Chainfeeds Guide:
Ethereum's ultimate dream is to be a world computer, and interestingly, Alpenglow shares the same vision.
Article Source:
https://mp.weixin.qq.com/s/6WFtqnXcumyHYvkL_YrwOQ
Article Author:
Zuo Ye
Perspective:
Zuo Ye: Starting from Bitcoin, the number and dispersion of nodes have been synonymous with the degree of decentralization of a blockchain network. To avoid centralization, the security threshold is set at 33%, meaning no single entity should exceed this proportion. Driven by capital efficiency, Bitcoin mining ultimately led to mining pool clusters, and Ethereum became the main stage for Lido and CEX. Of course, this doesn't mean mining pools and Lido can control network operations; under the "maintain network - earn incentives/management fees" model, they have no motivation to act maliciously. However, measuring network health must consider its scale. For example, in a small group of 3 people, a 2/3 consensus is considered effective, while pursuing only a minimum 1/3 security guarantee is meaningless, as the remaining two can easily collude, with low cost and high potential gains. In a large-scale network of 10,000 nodes, like the current Ethereum nodes shown by Etherscan, there's no need to pursue a 2/3 majority. Outside the incentive model, most nodes are strangers to each other, and the coordination cost for Lido and Binance to act maliciously together would be too high. If we reduce both node count and consensus ratio, can we "speed up and reduce fees"? Everyone might think of this, and Alpenglow is preparing to do just that, maintaining Solana's 1,500 node scale while lowering the security consensus to 20%. This can increase node confirmation speed, earn more mainnet incentives for nodes, and encourage node scale expansion, perhaps to around 10,000. Whether this results in a 1+1>2 effect or breaks existing security mechanisms remains to be seen. However, I like it; Solana should follow the American chain, conspiracy group, and centralization route, competing in the public chain space as Ethereum's counterpart. Alpenglow's theoretical approach is that in a large-scale node era, strong consensus numbers aren't necessary due to the PoS mechanism. Attackers would need to mobilize massive capital to control the network. At current prices, 20% control would require $20 billion for Ethereum and $10 billion for Solana. With $10 billion, why would anyone want to control a blockchain? They'd face resistance from the remaining 80% of nodes, unless it's a state action. In Alpenglow, the protocol variant called Rotor is essentially an orderly block message propagation, with no fixed Leader or Relay nodes. Votor is the node confirmation mechanism; for instance, in Alpenglow's vision, if the first round of node voting reaches 80%, satisfying the minimum 20%, it can pass quickly. If the first round is between 60-80%, a second round of voting can be initiated, and surpassing 60% would confirm it. If that fails, the Repair mechanism is used, but personally, this seems similar to the Optimistic Rollup challenge period. If it reaches this point, the protocol will likely fail, just as FDIC couldn't stop the Silicon Valley Bank run. Unlike using brute-force hardware resources to increase bandwidth, Alpenglow aims to reduce the block consensus generation process. If data blocks are kept small, around 1,500 Bytes, and generation time is very short (currently tested to potentially reach 100ms, just 1% of the current 10s), it could be groundbreaking. (1 s = 1000 ms, though this is questionable; in large-scale practical use, conditions won't be this ideal, but it's still impressive.)
Content Source
