Ethereum 2.0 — or ETH2 — is a frequently used term. It is considered crucial for the future of the second-largest blockchain. What it means exactly, however, is not so easy to put into clear words. We will try it nevertheless.
There is hardly a blockchain where so much happens as on Ethereum. The home of the overwhelming majority of tokens and smart contracts, the blockchain that has the greatest appeal to developers and companies.
However, Ethereum has not yet officially taken on its true, intended shape. This is to be achieved by taking the step to “Serenity”.
Actually, the transition to Serenity was planned by a simple hard fork. Since it is considerably more complicated than initially expected, this transformation will now take place in a lengthy process. The term “Ethereum 2.0” describes both the goal and the way to get there. The word is read in many crypto media, and there are dozens, probably even hundreds of articles on it. Most of them, however, are either too simple and general or too technical and detailed.
In order to understand what ETH2 means, three texts, in particular, have helped me: The Ethereum 2.0 Roadmap on ethhub.io, Ethereum 2.0 for Dummies, written by Chromatic Capital, a San Francisco-based investment fund, and the official design notes of Ethereum 2.0. All articles are very up-to-date.
Why ETH 2.0?
Ethereum as it is today is slow, explains Chromatic Capital. The blockchain only manages 15 transactions per second, “and any complex application that scales up will be extremely expensive. Visa, on the other hand, can handle thousands of transactions per second. Ethereum is slow because it’s decentralized and needs to stay decentralized.
You could rely on supernodes, on the “consensus nodes”, according to the design notes, “having servers strong enough to process every single transaction.” That would be the simple solution. The blockchain could stay as it is, and you would just have to parallelize the client processes more.
The Ethereum developers, however, reject this option, mainly out of concern that it would compromise decentralization and censorship resistance. We know this argument: the good old blockchain trilemma of decentralization, scalability, and security. It became the basis of all “second-generation blockchain” projects, for example, IOTA, which claims to be able to solve this trilemma. Most of these approaches, including ETH2, replace mining with another consensus mechanism.
The old design of Ethereum is thus more or less considered a failure. ETH2 is, therefore, according to Chromatic Capital, “a completely different project, with a zero-to-one paradigm, like blockchain slaving.” ETH2 is not just an upgrade — it reinvents chains. It is designed to make Ethereum more secure, scale to thousands of transactions, improve programmability — and all this without compromising decentralization.
How does ETH2 solve the problem?
ETH2 will solve the classic blockchain trilemma by three main achievements: sharding, proof of stake, and a new virtual machine.
What does this mean?
Sharding is a method of splitting databases across multiple servers, i.e. partitioning them. Just as a computer has several partitions on one disk, but remains on the same disk, the blockchain should be split into several parts without losing its coherence. The security of a “splitter” (“shard”) should be as high as that of the entire system.
Proof of Stake means that there are no more miners that form blocks by using energy. They are replaced by Stakers: nodes that prove that they have a certain amount of ETH. This is not only more energy-efficient but should also ensure faster finalization of a transaction and greater decentralization of the validator landscape.
The Ethereum developers have been working for years on a consensus process that will finalize transactions quickly and definitively and ensure that Stakers follow the rules. With Proof of Stake, Ethereum transactions should be finalized within seconds, while the blockchain is resistant to 51 percent attacks.
Finally, the new virtual machine should allow developers to create their own execution environment. This allows the rules of other blockchains, for example, BTC or Zcash, to be introduced on a shard. Ethereum should thus become a universal construct of various blockchains.
So far, there is no functional sharding of blockchains at all, while many smart people are arguing about whether Proof of Stake actually works. The Ethereum developers set out to solve several challenges of blockchain development and thus create an architecture that leaves the core constraints and tradeoffs of blockchain development behind.
The Beacon Chain
The center of ETH2 will be the Beacon Chain. This new blockchain to be created can initially coordinate up to 64 shard chains. Chromatic Capital describes it as ‘a kind of command and control center’ of the entire ETH2 network.
The Beacon Chain manages the “Caspar Proof of Stake” protocol. It notes validators and their stakes, randomly selects block validators and assigns them a random shard chain. The Beacon Chain applies the consensus rules, executes sanctions for dishonest validators, and serves as a link between the shards.
The validators of ETH2 vote by proof of stake, i.e. with their coins, on so-called “certificates”. As soon as a block of a shard chain has collected enough such certificates, it receives a so-called crosslink on the beacon chain. From this moment on the block is “finalized”, which means that it is no longer possible to undo it. It now offers the same security as a block on the Beacon-Chain, which is the main chain. The procedure is vaguely reminiscent of the consensus at Ripple and what IOTA is planning: The nodes on the network vote in some way on blocks.
As a reward for their work, the validators receive ETH2. This is the native coin on the beacon chain. In the long run, it will replace ETH, and will be distributed without a fixed limit to guarantee the security of the blockchain in the long run.
The beacon chain should go live in the second or third quarter of 2020. This will initiate phase 0 of the transition to ETH0. It will be possible to run a beacon client and pay its ethers — now called ETH1 — onto the beacon chain, where they will be switched to ETH2. This way is a one-way street, the ETH1 are effectively burned.
So if phase 0 is successful, there will be two ethereum blockchains: eth1 and eth2. For users, eth2 is not yet interesting at this stage, as no shard chains can be attached to it yet. It is more or less a skeleton without a body.
The Shard Chains
The first Shard Chains are expected to go live in 2021. These shards will initially store the data in raw form. They are not able to process data according to certain consensus rules or even to display account balances of users or addresses. They are only used as a laboratory to test the shard architecture, but can probably also be used as data storage to help DApps scale better.
The focus of Phase 1 is on the work of the beacon chain: It now starts to reference the states of the shards by cross-links and thus finalizes them. You can now imagine them as a blockchain, to which other — so far still non-functional — chains are continuously attached while rotating around them.
There is probably not much more to say about this phase. There will be hundreds or thousands of details, discussions, small and large updates to bring Ethereum to this point. Getting a new Proof of Stake system up and running and setting up a blockchain to act as a control center for various data chains is a huge challenge.
In this phase, the eth1 and eth2 blockchain will continue to work in parallel. Things get really exciting from phase 2 onwards.
Phase 2: The fulfillment of the vision
The following phase, according to the roadmap, will “fulfill the true vision of ETH2”: the shards will be able to process transactions. From this phase on, the ETH2 network should also be usable for real applications. In the further course of the project, smart contracts will also be activated on the shards. Then the new execution environment mentioned above will take effect, which allows to activate own rules on the shards, which for example emulate other blockchains like Bitcoin, Zcash, or ETH1.
Chromatic Capitals explains this with a nice picture:
“Just as ETH1 abstracted Bitcoin to implement smart contracts on a blockchain, ETH2 will abstract ETH1 to redefine the fundamental nature of blockchain computing”. Phase two “unifies the functionality of all chains.”
The shard chains, formerly pure data storage devices, will now become structured chains that combine the states of smart contracts with a consensus process. Developers can form their DApp on a shard, Ethereum becomes a multi-chain system governed by the Beacon chain.
Phase 2 is expected in late 2021 or early 2022. According to the roadmap, many parts of it, such as the conceptualization of the execution environments, are “still in intensive research and development”.
It is planned that ETH2 will be rolled out in six phases. However, from phase 3 onwards, according to the roadmap, “any kind of prediction is speculation, as much is likely to change.” At some point during these phases, ETH1 is expected to die of thirst. Whether this should happen as planned, or whether ETH2 should replace the old blockchain, because all the ethers and users have migrated step by step, is probably not yet decided.