Ethereum 2.0

If you follow tech news, you may have heard about something refered to as “Ethereum 2.0”. There’s a lot of buzz surrounding it, and for good reason. Ethereum 2.0 (or Eth2) is a series of large upgrades to the underlying protocols of the Ethereum platform. It’s intended to make the network operate faster, more securely, and more efficiently. It is currently being rolled out in phases, with us in Phase 0.

The Problem & Proposed Solution

In its current form, Ethereum suffers from a lot of the same problems as bitcoin. However, this problem is endemic to Proof of Work, and not Ethereum as a whole. Ethereum 2.0 is first step in a never-ending development process to make the network better.

Let’s look at the problems:

Innefficiency: Unfortunately, Elon Musk was correct when he said that cryptocurrency mining is environmentally very damaging. Ethereum is the 2nd largest mined curreny currently, and its operation is incredibly wasteful. Building a cryptocurrency platform capable of powering the world’s computations is going to require a radical restructuring to be more in line with traditional energy consumption. Ethereum 2.0 is expected to cut the network’s power consumption by ~99.5%.
Scalability: Without making sacrifices on decentralization or security, proof-of-work will never be able to handle the massive growth in transaction volume. In previous articles i’ve explained how block size limitations prevent transactions from occuring quickly without incurring high transaction fees. Using an architectural style known as sharding, Eth2 is expected to significantly increase its number of transactions/second without sacrificing decentralization.
Security and Centralization: As cryptocurrencies have grown in population, so has the barrier-to-entry for mining. Today, major parts of mining profits are centralized to a few big players with warehouses of computers in remote parts of the earth. This creates a massive security risk, and prevents new players from entering the game profitably. By switching away from mining, to a different consensus algorithm, operating a node and validating blocks becomes more decentralized, with a much lower barrier to entry. This is best for security because it reduces the chances of 51% attacks by increasing decentralization.

To solve all of these problems, Eth2 relies on 2 solutions: Proof of Stake and Sharding

Proof of Stake

I’m not going to explain here how proof of stake works, as I’ve already done so previously. I highly suggest checking it out here. The first part of the Eth2 upgrade is the switch from being Proof-of-Work based, to Proof-of-Stake based. This comes with all the aformentioned benefits of a more efficient consensus algorithm.

This however, is not happening all at once. It is a multi-step process that occurs over time. It starts with what is known as the Beacon Chain. Started in December 2020, the beacon chain is a seperate chain operating in parallel to the main Ethereum chain. It is operating with proof of stake, and is being used to build consensus. On pre-determined intervals, new empty blocks are being created and verified by all of the stakers. Slowly and over time as the software is developed and tested, transactions are going to be introduced and tested on this beacon chain. At a point in ~2021/2022, the original Ethereum chain and this new Beacon Chain are going to merge. When this happens the beacon chain will be the new main chain, and all transactions will be routed through it. There will not be any more proof-of-work blocks when this occurs.

This new chain in terms of functionality, will be exactly the same as the original chain. It will be able to handle full transactions and smart contract integration. This upgrade is about changing how new blocks are created, not what they can do.

If you are regular user, you won’t even realize this is happening. You will be able to keep sending transactions the same way you always have. You do NOT have to do anything as a user to prepare for eth2. Just continue to hold your coins. Scammers will attempt to get you to send them eth under the guide of exchanging for newer eth. Ignore these as they are scams.

Anyone can become a validator (the new term to replace miners), provided they stake their ether first. This must be done in multiples of 32 ether. If you have > 32 ether, you can only stake 32, and keep the rest to do as you please. Each multiple of 32 entitles you to run 1 validator node. If you stake 64 ether, then you can run 2 nodes, and receive ~2x the reward.

Keep in mind, this does NOT mean that if someone with 32 ether earns 6% interest, you will earn ~12%, only that you will earn 6% of your 64 ether. Each node is independent of eachother and earns roughly the same amount. You just operate multiple of those nodes and collect the sum total of their rewards.

If you choose to stake, the ether to stake WILL BE LOCKED and unable to be withdrawn until eth2 is finished. When this is is uncertain, but will most likely be within the next ~1.5-2 years.

