Recently, the Ethereum network has become increasingly congested as traffic and users on the network have been increasing, leading to long wait times and higher fees. Layer 2 solutions are being developed to relieve these issues, but which one is the best?
One of the main reasons Ethereum has not already achieved its goal of being the “World Computer” is because of its scaling issues. Currently, Ethereum can handle about 20 transactions per second or TPS, and the fees vary based on how fast a user wants a transaction to go through. The median transaction fee hit a record high above five dollars recently, which signifies how bloated the Ethereum network has become due to the high volume of DeFi apps, such as exchanges and yield farming. Though some may see this increased volume of activity on Ethereum as a positive, it also poses a high barrier to entry for cryptocurrency users attempting to interact with smart contracts who have a low account balance.
Cardano, EOS, TRON
Instead of using Ethereum, these users may use another smart contract platform, such as Cardano, EOS, or TRON, which boasts much higher transaction speeds at a much lower cost. To the average user, these blockchains seem like the obvious choice, as they appear to offer faster transactions with no downside. In reality, these blockchains offer quicker transaction times in exchange for a more centralized structure. For example, EOS only has 100 block producers, meaning each has control over 1% of the network, whereas Etheruem has no limit to the number of miners which can secure the blockchain. In order to mitigate these problems, Ethereum 2.0 is planning on having sharding capabilities, which will allow the network to scale to over 100,000 TPS on-chain according to Ethereum creator Vitalik Buterin.
However, the launch date of Phase 1 of Ethereum 2.0 is not set for a couple of years at the minimum, and Ethereum needs a scaling solution as soon as possible. This is where Layer 2 solutions come into play in order to make Ethereum a competitive smart contract development platform today.
What Are Layer 2 Solutions?
Before we can understand what Layer 2 scaling solutions are, we need to understand what Layer 1 is and work our way up. Layer 1 is the main blockchain, such as Ethereum, that uses itself for security. For example, Ethereum and Bitcoin have miners providing security for the network through proof of work. Layer 2 solutions are built on top of these Layer 1 protocols and piggyback off of them for security. They achieve this relationship by interacting through the blockchains using smart contracts. This effectively allows the Layer 2 solutions, assuming there are not any bugs, to be as secure as Layer 1. In the case that there is a bad actor on Layer 2, their actions cannot be immediately noticed, as there is not a massive amount of hash power securing these networks. Instead, when a bad actor is found through software that monitors the Layer 2 network, the data is sent to the Layer 1 protocol, which acts accordingly to punish the bad actor and reward the funds to the correct user.
If a user wants to withdraw their funds from a Layer 2 solution, they must wait through a specified window. The length depends on the protocol, from seconds to days to weeks, in order to give software time to check for any bad actors relating to your transactions.
What Layer 2 Solutions Exist?
In July 2020, Reddit collaborated with the Ethereum Foundation in order to launch “The Great Reddit Scaling Bake-Off.” This challenge was meant to find a scaling solution for Ethereum which would allow for Reddit’s Ethereum-based digital assets, Community Points, to be transferred quickly and cheaply. They asked each interesting project to provide documentation, cost estimates, and other crucial information to be considered a contender for the project. The actual requirements included: 100,000 transfers over a five-day period, no single point of failure, a simple and free experience for users, interoperability with third-party wallets and apps, and full security. The contest, which is still ongoing, offers no monetary reward for winning the Bake-Off, but the brand recognition and association with one of the internet’s largest sites is a huge benefit for the winner of the competition.
Twenty-two unique projects entered the contest, but we will only be focusing on the largest and most viable solutions, which include: OMG Network, Loopring (note: did not enter the challenge, but a viable and popular scaling option), Matic Network, and Raiden Network. Each of these Layer 2 solutions offers a different approach to solving the scaling problem, but only one will be adopted by Reddit and claim the title of the scaling solution for a website with millions of monthly visitors. Let’s take a deep dive into each scaling solution and determine which one is the best by looking at metrics such as their team, technology, TPS, cost per transaction, roadmap, and progress to this point.
OMG Network, previously OmiseGo, is one of the oldest scaling projects for Ethereum. OmiseGo was originally created by Thailand’s payment company Omise. In 2017, Omise conducted an ICO and got interested from Vitalik Buterin, as they were planning to implement the Plasma scaling solution he had recently co-authored a paper about. Vitalik served on the board of advisors, and it is unknown if he still does. In their ICO, Omise raised $25 million for 65% of the total supply of their OMG token. A generous 5% of the tokens were airdropped to ETH holders, and the rest of the 35% was reserved for the team.
