Today, Ethereum is a Proof of Work blockchain that enables developers to build and deploy decentralized applications. With eth2, the protocol is transitioning to Proof of Stake and improving its scalability and decentralization without sacrificing security.
eth2 has been envisioned since Ethereum’s genesis in 2015. However, it took many years to rework the current protocol design so that Ethereum can remain accessible for regular users to run validators and secure the network. eth2 is optimized for hundreds of thousands of validators with low startup costs and forgiving slashing and downtime penalties.
eth2 represents the first time a major protocol has attempted to completely rebuild itself from scratch under a new consensus mechanism. Ethereum’s mainnet is currently the second-highest market cap blockchain network, second only to Bitcoin. With millions of users, tens of thousands of dapps and companies, this change is no small feat.
The eth2 research team has planned a phased transition that will allow all current users to shift to the new consensus mechanism without interruption. During the early phases of the transition, Ethereum 1.x and eth2 chains will run in parallel, with their own state and inflation; eventually, Ethereum 1.x will be merged into eth2.
According to a recent report from Consensys, almost 50% of users with greater than 1 ETH plan to stake (i.e. using a pooling provider like an exchange), and of those tokenholders with at least 32 ETH, almost 90% plan to stake some portion of their holdings. Helping eth2 launch supports the future of Ethereum. It’s also an opportunity to secure the chain and earn staking rewards.
“The transition to eth2 is incredibly important for the crypto ecosystem because it allows one of the most important and most adopted blockchains—with millions of users and tens of thousands of dapps—to introduce new scalability options for the protocol and that’s a big deal.” —Joe Lallouz, CEO of Bison Trails
Phase 0 is the genesis of the eth2 PoS chain. The blockchain will be producing blocks, managing validators and their stakes, performing consensus, selecting block proposers, organizing validators into committees to vote on the proposed blocks, and applying rewards and penalties such as slashing. However, there will be no transactions, smart contracts, shards or any real usage outside of securing the chain, during this phase. The goal of Phase 0 is to bootstrap the new consensus.
The Beacon Chain is not a testnet; it is a mainnet with limited functionality but real consensus and consequences. Real ETH will secure the Beacon Chain. In order to get ETH into the chain, the Ethereum Foundation built a deposit contract. This smart contract lives on the Ethereum 1.X blockchain and allows users to move ETH from Ethereum 1.X onto the eth2 Beacon Chain in chunks. 32 ETH are needed to create a validator.
The contract is unidirectional: once you move your ETH, it can not return to Ethereum 1.X until Phase 1.5. The Beacon Chain constantly watches this smart contract and when there’s a new deposit with certain data fields, an equivalent amount of Beacon ETH (BETH) is created on the Beacon Chain and credited to a specified address.
In order for the Beacon Chain to launch its genesis block, at least 524,288 ETH must be deposited in the deposit contract to create a minimum of 16,384 validators. When this minimum is reached, the network will launch!
Phase 1 doesn’t have a set release date. Given the amount of research and development needed to get there, we estimate the earliest Phase 1 will launch is later in 2021. 64 shard chains that run in parallel and interoperate will be introduced. This phase will test eth2’s ability to come to consensus on 64 concurrent chains. Shards also introduce an additional scalable data layer, which can provide massive scalability via technologies like "rollups."
Phase 1.5—the merging of Ethereum 1.X and eth2 into a single chain—is one of the most anticipated moments of the transition process. The exact method for the merger has not been finalized but there are many promising possibilities.
The current plan is to make the Ethereum 1.X chain a single shard of 64 in the eth2 Beacon Chain. This plan prevents a break in continuity, state, or data history. ETH token holders won’t have to undergo any token transfers or do other work. Dapps won’t have to make any changes to adapt to the new chain. When Phase 1.5 starts, the Ethereum 1.X chain will stop existing as a PoW network and eth2 will be fully operational on a single shard. We anticipate this phase will begin in 2022.
Phase 2 is the furthest away and the least defined. By Phase 2, PoS and sharding will be successfully implemented and transactions and smart contracts will be enabled on the remaining 63 shards. In short, the Beacon Chain will be fully functional. We anticipate this phase will begin in 2023.
eth2 is a unique Proof of Stake protocol design with notable innovations.
“With millions of users, tens of thousands of dApps and companies, and billions of dollars of value involved with the protocol, the transition to eth2 will be no easy feat. As a platform providing validators for the earliest adopters to join the network, we’re excited to power such a historic moment.” —Joe Lallouz, CEO of Bison Trails
Proof of Stake consensus on eth2 functions using two key processes:
When the network is functioning normally—when two-thirds of validators vote on new blocks—the blocks are added to the head of the chain by GHOST and are justified and finalised by FFG in approximately 12.8 minutes.