I’m not gonna walk through the technicals of staking, but it is basically a 2 step process. I will give you a top-level explanation of how it works.

You derive a new set of encryption keys. These are your validator keys, and are used to prove your identity in relation to your staked coins. They are generated with a 24-word mnemonic phrase. Even if you don’t validate a single block, when the time comes you can withdraw your 32 ether exactly as long as you have these keys
Send 32-ether or multiples of it to the deposit contract. This is a contract on the main chain that keeps track of the people who have staked their funds and how much.
Run the validator software on a compatible machine. You have to connect to a full blockchain node which you can either do locally or with infura. There are several different softwares built by different teams such as: Teku, lighthouse, Prysm, etc.
Rake in those sweet rewards.

One of the main benefits of Proof of Stake as well, is that it enables a sharded blockchain.

Sharding

While Proof of Stake is meant to solve a lot of problems for Ethereum, it does not help with scalability. This is where sharding comes into play.

Sharding is a common practice in database design and computer science. It involves splitting the information horizontally into “shards” to reduce congestion and distribute the computational load. Imagine a database. Instead of one server holding millions of records, it would distribute fractions of the storage requirements to other servers.

These servers are constantly communicating to receive information when necesarry. In a blockchain context, it involves splitting the blockchain into 64 smaller chains, known as shards. Each node will only have to process and store information on their shard, instead of the entire chain. All of these shards are coordinated by the beacon chain (hence the name).

Original diagram by Hsiao-wei Wang, design by Quantstamp

This allows greater throughput as nodes only have to validate and store transactions on their individual shard. This means that transactions can also occur much faster without sacrificing centralization or security. Distributing the storage and load therefore prevents the nodes from having to store incredibly large amounts of chain data as time goes on, preventing decentralization and making it easy for anyone to run a node regardless of hardware.

When sharding goes live, the original ethereum chain and all its data will be converted into one of the 64 shards.

However, some schematics as well also have the shards only for storing data, and not executing contracts. This is also a valid way of distributing load, because it means that the on-chain data will be able to grow faster without worrying about centralization risk. If the chain grows ~2x as large everyday, but each we split the data storage responsibility upon 2 different shards, the ability for a node to process without overburdening the storage capacity roughly comes out about even.

The original version of this design included making several of these chains entirely executable, like the main ethereum chain. This would mean that say you have 5 shards operating in parallel, fully executing transactions and communicating with each other, meaning a huge jump in speed. This is because many computationally-strenuous contracts can be processed on a shard without bogging down the others.

Once again, as a holder of Ether, you will NOT have to do anything to prepare for this update. If you are running a node or a developer, consult the documentation for your node software for instruction.

The exact schematics of how these shards will operate in terms of execution vs. simple data storage is constantly evolving alongside other technologies and updates. Sharding is currently planned as “phase 2” of eth2, but it is entirely possible that with other scalability solutions, it will become unnecesarry.

Rollup-Centric Roadmap for Sharding

The original road map for introducing sharding was focused on layer-1 scalability. This means that changes to the blockchain, to increase transactions-per-second, were focused on the main-layer of Ethereum (how to increase transactions per block per second). It was previously thought that layer-1 solutions alone could bring sufficient to enough scalability to Ethereum. Recent spikes in transaction fees and congestion, alongside new long-term projections, have thrown that into question. It is no longer believed that layer-1 alone is sufficient to scale Ethereum to hundreds of thousands of transactions-per-second (TPS).

It was the decision of the community therefore, that the long-term road-map for improving Ethereum needs to work alongside layer-2 solutions (rollups). This has led to a phrase coined by Vitalik for the future upgrades for Ethereum

A Rollup-centric ethereum roadmap

This phrase means that future updates for Ethereum are intentioned to work with L2 rollups, to enhance their effectiveness. This is opposed to ignoring, or fighting against them. The way that sharding and future updates are designed are meant to make rollups even faster and cheaper, for the good of the community. This is where shards come into play, as an example.