Plasma, the Layer 2 scaling framework of the OMG Network, was envisioned by Vitalik Buterin and Joseph Poon in 2017. This framework is not specific to OMG but is instead can be implemented in many projects. The basic idea of Plasma is that it is faster to do transactions off the main Layer 1 blockchain and instead of doing them on a sidechain.
The Plasma Whitepaper Header
The Plasma MVP, or Minimal Viable Plasma, allows for fast transactions but not complicated smart contract execution. This is the simplest implementation of Plasma and is similar to what the OMG Network runs. It relies on an operator, which is a single party that runs the plasma chain and creates blocks. Even though this seems insecure, the design ensures that the operator cannot act maliciously. When a user sends a transaction, they must also include two confirmation signatures, which ensures the legitimacy of each transaction.
Unfortunately, this is an unpleasant process, so the OMG Network modified the MVP and created More Viable Plasma or MoreVP. This version does not require confirmation signatures and implements extra security features to ensure the security of the network without requiring the confirmations. Eventually, OMG Network aims to move away from the operator and instead use a system of validators to secure the network.
Staking the OMG token
These validators would stake the OMG token and receive fees in exchange for powering transactions on the network. Their roadmap does not have a timeframe for this full proof of stake model, so it’s anyone’s guess as to when it will be released. However, they do have a live proof-of-concept for the Bake-Off which is capable of thousands of transactions per second at no cost.
One of the greatest successes of the OMG Network has been its integration with Tether and Bitfinex. Deposits and withdrawals using USDT can now take place on the OMG Network, which “supports thousands of transactions per second and reduces transaction costs to a third of Ethereum” claims Vansa Chatikavanji, CEO of OMG Network.
In summary, OMG Network is well on its way to being a prominent scaling solution for Ethereum and already has the support of Vitalik and the largest stablecoin ecosystem.
Loopring was created in 2017 by Daniel Wang and Jay Zhou as a high-speed decentralized exchange, or DEX. Wang, CEO of Loopring, has a Master’s Degree from Arizona State University in Computer Science and an MBA from the University of Minnesota. Zhou, COO, has no formal Computer Science education but has worked at Paypal and as a CS Project Mentor for Stanford University, where he gives blockchain advice to students. They held an ICO for loopring in August 2017, where they raised $45 million.
Initially, the project focused heavily on being a non-custodial decentralized exchange through price-matching and ring-sharing features. The Loopring LRC token’s initial purpose was as a way to receive fees from the protocol through a profit-sharing initiative. Another purpose of the LRC token is that it must be staked by DEXes who want to implement Loopring’s technology.
In December 2019, Loopring introduced their 3.0 update, which brought the important scaling technology called zero-knowledge rollups, or zk-Rollups. Zk-Rollups attempt to solve the problems of Plasma, mainly the long waiting period to withdraw from the Plasma network and the requirement to watch the network for malicious actors. Instead of creating one transaction per transfer, like Plasma, this protocol “Rolls up” hundreds of transfers into one transaction, creating a smaller and cheaper transaction.
Then, using the technically complicated zero-knowledge proofs, the transactions can be published to the Ethereum blockchain without requiring data about the transaction, hence the “zero-knowledge” aspect. There are two users in the zk-Rollup ecosystem: transactors and relayers.
Faster than Plasma!
The transactors are those who make transactions, which consist of a “to” address, a “from” address, the value transacted, a network fee, and a nonce (unique cryptographic number). These transfers are then sent to the relayers, who use lots of computing power to create the rollup and transfer it to the Ethereum mainnet. This system is faster than Plasma and does not require the same time-sensitive fraud-prevention methods. However, the relayers are required to use a lot of computing power to create the zero-knowledge proof.
In June 2020, Loopring announced Loopring Pay, which uses zk-Rollups to transfer Ethereum and ERC20 tokens between users “instantly, for free, and with the same 100% Ethereum security guarantees”. Loopring pays for their users’ transactions fees, which cost $0.00006 per transfer. The main caveat of this system is, like Plasma, it requires users to use the solution. So if a certain exchange or address does not have their address registered with Loopring and is not a member of the zk-Rollup, they cannot receive or send these payments. Fortunately, Loopring has help from Vitalik, who has been tweeting about Loopring and zk-rollups.
The Loopring smart contract currently has over $20 million in funds locked, which is somewhat significant but has lots of room to grow.
As Loopring looks toward the future, they hope to continue making developments to increase the effectiveness of their DEX protocol and adoption of their Loopring Pay platform.