If there is an attack, or a large number of validators are offline, GHOST will continue to add new blocks to the chain while FFG stops finalizing blocks until the network is stable. GHOST may change the head of the chain during this period. If blocks continue to be produced but not finalized, eth2 uses a mechanism called an inactivity leak to slash the balances of offline validators until enough stake is online to achieve finality.
eth2 will have a Beacon Chain in Phase 0 and 64 shard chains in Phase 1. The Beacon Chain holds the shards together and enables consensus and cross-shard communication. It has a registry of validator addresses, the state and balance of each validator, attestations, and links to shards; it is the central controller of the eth2 system.
eth2 has 12-second-long slots and epochs of 32 slots / 6.4 minutes. A slot is a chance for a block to be added to the Beacon Chain and shards. A block is not required for every slot. In optimal conditions, for every slot one Beacon Chain block and 64 shard blocks will be created.
Every slot, a validator is pseudorandomly chosen by RANDAO (a random beacon) as a block proposer. At least one committee is also chosen as attesters.
A committee is a group of validators with a target size of 128 validators. As attestors, the validators review the proposed beacon blocks and shard blocks, and vote on them. These votes, called attestations, are recorded on the Beacon Chain and determine the chain head of the Beacon Chain and each shard.
This process ensures a block was made correctly. The more attestors involved, the more confident there is that the block is well made. The process also provides assurance that the next validator selected will produce a block on the right chain.
As the number of validators increases, more committees are assigned to a slot. The maximum number of committees per slot is 64. When shards are added, committees will not just attest to a specific slot, but will also crosslink to a particular shard to keep the protocol in sync.
The block in the first slot of an epoch is the checkpoint. This is how the chain periodically agrees on finality. It is finalized in two epochs (12.8 minutes).
While validators are the entities participating in consensus, validator clients are the nodes that hold the validators. The validator clients connect to public-facing beacon nodes to enable the validators to participate on the chain. Beacon nodes follow and read the full Beacon Chain, similar to an Ethereum GETH node does today.
“At Prysmatic Labs, we're excited that Bison Trails is supporting eth2 and integrating the Prysm client into their infrastructure platform. Bison Trails has been an active participant in our testnets and we look forward to ongoing collaboration with the Bison Trails team as we work toward the release of eth2 Phase 0.” —Preston Van Loon, Co-founder Prysmatic Labs
eth2 allows participation on the network for a broader set of token holders than most PoS chains. 32 ETH represents a relatively low staking requirement and low entry point for running a validator that earns rewards. Please note: there is no delegation or stake pooling at the protocol level.
Only 900 validators can be activated a day (4 per epoch) until 327,680 active validators are in the network. Once that goal is achieved, 1125 validators a day (5 per epoch) can be activated in an epoch.
The number of validators that can be activated every day increases by 225 (an additional 1 per epoch) with every additional 64k validators. The same pattern applies to validator exits. This could result in congestion—a wait of a few hours or days—if validators attempt to join or leave the network simultaneously.
Validators earn three types of rewards for participating in consensus.
Validators remain active on-chain until they exit. Exiting can happen in one of three ways: voluntarily, due to insufficient balance, or due to slashing.
If a validator choses to voluntarily exit, they get placed into the exit queue and exit 4 or more epochs later. Validators can lose funds due to penalties—such as being offline for too long—which decreases their starting balance of 32 BETH. If the balance falls below 16 BETH, the validator is ejected from participation. A slashed validator immediately begins the exit process. Please note: It is possible to deposit more ETH into the deposit contract to add BETH to a validator’s balance, to prevent a low balance.
Until Phase 2, it is not possible to withdraw the BETH balance—initial deposit plus any earned rewards—on a validator even if the validator exits the active set. Once a validator has exited, it will not be able to rejoin.
BETH (shorthand for ETH on the Beacon Chain)
|Total Initial Supply||524,288 BETH (needed to launch Phase 0)|
|Total Planned Inflation||Infinite|
|Maximum Token Supply||Infinite|
|Token Price at Sale||N/A|
|Inflation||Variable, but expected to be:
.17% in the short term
.54% in the medium term
|Target staking rate||N/A|
|Expected number of nodes||10,000+ validator client nodes, 100,000+ validators|
|Expected rate of reward||Variable, but expected to be:
18.10% in the short term
5.72% in the medium term
|Maximum Stake||32 ETH per validator|
|Minimum Stake||32 ETH per validator|
|Lock-up Period||Variable depending on amount of simultaneous unbonding. Minimum of 27 hours.|
|Reward re-invested method||No compounding beyond an effective balance of 32 ETH|
|Reward payment frequency||All rewards and penalties are applied at the end of each epoch (6.4 minutes)|
|Reward Payouts Mechanics||Rewards accrue on validator until withdrawn by the withdrawal key|
The inflation rate on eth2 decreases as the staking rate increases. Because so much depends on how many people lock up ETH to support the transition to eth2, it is difficult to clearly identify a rate of inflation and reward rate. Given the risks and opportunities apparent with eth2, we assume that:
Rewards are designed to incentivize early validators starting at a 23% rate of reward at network launch. Anyone considering participating on eth2 benefits from running their validator as early as possible.
