
Soteria theme sharing:
Ming Guo | The past and present of DAG (1)
https://bihu.com/article/1145448479
Q1. What is the purpose of dag? Can't follow the vine, encrypted backup with tail and no head?
A: Different types of DAG have different purposes, but what they have in common is to connect to more nodes, whether the node is a transaction or a block (wallet or miner), because one of the most basic characteristics of a distributed network is that nodes Sufficient connectivity is required to facilitate consensus.
In addition, DAG is a "directed acyclic graph", so it has a "head and a tail". The head is the "creation node", and the tail is dynamically changing.
Blockchain data is "unencrypted", unless it is an anonymous blockchain; but it contains a so-called Merkel Tree with digital signatures.
Q2. What is the biggest difference between "Satoshi Nakamoto Consensus" and "Byzantine Fault Tolerant Consensus"?
A: I think the biggest difference between the "Satoshi Nakamoto Consensus" and the "Byzantine Fault Tolerant Consensus" is that the Satoshi Nakamoto Consensus guarantees "probabilistic security", not the "local security" of the "Byzantine Fault Tolerant Consensus". This makes the network scalability (not performance scalability) of the "Satoshi Nakamoto Consensus" much better than the "Byzantine Fault Tolerant Consensus" network. The Nakamoto consensus can have a scalable network with unlimited nodes, while the nodes of the Byzantine consensus network are usually limited, on the order of dozens or hundreds, and it is said that Google can reach thousands.
Q3. What are the main advantages of DAG and main chain?
A: If you are asking about a blockDAG such as Soteria DAG, compared to Bitcoin, which discards most of the mined blocks and only leaves the longest main chain algorithm, Soteria DAG will write all legal blocks into the last In the blockchain, this is what we call "inclusiveness". Its advantage is that it does not waste the work and energy of miners, and it helps prevent the concentration of computing power in the long run.
Q4. The blocks of ordinary blockchains are also issued sequentially. The blocks are calculated and then connected, right?
A: Absolutely correct. It is precisely because of this that many side chain blocks cannot be linked, and are eventually abandoned to ensure more important security and order. blockDAG is more inclusive of concurrent blocks. Therefore, under the condition of ensuring safety, the overall system performance is improved.
Q5. Does BlockDAG still need POW? Or Satoshi Nakamoto’s set of POW?
blockDAG also needs POW.. We will elaborate on our POW.. the difference from Satoshi Nakamoto in the next few sharings.
Q6: What are the potential areas for improvement in DAG?
A: The different widths (k) of the DAG have a lot to do with the network speed and node processing speed. It is the same algorithm, such as Phantom or Greedy Phantom, there is still a lot of room for adjustment. We have written a lot of developer tools , I hope everyone can test together to find the most suitable solution for the current network and system links.
Q7. How do you view the proportion of the encrypted economy market corresponding to the PoW and PoS mechanisms in the future?
A: I actually don't know the ratio of PoW to PoS. I believe in PoW, because after all, it is the only successful example of battle testing in the past 10 years. PoS still needs the test of time. And after Libra came out, many PoS projects faced greater challenges.
Q8. Liu shu: How does blockdag prevent 51% attacks and double spending?
A: Ming Guo: blockDAG is the consensus of Satoshi Nakamoto. Therefore, the idea of preventing 51% attacks is the same as that of the Bitcoin network. It is necessary to prevent common computing power attacks, such as selfish mining, etc.; in fact, due to the The longest chain winning strategy, the actual situation is that there are many chains being mined, so the longest winning chain usually does not need 51% computing power to win other "sub-longest chains", such as 37% to 40%, then blockDAG BlockDAG is more effective than Bitcoin against this attack because it accommodates all legal "sub-long chains".
Q9. Liu shu: What is a legal sub-long chain? Even a forked block is acceptable?
A: Ming Guo: The legal "sub-longest chain" means that everyone is mining their own "private chain" and then publish it together, so the longest one wins.
Liu shu: That still doesn't solve the problem of 51 attack and double spending! Still the longest chain.
Ming Guo: What I just said is the case of Bitcoin’s computing power attack.
Liu shu: Oh oh oh, how does blockdag solve it?
Ming Guo: blockDAG is "weighing" these "forked" blocks at any time. We call it "coloring" and topological sorting, so this problem can be solved very efficiently. I repeat that the 51% attack itself is not a necessary The problem to be solved is the security "valve" of the Nakamoto consensus. If this valve is lost, the security of the chain will be lost.
Liu shu: I agree that the 51% attack itself is not a bug, but it will bring economic losses to ordinary investors and exchanges. If a transaction in block a is transferred to account M, and the same transaction in the forked block a is transferred to account N, how to solve it?
Ming Guo: The actual situation is that because the longest chain wins, selfish mining, many computing power attacks do not need 51% computing power to succeed, which can be optimized. blockDAG is an optimized algorithm structure for this kind of attack. The flower problem is solved by sorting, and the one ranked first wins.
Liu shu: Are these forked blocks recognized or not? If admitted, how to record block A and block A'?
These questions of Ming Guo all involve the details of the algorithm. In the next sharing (next Tuesday) we will share the algorithm of Soteria DAG in more depth, and these questions will be easy to understand. According to the algorithm of Soteria DAG, all the forked blocks are recognized, so it is called "inclusive" block production. We "color" all the forked blocks and then sort them. The "blue" is the "legal" area Blocks, in the front, dyed "red" are "suspicious" blocks, in the back, under certain circumstances (such as many generations ago) will be discarded.
So come and listen to our more in-depth Soteria DAG technical sharing next Tuesday, August 13th!
Liu shu: Looking forward to it very much.
Related LinksRelated Links
Claire: Thank you Mr. Ming Guo, the keynote speaker, for his wonderful and detailed sharing, which is easy to understand and easy for novices to understand!
(full text)
(full text)