Opinion | Ethereum will continue to develop after Phase 2, the future blueprint is still unclear but the future looks bright

Author: Grant Hummer

Translator: Sun Wei

Production: Blockchain Base Camp

Editor's Note: The original title was "Ethereum 2.0 Bright Future"

Anyone who has followed Ethereum for some time knows that the history of Ethereum's development has been accompanied by promised extensions, bounced tickets and poor communication of future plans. This has led many to classify Ethereum as a failure of experiments. This pessimism is understandable. Today's Ethereum (ETH1) runs slowly, the speed of the entire network is limited, it can only process 15 transactions (TPS) per second, and it costs a lot to perform large-scale complex operations. In contrast, Visa can handle tens of thousands more transactions per second.

Why is Ethereum slow and expensive? Simply put, it's because of (currently) high decentralization costs. Decentralization is expensive. Like most blockchains today, every node on Ethereum (defined as a computer connected to the network, such as a computer running Ethereum software) must be networked for all operations to ensure all participation. All follow the rules. This consumes actual energy and resources. In terms of time cost, the nodes running Ethereum are spread all over the world-after all, this is a public network that provides numerous incentives for participants-at the same time, the communication delay between different nodes is high and the computing power is different. Therefore, the network needs to maintain sufficient latency so that slower nodes can “follow” and continue to contribute to the network's decentralized voting. If the current architecture is designed, the Ethereum network will process too much data in a very short time, and the user's hardware (such as a notebook or personal server) cannot keep up, and the only functional node on the network is only a large data center Already. This will seriously damage the decentralization of Ethereum, because these data center nodes can easily constitute a monopoly and take over the network or exercise censorship control over the network. Many encrypted networks today, such as EOS, experience this problem in practice.

It is important to maintain sufficient levels of decentralization in the public blockchain. Without decentralization, using a blockchain would not make much sense, because a centralized blockchain can easily review the transactions of the general public, and it is much less efficient than a conventional distributed database. For the new readers of the blockchain, we can recall: Before Satoshi Nakamoto invented the new form of decentralized management in Bitcoin, virtually no system could implement "detrust control" ——Because all participants will influence the decision, but no one party can have all or most control. Ethereum's creator Vitalik Buterin's famous answer about weighing decentralization and performance states that anyone can build a high TPS system as long as it is a "centralized garbage dump braving steam".

Suppose it is possible to construct a blockchain without the need for each node to process transactions of other nodes, and also reduce the communication consumption by processing only a small part of the entire network transaction?

We have ETH2.

ETH2 is the next generation of Ethereum, and even it is a bit inappropriate to call it Ethereum. This is a completely different project, using a new 0-1 paradigm on the large-scale operation of the blockchain. ETH2 aims to improve the scalability, security, and programmability of Ethereum. The processing of 15 transactions by a single chain has become history. ETH2 can process thousands or even tens of thousands of transactions (or even more) through a single chain per second without loss of decentralization. In fact, unlike the Proof-of-Work (PoW) mechanism currently used in Bitcoin and ETH1, ETH2 will introduce a more economical and secure consensus mechanism, namely Proof of Stake (PoS). In traditional PoW blockchains (such as Bitcoin), new Bitcoin production and transactions are processed by miners, that is, individuals and organizations that use expensive hardware to solve complex mathematical problems. Miners guarantee network security in exchange for transaction fees. In the PoS mechanism of the blockchain (such as ETH2), the situation is different. New Ether is produced and traded by validators who lock their own Ether to provide security for the network. In fact, the security provided by the verifier depends on the value of the network itself. If a verifier behaves improperly (such as approving a malicious transaction), it can significantly reduce its ether. This mechanism gives validators an incentive to comply with the rules of the agreement.

Another important reason why PoS security is outstanding is that PoW mechanisms are vulnerable to so-called "spawn camping" attacks. If an attacker can accumulate enough mining hardware to attack Bitcoin or other PoW chains, Bitcoin cannot prevent further attacks, because the network will continue to restart or hard fork, and will eventually only be unlimited by the same mining hardware. Attacks. In contrast, Ethereum has greater flexibility in responding to such attacks-it can hard fork and cut the rights of the attacker, which is equivalent to burning the attacker's bitcoin mining farm.

