Where is the Ethereum server
Ethereum uses fog computing. The data processing and applications of fog computing are concentrated in the devices at the edge of the network, rather than almost all stored in the cloud, which is an extension concept of cloud computing, which is more in line with the spirit of decentralization of blockchain. If all the storage and operations still stay on the centralized server, then it will not be a real blockchain
recently, the hot "ethereal cat" on Ethereum network has led to the congestion of Ethereum network. The problem is that the TPS concurrency per second of Ethereum is too low. The emergence of ethereumfog with fog computing will break the embarrassing situation of Ethereum network congestion, and add powerful decentralized storage and additional computing power to the chain.
Ethereum uses fog computing. The data processing and applications of fog computing are concentrated in the devices at the edge of the network, rather than almost all stored in the cloud, which is an extension concept of cloud computing, which is more in line with the spirit of decentralization of blockchain. If all the storage and operations still stay on the centralized server, then it will not be a real blockchain
recently, the hot "ethereal cat" on Ethereum network has led to the congestion of Ethereum network. The problem is that the TPS concurrency per second of Ethereum is too low. The emergence of ethereumfog with fog computing will break the embarrassing situation of Ethereum network congestion, and add powerful decentralized storage and additional computing power to the chain
the concept of fog computing may be unfamiliar to many chain friends, but it is the most important aspect of ether fog to attract investors. In fact, I have heard about the concept of fog computing for a long time, and I have always thought that fog computing is the future development direction. When it comes to fog computing, we should start with cloud computing. In fact, the cloud computing we see now evolves from the cluster of independent servers, that is, from the original centralized independent servers to large server clusters, but in the final analysis, it is still a form of centralized computing. Compared with cloud computing, which is the evolution of centralized computing, fog computing is to share computing tasks with various computing power devices around us, and realize the evolution of decentralized and distributed computing, which coincides with the philosophy of blockchain itself.
How to use the Linux system mining, to use the service of the HA Yu miner, only need two steps, you can mine on the Linux system
-
open the website, input the mobile phone number, select how many CPUs you want to use for mining, the default is to use 50% of the CPU for mining, click to generate your exclusive command and
in two steps, you can mine on Linux system, the money you dig only needs to go to the website of hayu miner, enter your mobile phone number to withdraw< br />
there is no central server in Ethereum. Instead, there are many equal nodes connected by P2P protocol, which store all the data in many nodes. When a user initiates a transaction, the transaction will be broadcast out through P2P protocol. The miner node verifies, packages and further broadcasts the transaction to the whole network. After confirmation in the blockchain, the operation is considered to be unchangeable
in the articles on blockchain on the Internet, the two words distributed and decentralized are mentioned, sometimes slightly different, sometimes mixed. The author thinks that if we want to distinguish accurately, the distributed system emphasizes that multiple components work together by sending messages, and the decentralized system emphasizes that there is no central node to control the operation of the whole system. Therefore, we think that Ethereum is both decentralized and distributed, or running a decentralized program on a distributed platform.
in front of the article: the video version is slightly different from the text version. If you want to see my affectionate interpretation, please see the video version (meow understands blockchain phase 22 | sharding: Ethereum is too slow, "disk" him!), The logic of thinking is strange. Please see the extended version
recently, Ethereum has shown an overwhelming positive trend e to the Constantinople upgrade, while Ethereum's upgrade road is like upgrading and fighting monsters, falling into rabbithole. No one knows how deep the hole is. Since it is a long way to go, it is of vital importance to make every step well and accurately. After overcoming this difficulty, Ethereum's next technical difficulty, sharding fragmentation, has been put on the table again. This issue of "meow understands blockchain" will take you into sharding, the magic weapon that makes Ethereum fast
what is sharding
slicing technology is not a new concept. At first, it is an optimization scheme for large central database. Specifically, it divides the data in large database into many shards according to certain rules, and then stores these data in different servers to rece the data access pressure of each server, So as to improve the performance of the whole database system
let's take a small popular example:
for example, meituan, Didi taxi and other software that we often use can be divided according to the "city". Because the data of different cities do not need to communicate, the data of different cities can be stored in different databases, so that the database server can be deployed to the nearest node from the corresponding city, Can also improve the speed of access, why not do it
from the above examples, we should have a preliminary understanding of the concept of fragmentation. What about fragmentation in blockchain scenarios
taking Ethereum fragmentation as an example, in the original single chain system, the overall performance of the public chain depends on the performance of a single node. After fragmentation, each node only needs to undertake the work of the whole network, and each fragmentation works in parallel. According to vitalik, each shard is like a separate galaxy, and each fragmentation is like an independent universe, In this way, the efficiency will be improved naturally! Originally, the total network TPS of Ethereum chain was about 20. Now, if it is increased to 100 pieces, the total network TPS can be increased to 2000. Similarly, the total network capacity will be increased to 100 times of the original
"each node only needs to undertake part of the work of the whole network", which will lead to several major problems. 1. How to determine which piece of work this node is responsible for? 2. Which transactions should be classified into which segments? 3. Does each node only need to store its own transaction information (ledger)
according to the implementation of the above problems, we can divide the fragmentation into three types: network fragmentation, transaction fragmentation and state fragmentation
Network fragmentation: how to divide the nodes of the whole network into different partitions
transaction segmentation: how to divide the whole network transactions into different segments
State fragmentation: how to make each node only maintain the account book in its own fragmentation without affecting the security of the whole system
the difference and connection between main chain and fragmentation chain
we have understood the types of fragmentation, so what's the difference between main chain and shard chain
turn left | turn right
the realization of Ethereum fragmentation is a long process. Even vitalik himself said that it will be graally realized by stages. We'd better look forward to whether fragmentation can move from theory to practice
after enter, the indicator light next to the network interface corresponding to eth0 will flash, so that you can quickly determine the location of eth0 network interface (press Ctrl + C to stop flashing)
also determine eth1 and eth2...