Ethereum is getting closer to the second generation of its design, known as ETH 2.0.
Ethereum 1.0, like Bitcoin, has used Proof of Work (PoW) for security and consensus. However, the Ethereum community will soon be transitioning away from PoW and will be implementing its own Proof of Stake (PoS) algorithm. We wrote about the tradeoffs for both approaches in a previous post. Check that out if you want to learn some more about the positives and negatives of PoW and PoS.
Improved Accessibility and Scalability
The Ethereum community sees PoS as a superior way to scale Ethereum to the levels of usage, adoption, and performance its supporters are aiming for. The community also wants to make the barrier to participate in consensus and security much lower than it currently is within its current PoW paradigm.
The barriers to be a PoW miner are quite steep for mature blockchains like Bitcoin and Ethereum. It’s very difficult for an individual or a small group of individuals without access to a lot of capital, expertise, and low electricity costs to be competitive miners. Proof of Stake lowers the barrier to participation by removing the need for expensive, specialized hardware and the need to access abundant, low cost electricity. Instead, validators (miners) will download the Ethereum 2.0 client software and put up at least 32 ETH (stake) for potential loss. This opens the door to a larger pool of participants, aiding in keeping the network decentralized. Add in the development of staking pools and staking derivatives that allow holders with less than 32 ETH to participate in staking and you’ve opened the door to almost everyone to participate in at least a small way.
PoS validators will be randomly selected to propose and attest to the validity of blocks on the Ethereum blockchain. Honest validators will be rewarded with new ETH and transaction fees. If a validator is dishonest or incompetent, all or some of their staked ETH will be slashed. Economic rewards for good behavior, economic punishments for bad behavior.
PoW isn’t inherently unscalable, but it’s not as flexible as PoS can be. PoS’s flexibility allows protocol developers to experiment with techniques like sharding that, if successful, could increase the throughput and performance of Ethereum immensely.
Because of the nature of PoW blockchains, transactions and blocks can only be processed sequentially. All transactions have to “wait in the same line” to be verified. With PoS, via a technique called “sharding”, Ethereum developers think they can “parallelize” transaction processing. This means instead of processing transactions one at a time or “sequentially”, transactions and blocks can be processed simultaneously without waiting in the same line.
Imagine two theme parks: One theme park has one line and can only process one person at a time. The other theme park has 15 lines that can process one person in each line simultaneously. This is a 15x improvement in throughput! The second theme park is parallelizing its check in process, whereas the first park is using sequential processing.
Sharding
Sharding is new to blockchains, but it isn’t a new concept. Database architects have used it for years to dramatically improve the performance of databases. Both proof of stake and sharding research are attempts to maintain or improve security while increasing throughput and lowering the barrier to participating in consensus. Ethereum related PoS and sharding research started as early as 2014 but Ethereum won’t be the first blockchain to implement it. Researchers in 2019 from MIT created a cryptocurrency named Vault, that was the first to incorporate the concept of sharding to blockchains.
Sharding is a way of splitting up responsibility amongst the nodes on the network. Traditional blockchain architectures have every node process, store, and validate the same data. Sharding divides these responsibilities between many nodes, which allows transactions to be processed in parallel rather than sequentially in a traditional blockchain.
Each “shard” has its own state and transaction history within a larger network of shards that also have their own unique state and transaction histories. Each shard is a nested blockchain: a small blockchain operating and coordinating with other small blockchains to create a large, emergent Blockchain, known in the Ethereum 2.0 development community as the “beacon chain”.
You can think of the first iteration of the Ethereum blockchain as having a single giant shard. Eth 2.0 will break that single shard into 64 different shards that share data with the beacon chain and communicate with each other. In theory, instead of 1 block being processed at a time, 64 blocks will be processed simultaneously or in parallel.
Within each individual shard, there are nodes called “proposers” that create a cross-link on the main beacon chain. These cross-links are data structures that contain important information about the shard a specific proposer is responsible for. Cross-links contain mini-descriptions of the state and the transactions of a specific shard. Cross-links and inter-shard communication allow the Ethereum network to create a shared state amongst the different shards.
Although sharding has the potential to dramatically improve the performance and scalability of blockchains, it doesn’t come without its risks and challenges. If just a single shard is compromised the entire network is put at risk. In addition, reliable, incorruptible inter-shard communication is absolutely crucial for this to work. The complexity of making this all work is immense, but Ethereum developers are confident they can use it to take Ethereum to the next level.
There are still many open research questions on this front but Ethereum developers seem to have a good handle on the problem space and what needs to be done. However, execution is always tough.
Transition Phases to Ethereum 2.0
Phase 0
This phase focuses on launching the main beacon chain and getting it to come to consensus with tens of thousands of validators participating in the process.
As the beacon chain goes live and Proof of Stake is implemented, the original Ethereum PoW blockchain will continue to exist. The beacon chain will not be useful to the majority of Ethereum users in Phase 0; it will be unable to process transactions, execute smart contracts, or host dapps. This is done purposively, ensuring PoS undergoes considerable testing as a live, functioning network before dapps and users begin transacting by the millions on a daily basis. To ensure the launch of Ethereum 2.0 does not disrupt the existing Ethereum ecosystem, therefore, the original Ethereum 1.0 blockchain will continue to run in parallel and receive upgrades during Phase 0. In Phase 1.5, the two blockchains will merge to ensure continuity and one single Ethereum network.
Phase 1
The primary improvement of Phase 1 is the implementation of shard chains, the scalability solution mentioned earlier. For Ethereum 2.0, sharding will result in the Ethereum blockchain being partitioned into 64 separate chains (called shard chains) that run parallel to one another and interoperate seamlessly. This setup should be able to process 64 blocks at a time.
Phase 1.5
An important milestone within Phase 1 is the merging of the original PoW Ethereum blockchain with the new PoS chain. This moment is being referred to as Phase 1.5 in the Ethereum community and is the next big step to be made. The legacy PoW Ethereum chain will be brought into Ethereum 2.0 and will exist as one of the 64 shard chains alongside the beacon chain, meaning there will be no break in continuity or data history.
This also means that ETH holders will not have to undergo any sort of token transfer or swap between Ethereum 1.0 and 2.0; rather, they will be able to use their ETH on Ethereum 2.0 seamlessly, with no anticipated extra work on their end or risk of their ETH becoming obsolete. Though the history of the original PoW chain will still exist, it will no longer need the PoW consensus mechanism to be maintained; it will operate as any other PoS shard on Ethereum 2.0.
Phase 2
With both PoS and sharding successfully implemented, Phase 2 will enable––among other things––ether accounts, transactions, transfers, withdrawals, and smart contract execution.
Coda
Ethereum blockchain development continues to progress at breakneck speed and will do so into the future way past the completion of Phase 2. We’re proud to support the asset holders and app developers of this network and look forward to Ethereum’s move from a PoW blockchain to a PoS blockchain.