Rewards do not compound on eth2. They are earned based on the effective balance on the validator, which is capped at 32 BETH. As long as the validator is earning rewards and not being penalized, the true balance will start at 32 BETH and increase while the effective balance will remain consistent at 32 BETH. After Phase 2 is live, validators will be able to withdraw all their rewards, combine them, and run additional validators.
On eth2, validators can experience penalties and slashing. Penalties are significantly less severe than slashing.
Penalties act like negative rewards. If a well-performing validator can make a 10% reward rate in a year, a poorly-performing validator could lose 10% in the most extreme circumstances, such as being offline regularly or voting on blocks that do not get finalized. In most circumstances, the penalty for being offline is simply not earning rewards. If a validator is online greater than 53.6% of the time, assuming no inactivity leak, it will break even.
Slashing carries a heavy punishment for validators that commit a significant offense, burning a minimum 1 BETH and a maximum of the entire stake. The validator is automatically ejected from the Beacon Chain active set.
The two slashing conditions for validators are:
Emitting a slashable vote is uncommon for validators that correctly follow the protocol: forming a slashable message without malicious intent only occurs as a result of some bug or accident. Because these errors are bound to happen, eth2 minimizes the slashing amount by destroying stake in proportion to the number of validators slashed around the same time.
This method assumes that a small number of validators commiting a slashable offence around the same time is unlikely to be an attack on the system as a whole. These validator mistakes are punished lightly (a minimum of 1 BETH). However, if a large number of validators commit an offence at a similar time, a large amount of their stake is burnt (up to their full balance) because an attack is assumed. This is called correlated slashing.
All slashed validators, whether due to a mistake or a malicious act, are forcibly exited from participating in the protocol and can not return until Phase 2. This prevents validators from further harming themselves by being slashed again and from harming the network in case of an attack.
Beyond penalties and slashing, there are two other forms of punishment on eth2: inactivity leak penalty and forced exit.
To support eth2, Bison Trails will offer a suite of enterprise products including:
The Bison Trails platform will make it easy to interact with the Beacon Chain, stake ETH, and automatically manage validators, validator clients, and beacon nodes.
Bison Trails’ autoscaling software will manage customers' infrastructure automatically, as network requirements change and customers choose to add more validators. This innovation will enable Bison Trails’ customers to continue earning rewards without the hassle of manually managing participation.
“The transition to eth2 will be hard, but more significantly, it’s momentous for Proof of Stake, for Bison Trails, and really important for crypto. eth2 is the first time a protocol that is this big tries to move to Proof of Stake and adds support for the scalability features that a lot of the new protocols are introducing. As a platform providing validators for the earliest adopters to join the network, we’re excited to power such a historic moment.
As a validator, you’re part of the moment. As a platform where people can launch validators, we’re powering the transition to eth2. We’re doing the work to enable people to participate.” —Joe Lallouz, CEO of Bison Trails
Contact us to learn more about running an eth2 validator on the Bison Trails platform or about any of our enterprise offerings in support of eth2.
Bison Trails is a blockchain infrastructure company based in New York City. We built a platform for anyone who wants to participate in 19 new chains effortlessly. We also make it easy for anyone building Web 3.0 applications to connect to blockchain data from 27 protocols with QT. Our goal is for the entire blockchain ecosystem to flourish by providing robust infrastructure for the pioneers of tomorrow.
eth2 Update 007Dec 3 2020
eth2 Update 006Nov 25 2020
Substrate Ecosystem Update 003Nov 25 2020
Now Available: Diem (Libra) QTNov 24 2020
eth2 Insights: Network PerformanceNov 23 2020
eth2 Update 005Nov 18 2020
eth2 Insights: Validator EffectivenessNov 16 2020
tBTC and DeFi: How to Get InvolvedNov 12 2020
Coinbase Custody Expands Bison Trails Integration to Add Staking Support for CeloNov 12 2020
eth2 Update 004Nov 11 2020
QT Archival: When a Full Node is Not EnoughNov 11 2020
Bison Trails Newsletter 009 • October 2020Nov 10 2020
View More →