深入Monad：重新定义交易速度的 EVM 创新

Digital currency, the platform of Shanouba Virtual Coin Exchange, has been talking about the speed of transaction processing. Recently, we have been discussing how to improve the operation mode explained in detail in this article, which has always been an indicator of our concern. We hope that the blockchain can support higher to accommodate more users and applications. The following picture shows the number of Ethereum. So far, there is no standard reached by a chain. What needs to be remembered is only a general overview to measure scalability. Because not all transactions are equally complicated, it is not very accurate to only use data, but for convenience, we will still regard it as an indicator to measure scalability. So what are the ways to improve it? One way is to build a brand-new system from scratch as usual, and give up compatibility with Ethereum virtual machine for speed. It adopts multi-threaded execution instead of single-threaded execution, just like multi-core and single-core parallel processing transactions and adopting different consensus mechanisms. The second way is to use. The third method to expand Ethereum by executing under the chain and using centralized sorter is to decompose it into separate parts and optimize it to improve the scalability. It is a new and compatible blockchain, which recently raised hundreds of millions of dollars. It chose to build from scratch instead of directly using the existing version, and adopted the third method to improve the scalability. Next, we will discuss some major changes introduced and execute Ethereum virtual machines in parallel to execute transactions. Before one transaction is executed, the next transaction must be completed. Waiting can be an example. Imagine a platform of a motorcycle assembly warehouse. Many trucks transport motorcycle parts. Each truck is equipped with all the parts needed to make a motorcycle. The assembly warehouse has four different functions, and each function is unloaded, classified, assembled and loaded by a special team. In the current setting, there is only one platform and one place for loading and unloading goods, so when the truck stops, motorcycle parts will be unloaded, classified, assembled and loaded on the same truck. His team are all waiting, so if their work is regarded as different slots, each team will only work once every four slots, which leads to serious inefficiency and highlights the need for a more simplified way. Now imagine that there are four platforms with independent loading and unloading areas. Even if the unloading team can only handle one truck at a time, they don't have to wait for the next three slots to start their work. They can directly move to the next truck and start their work. So can the assembly and loading team. In this way, when the truck finishes unloading, it will drive to the loading area and wait for the loading team to load the assembled motorcycle, so the warehouse with only one platform and loading and unloading area will perform all operations in sequence, while the warehouse with two platforms and different loading and unloading areas can process tasks in parallel. It can be regarded as the warehouse infrastructure with multiple truck platforms, but it is much more complicated than this example. When there is a dependency between trucks, the complexity will increase, for example, if a truck does not have a motorcycle factory. What about all the parts needed? Transactions are not always independent, so when they are executed in parallel, it must deal with interdependent transactions. How to do this? It performs an operation called optimistic parallel execution. The protocol can only execute independent transactions in parallel, for example, considering a transaction, in which Joel's balance is Joel's sending a West Germany foundry to Saura, Joel's sending a West Germany Shrock to buy. All these transactions are executed in parallel with pending results. These results will be submitted one by one if. If the output of the pending result conflicts with the original input of any transaction, the transaction will be re-executed and independent of each other, so their pending results will not conflict with the input of other transactions, but the sum is not independent. Please note that since all transactions start from the same initial state, Joel's balance will become the pending settlement after Joel's balance is sent to West Germany, and his balance will be submitted one by one to ensure that the output will not conflict with any input. If Joel's new balance becomes a conflict between the output and the input after submission, Joel's balance becomes so far after the current input is re-executed. The obvious question is how do we know that it is not necessary to re-execute most transactions? The answer is that re-execution is not the bottleneck, but the bottleneck lies in accessing the memory of Ethereum. It turns out that the way Ethereum stores its state in the database makes it difficult, time-consuming and expensive to access the state. This is where another improvement comes into play. When a transaction needs to be re-executed, all the inputs have been cached in the cache memory, and the access speed of the cache memory is much faster than that of accessing the overall state. There are 10,000 in the test network, but only about 1,000 in the main network claim to have realized 10,000 in its internal test network. Although this does not always represent the actual performance, we can't wait to see the performance in practical applications. 比特币今日价格行情网_okx交易所app_永续合约_比特币怎么买卖交易_虚拟币交易所平台

注册有任何问题请添加 微信：MVIP619 拉你进入群

文章转载注明 网址：https://netpsp.com/?id=63690