer look at it may challenge the status of some of the technical characteristics of the square bar bar.n
nn
nPerformancen
nThe low transaction throughput is a well-known problem for the etherchards and other work volume proof (PoW) block chains. Although the APF plans to replace PoW with a purely proprietor Proposal (PoS) solution, the first phase will only add PoS levels to the PoW-enabled block chain, and will not improve transaction throughput. While DFINITY has the advantage of having no legacy problems, a pure PoS system will be introduced, with faster block generation times (ie, generating more blocks per minute) with higher gas caps (ie, each block contains more Transaction) increases the number of transactions processed per second by 50 times.n
nDFINITY realizes the rapid generation of blocks through its threshold relay technology, and uses threshold signatures to quickly reach a group of verifiers in peer-to-peer relay networks. A threshold signature is a group signature that can only be composed of a combination of signatures of some members of the threshold, thus representing the encryption of the agreement of these members. DFINITY uses the BLS group signature to have a very important attribute, that is, the signature bit is always the same, regardless of the contributor who is. This allows the network to reach protocols quickly and independently with random values ​​(ie signatures), which is the key to quickly and regularly generate blocks (for example, generate one every 5 seconds) and significantly increase transaction throughput.n
nDFINITY said that the Probabilistic Slot Protocol may increase the number of transactions on each block by a factor of 25 (compared to the Ether Square). The protocol divides each block time to the priority slot of the identification block chain producer. The higher the slot priority, the greater the block weight. Therefore, in the case where multiple blocks are generated within a given block time, the network selects the active chain with the largest cumulative weight. The random values ​​generated in the previous block time allow all honest nodes to agree autonomously with the slot priority and block weights, thus becoming the most efficient chain of weight. The chain consisting of blocks is valid only after notarized by group threshold signatures, and the notarization of each block time quickly eliminates the lighter weight of the chain so that the transaction confirmation time requires only two blocks (about 7.5 seconds) Then
nIn the world of smart contracts, high performance but low security means that an attacker can drain your money more quickly. The attack threshold relay protocol is expensive because the threshold signature group for each block is different and is randomly selected, so the attacker must have sufficient bribes to control enough signers to influence the consensus. The greater the group, the more difficult it is to hinder the consensus of the honest verifier, and the higher the price. The security depends on the use of unpredictable and tamper-resistant random sources to select the authentication group members, which are implemented through the threshold group BLS signature. The group signature from the previous block time is used to randomly select the validation group of the current block and then generate the group signature used in the next block time, and so on.n
nTherefore, the security of the group signature key is the key to preventing the randomness of the process of notarization and protection of the threshold. DFINITY uses a secure distributed key generation (DKG) protocol that enables group members to interactively generate group signature keys without any centralization rights. Although this protocol has proven to be secure in a reliable network, this assumption does not apply to a relay network that contains an opponent, and the key can be destroyed by an attacker who has sufficient resources to control a single group. The DFINITY claims that the threshold relay consensus process can tolerate the high failure rate of the DKG protocol and continues to function, and may also disclose the maximum tolerable failure rate and the number of non-faulty production damage group key details. The computational and communication costs of the DKG algorithm are important for effectively preventing the population size required for attacks on threshold resolution (about 400 members), so DFINITY may face some major challenges in optimizing and designing key generation protocols, These protocols play a better role in a relay network with a confrontation node.n
nTheoretically, the ethertop can use these features that make DFINITY superior, and it is really possible to do so, provided that DFINITY proves that these features are effective in protecting their billions of dollars while improving performance. However, it now appears that Ethernet will continue to improve its performance through its current approach, using its own PoS protocol (CASPER) with focus on usability and other extensions.n
nn
nExtensibilityn