In addition, ETH2 enables developers to create their own transaction processing methods, ie execution environments. As a result, developers can use different blockchain rules within Ethereum when needed. In order to greatly simplify the execution environment, ETH2 gives users the right to use Bitcoin rules, ZCash rules, ETH1 rules, and many other possible rule sets for transactions.The scale is several orders of magnitude higher than the current, and at the same time, the same large, Capitalized validators provide security protection. ETH2 can achieve this through "shards": Each shard in ETH2 is similar to a blockchain, with its own unique generator and validator, but each shard is closely connected with other shards and can be mutually connected Communication to form a large network fragmentation chain. In this way, the designated verifier on ETH2 does not need to process a single transaction in the entire network, and only needs to process and verify transactions on a single shard. This is an innovative technology that makes it more meaningful for participants using client hardware to participate on the ETH2 network.

This is important: each shard in ETH2 will share the same security with other shards, and in order to break a single shard, the entire system has to be broken. On Cosmos, each chain is responsible for its own security, thus forming a decentralized and vulnerable network. Compared to platforms like Cosmos, this security model provides better security. In this way, in order to destroy ETH2, attackers must purchase and mortgage billions of dollars worth of ether, and if the price of ether rises, the amount will increase.

This is an image of a traditional blockchain network-a single "chain" of data blocks. No need to bother to understand the meaning of block header or transactions list, this picture is only for illustration:

This is an image of ETH2, but there are actually more than two shard chains, but 64. All shards are coordinated with each other using a beacon chain. Similar to the picture above, don't worry about the technical details:

ETH2 is built by 9 different engineering teams, all of which have received funding from the Ethereum Foundation and received extensive support from the Ethereum community. These teams are building clients–something similar to a web browser (such as Chrome or Firefox), except that the client is used to access and participate in the Ethereum network. The diversity of clients is one of the core principles of Ethereum. The prerequisite criterion is that if one or two clients crash or there are bugs, the entire network will not crash. Each client targets a different niche usage, but all clients can participate in the network. For example, one client is optimized for smartphones, and the other is built for enterprise use. The engineering team builds ETH2 based on specifications created by talented researchers (mostly PhDs in computer science or the like) from the Ethereum Foundation, ConsenSys, and the broader academic community. Much of the research is conducted on the ethresear.ch website in an open source way, and anyone around the world can post effective technical ideas, suggestions or criticism. After years of iterative iterations, these specifications have reached their current state: for a while researchers thought that a practical and effective design had been made, but then found some defects that led to the need to redesign some or even the entire system. It is no exaggeration to say that the ETH2 research team is the most experienced and talented protocol design team in the entire blockchain field.

The following is a list of ETH2 research teams (in random order):

ChainSafe Systems

  • Headquartered in Toronto, Ontario, the company has more than 30 people, of which 5 developers are responsible for the research of the ETH2 client;
  • Grassroots developers, met at the Toronto Ethereum Developer Exchange;
  • Built Lodestar client written in JS.

PegaSys

  • ConsenSys (the largest company in the Ethereum ecosystem) with a team of more than 50 people, with members scattered around the world;
  • Constructed a business-friendly ETH1 client Pantheon;
  • There is a team of researchers working on the ETH2 specification;
  • Constructed a Java-based client Artemis, optimized for enterprise use.

Harmony

  • A team of 4 in Russia;
  • Building a Java Ethereum client;
  • Limited public information to date;
  • May be merged with PegaSys / Artemis.

Parity Technologies

  • A team of 60+ people in Berlin with a large number of remote employees (not everyone works on ETH2);
  • Received $ 5 million from the Ethereum Foundation for constructing the ETH2 client Substrate Shasper;
  • At the same time, it is also the team behind the interoperable blockchain project Polkadot, which is considered by some as an Ethereum competitor
  • Building and maintaining Parity, the second hottest ETH1 client

Prysmatic Labs

  • A team of 6 people scattered across the United States;
  • The update blog is very active, and its blog is a good way to learn more about and follow up on ETH2;
  • Built the client Prysm, which can be used for multiple mainstream uses, similar to Geth in ETH1.

Sigma Prime

  • 4-person team based in Sydney, Australia;
  • Strong academic and software development background;
  • Built the client Lighthouse, suitable for mainstream use cases.

Status

  • A team of 8 people scattered in Europe;
  • Develop lightweight client Nimbus for smartphones or other "light" environmental uses.

Trinity

  • A four-person team signed with the Ethereum Foundation;
  • The built lightweight client will be used for prototypes of more advanced implementations.

Nethermind

  • A team of 7 in London and Poland;
  • Build a Windows client for ETH2 in .NET.

The release of ETH2 is divided into multiple stages, but the most important of these are the first three stages. These stages staggered the launch of the various components of ETH2, not only because it took time to build each component correctly, but also because the gradual introduction of changes would significantly reduce technical risk. For a multi-billion dollar crypto network, "fast action to break the dilemma" is not wise. Unlike centralized technology companies that can quickly roll out fixes, encrypted networks with severe bugs may experience severe damage for a long time, as the network participants must manually update to the latest client version, the worst In the case of an attacker (such as an attacker trying to get new Ether out of thin air), a large number of transactions must be rolled back to ensure the security of the network. This will have a huge negative impact, as caused by DAO hackers in 2016, causing Ethereum developers and users to separate their own chain, Ethereum Classic.