The short-term immediate vision for sharding is to be data-shards, not execution shards. This means that a variety of shards will only hold data, posted on-chain by rollups. Other shards will hold execution information, blocks, account balances, etc. Try to follow me here, cause it gets a little complicated.

A rollup is just a block of raw data, stored on-chain, with transaction info. In a world without rollups, there is still a block-size limit. Only so much data can be put into a block before it runs out of space. Once rollups get larger, from more transaction volume, we end up back at square one. Rollups go back to fighting for block space and transaction fees go up. The solution, is to minimize the amount of data posted directly on chain, by moving it somewhere else.

Instead of posting the rollup itself to the chain, you could put it on a data-shard. The data shard then only needs to pass the identifying information to the execution shard, so that it can be recovered, and verified. The execution shard then, only needs to store the reference to the rollup, not the rollup itself. If you’re a developer, it works like a pointer.

Imagine a rollup containing 1000 transactions takes up about 55 kilobytes (I just made that number up i don’t know how much it actually is). You could store that on shard-45, which gets reported to the execution-shards. This shard gets a much smaller reference that says “the info you are looking for, that is verified to be legitimate, is on shard-45 and has the following identifier”. This reference information is obviously much smaller, containing something like only 5 kb of data. That makes the same amount of data 11x more efficient in transactions-per-second.

If you move all data to rollups, then every block becomes orders of magnitude more efficient, because the max amount of data that can be included, is.

But why do we need rollups? Why can’t we just use layer-1 if 64-shards means 64x more transactions?

We need rollups for a couple of reasons:

They are always going to be more efficient than using layer-1. Because you fit more transactions into the same space as 1 normal-L1-transaction, the transaction fee will always be lower. It’s simple math. A sole focus on layer-1 just doesn’t provide the same benefits as integration with layer-2.
Rollups are, at least in the short-term, the best solution for high transaction costs. They are already offloading billions of dollars in transaction-costs to L2 right now. Sharding is going to be a multi-year process that may change over time. A focus on supercharging them right now will make it significantly easier to adapt going forward.
Delegation of problem-solving. By offloading various responsibilities to rollup-teams, the Ethereum foundation and community can spend more time focusing on solving a variety of other necesarry problems, like security, post-quantum encryption, etc. This is a good long-term strategy to help Ethereum move faster into the future. It is also best for decentralization because it means no single entity is responsible for making Ethereum better, or ruining it.
Sharding is hard. Shards cannot talk to each-other, and parallel execution is very difficult. It’s easy to say “well just add more shards every time you want to increase TPS”. However, with every extra shard processing transactions, the organization between them gets tougher. All the shards need to agree on the current-state of the network. How shards interact and process in parallel is a mathematically, conceptually, and computationally, difficult undertaking. While it is hoped that someday all shards may be executable, focusing on roll-ups now will be the best way to reduce network-congestion.

The exact specifics of sharding are changing very rapidly with new research. It’s possible that my explanations may be slightly wrong, or that things have changed since I wrote this. Feel free to consult ethresear.ch for more cutting-edge info about what’s happening with ethereum 2.0.

The Future Post-2.0

After Eth2 is finished deploying, the future of Ethereum is uncertain. The foundation and the community will continue to develop updates to the network over time, as new issues arise. Vitalik himself has also put together an extensive roadmap of issues and solutions that Ethereum will continue to work towards in the years to come such as: Post-Quantum Encryption Updates, More Layer-2 Support, Client updates, etc.

Vitalik has created the following roadmap

Image Source: Vitalik Buterin Website, vitalik.ca

“Does Ethereum 2.0 mean the price of Ether is going to go up?”

I don’t know. It’s entirely possible. The last bull run began last year in anticipation of Eth2.0 Beacon Chain launch in December. I myself am very bullish on Ethereum for the foreseeable future, but I won’t make any predictions on price just yet. However, for the long-term health, sustainability, and scalability of the network needed to flourish, this is a very exciting time for all holders and developers.

If you’re still feeling a bit confused by all this, don’t worry. It’s a complicated subject and it took me a very long time to even understand it at a basic level. Keep working hard and feel free to check out this great video by Finematics.