nDFINITY is committed to expanding the network by dividing the consensus, validation and storage into different levels of architecture “almost infinitely”. The consensus layer has no trading blocks. The storage tier is divided into multiple chains, each of which is responsible for handling transactions that update the fragmented state. The validation layer is responsible for combining all the hashes into a structure similar to the Merkel tree, so that the global state hash is stored in the top chain block. While this architecture provides a solid theoretical basis for the expansion, there are still some major issues that need to be addressed in practice, such as transactions that affect multiple fragmented states.n
nThe Ethernet expansion plan also includes state fragmentation, but it may take several years to deploy such a work system on the main web. The ethertop will have to migrate from the fully replicated global state to the fragmented state, and DFINITY has the opportunity to implement fragmentation from the beginning. DFINITY will be able to successfully implement state fragmentation for the first time, which gives it an important expansion advantage. However, DFINITY’s latest development plan shows that this result will appear in its third (Tungsten) main release, which means it will take place a few years later.n
nn
nGovernancen
nDFINITY describes its governance mechanism as “the law of AI”, which is opposite to the “code of the ether” rule. However, this slogan is not enough to make an effective comparison; DFINITY said “AI” is essentially a mechanism based on mobile democracy, and the response to the DAO hacking incident led to community division, the “code That is, the law “of the believers left.n
nThe real difference is not the idea, but the mechanism of the implementation of the idea, which is the recent Parity wallet black event to our inspiration. Ideally, the ether is an anti-hacker, but there is no formal mechanism or automated tool to stop or revoke the invasion, can only rely on white hat hackers to restore some of the normal return. DFINITY is also an anti-hacker concept, is planning to introduce a more formal and automated chain governance mechanism, you can in a variety of attacks to make the case back to normal.n
nDFINITY’s Block Chain Neural System (BNS) is powerful in terms of executive management, such as management protocols and software upgrades, modifying economic parameters, freezing “rogue” smart contracts, fixing errors in contracts, and even reallocating DFINITY network tokens. Whether these actions should be taken and when these actions should be taken are subjective and often controversial, BNS can only play a role when the community adopts a proposal for action through a reciprocal mobile polling process. For any given proposal, a user who has obtained the right to vote by running a token on the network and who runs the “neuron” on the network can vote directly on any proposal or choose another neuron to represent himself and vote for it. The software will automate everything and allow the user to select different agents for different types of proposals and dynamically update these choices. The purpose of this approach is to make all proposals carefully examined by the collective wisdom of the community and to be adopted only in the best interests of the community (“community interests are the law”).n
nDFINTY’s automated governance may represent an attractive alternative to certain types of applications if the voting process is able to achieve real community consensus and the community does not break when BNS exercises its power. While BNS has a powerful tool to solve a wide range of governance problems, it still can not guarantee that these tools will be used to solve specific problems with the application; the proposal must pass. However, if the community builds a consensus, there is a repair mechanism that everyone agrees with, and the mechanism will not be abused, so it will bring great benefits.n
nIn view of the frequent involvement of the APF, it may adopt some more formal and automated governance mechanisms. Although a lot of value is in danger to make it an urgent need to take action, but it also means that the behavior of the community split will bring real economic pains. Therefore, the Etherhouse may be very cautious, while the less valuable platform will use a very powerful and highly automated mechanism for the experiment.n
nn
nin conclusion n
nDFINITY is unique in the “Tower of the Pacific competitor” field, because it is not a completely different platform, but compatible with EVM’s ether square. This allows DFINITY to take advantage of all the development and innovation from the large ethercom community, and DFINITY’s contribution to these projects is also beneficial to the Ether Square. While it may seem good to have more choices for the entire community and DApp developers for the entire community, people must also consider whether it is a strategy of “hugging, expanding and eliminating”.n
nThe threshold relational consensus and BNS functionality added by DFINITY and the public chain and private chain interoperability it provides should play a good role in the private license chain. But the design of the public chain without permission is very different, it is unclear whether an architecture can control both. These two platforms are likely to eventually coexist as an interoperable sister platform, evolving into a DFINITY-dominated private chain and the ether-based area continues to dominate the public chain.n

Leave a Reply

Your email address will not be published. Required fields are marked *