Phase 0: May be delivered by the end of the second quarter of 2020, or the beginning of the third quarter. Introduce the beacon chain as a kind of "command and control" center for the entire ETH2 network. This beacon chain will organize validators, assign them to different shard chains, and be responsible for processing transactions for that shard before randomly shuffling to another shard. Reorganization is an important part of the security of ETH2. The lack of reorganization will cause a small number of validators to be bribed or destroyed, infecting the operation of ETH2. The beacon chain is also responsible for proving the accuracy of the data blocks generated by different shards, so that ether, tokens, and data can be exchanged between different shards. To use a California term from the 1970s to describe it: a beacon chain is a blockchain that manages other blockchains.

If the beacon chain is command and control, the verifier handles construction work. A validator is a participant on the PoS network, responsible for processing and booking network transactions, and forwarding these content to each other and to other participants. They can get transaction fees and newly produced currency for this. The verifier can be said to be the most important role in the encrypted network, and designing an incentive mechanism for them is also the core challenge that ETH2 wants to solve. The beacon chain is also responsible for cutting down on misbehaving validators.

Phase 1: It will be delivered in 2021, starting the shard chain by itself, but cannot process or build data on the shard. These data will exist on the shards in the original form (that is, data blob), mainly to ensure that the shard chains can communicate with each other and pass information to the beacon chain. Shards cannot perform any calculations on the data. Phase 1 initially felt like an optimized testnet with limited use, but this was not the case. New or existing DApps can take advantage of these data stores to increase scalability. For example, the decentralized Twitter DApp can store data on the shard chain and perform most of the calculations off-chain, while ETH1 provides a high level of security for it. In Phase 1, ETH1 will still run normally, while the 1.0 and 2.0 chains will run simultaneously.

At the end of December 2019, Vitalik had proposed to speed up the transfer of ETH1 into the beacon chain so that Ethereum could realize the usefulness of PoS and integrate with ETH2 faster. Initially, ETH1 will exist as a special shard (not an execution environment), but in phase 2 it will transition to an execution environment. There are currently no plans to indicate when this will be achieved, but it looks like it will appear in Phase 1.

Phase 2: It is expected to start in 2021 or 2022, which will also realize the true vision of ETH2. By then, the shards can process transactions and the ETH2 network can be used in actual business applications. Fragmented smart contracts will be introduced in Phase 2, accompanied by the corresponding execution environment for running smart contracts. As stated earlier, the execution environment enables developers to create arbitrary rulesets that process their transactions. This also means that protocol developers can create imitating Bitcoin, Zcash, ETH1, or any other desired method to create an execution environment for the calculation of blockchain transactions. In the same way that ETH1 summarizes Bitcoin to launch a smart contract, ETH2 will summarize ETH1 to redefine the nature of blockchain computing.

There are still a lot of unresolved research issues on some components of ETH2, such as how to maintain the provision and storage of data in a decentralized manner (avoiding a small number of data providers monopolizing and controlling the network), how transactions across shards operate, and the execution environment How to manage and more. However, research on these issues is progressing rapidly.

Once again, the three phases of ETH2:

Phase 0:

  • Delivered at the end of the second quarter or the beginning of the third quarter of 2020;
  • Enable the beacon chain and implement PoS verification on the beacon chain;
  • ETH1 will continue to operate normally.

Phase 1:

  • Delivered in 2021;
  • Start shard chains, allowing data to be stored on these shard chains, but transactions are not processed on the shards;
  • ETH1 may continue to operate normally or be incorporated into ETH2 as a special shard.

Phase 2:

  • Delivered by the end of 2021 or early 2022;
  • Allow transactions on shards;
  • After a period of time, ETH1 will be merged into ETH2 as the execution environment.

Ethereum will continue to develop after Phase 2, but the future blueprint is still not clear. Some possible directions include: using ZK-SNARK to increase scalability, further development of lightweight clients, and secondary sharding (basically referring to resharding within a shard). In the long run, it is expected to use a consensus mechanism such as CBC Casper to abstract the shard itself.

From the author's perspective, the future of Ethereum is bright. Ethereum has a large number of researchers, developers, users, and projects. No other smart contract chain can match it, and it is time to intervene. Ethereum is facing the biggest and most important upgrade possible, and if successfully implemented, it will create a lot of value for the entire world.