Matic Network was created by Jayanti Kanani, Sandeep Nailwal, and Anurag Arjun as a Layer 2 scaling solution using the Plasma framework and Proof-of-Stake in order to effectively scale Ethereum while not sacrificing decentralization. Kanani, the CEO, has a Bachelor of Engineering in Information Technology from Dharmsinh Desai Institute of Technology in India, and previously worked as a data scientist at Housing.com. Nawal, COO, used to work for Deloitte as a consultant and has an MBA from NITIE. Arjun, CPO, has 11 years of experience in the technology sector with a focus on financial data. Together, this team has built a capable scaling solution for the Ethereum network.
In 2019, the team did an ICO for their MATIC token and raised $5.6 million for 32% of the total token supply. This ICO happened through the Binance launchpad, which offered immediate liquidity for the token on one of the world’s largest cryptocurrency exchanges. The MATIC token has an important use in the network in paying transaction fees and participating in the network through proof of stake.
Matic’s Layer 2 solution
Matic’s Layer 2 solution is a variant of the Plasma framework MoreVP, which is the same as OMG Network. Unlike OMG Network, the framework is only meant to guarantee the security and legitimacy of main chain assets, while other transactions are secured through a proof-of-stake network. This proof-of-stake network is built on top of Tendermint, the same solution as blockchain interoperability platform Cosmos. In order to achieve a high transaction volume, Matic uses a Block Producer layer, which quickly produces blocks that are then checked through the decentralized Proof of Stake validators. With this process, Matic has a theoretical maximum of 65,000 transactions per second, though this number is likely to be much lower in a real setting.
Matic describes their solution in four steps. First, a user deposits their cryptocurrency asset into the Matic contract on the main chain, Ethereum. Then, the tokens will be shown on the Matic chain, which is much faster and cheaper. Users can then transfer these tokens to any other address on the Matic network cheaply and instantly. Once they are ready to withdraw, they can do so through the same contract that they used to deposit their tokens. Stakers on the network verify transactions and use a checkpoint system to keep the main Ethereum network updated about the status of assets on the Matic Network, which requires a ⅔ majority of stakers. The stakers also choose block producers, who provide fast block generation (~1 second) and have a decently sized stake in the network. As another incentive to prevent fraud, the network also offers Fraud Proofs, which allow any user to claim that a transaction is fraudulent and if correct, funds of the malicious actors will be slashed and the claimer will receive those funds.
All of these details about the Matic Network focus on a one-sidechain design, but there could be an infinite number of side chains being watched by the checkpoints. A practical example would be a business that wants to transact all of their data on a sidechain, then update the Ethereum blockchain with valid data as-needed, all while knowing their transactions are valid and immutable.
Not only does Matic offer payment solutions, but also tools for Ethereum dApps to migrate to their blockchain, which would provide incredibly fast and cheap transactions. This is an innovative solution that the other two previously covered solutions do not offer. If Uniswap, for example, migrated to the Matic Network, fees would be pennies instead of $50 during times of high gas prices. Matic already has several dApps being built on their platform, such as Decentraland and Cryptostaw, a “Venmo using DAI”.
On June 29th, 2002, the recently-launched Matic mainnet released their staking feature, allowing MATIC holders to secure the network and make it a truly decentralized blockchain. In the future, Matic Network aims to add more features to their blockchain, bring on more dApps, and increase adoption.
Raiden Network was established by Brainbot Labs in 2017 as Ethereum’s version of Bitcoin’s Lightning Network. The founders held an ICO in 2017, raising $33 million for 50% of the tokens. As one of the older scaling projects for Ethereum, Raiden is an established, yet forgotten, name in the scaling space. Their method of choice is payment channels, which allow for two participants to transfer funds between themselves a theoretically unlimited number of times for free. For these payment channels to work, each address must agree to open a payment channel.
Then, users can deposit funds into this channel. Once the funds are in the channel, they can be sent in between the addresses with ease. An example of this would be if an individual opened a payment channel with their local coffee shop. Both parties would agree to the channel, and the consumer would put a set amount into the channel, like $20, which would still sit in his wallet in the channel. Each time the consumer makes a purchase, he would send the money in the channel to the coffee shop, and this transaction would be instant and free. The consumer can never spend more than is in the payment channel, though.
Once one of the parties wants to remove the money from the channel and put it back onto the blockchain, they can close the channel and the funds will be delivered to the correct addresses.
Even though this solution appears to only work between two parties, there is a workaround. There does not need to be a direct payment channel between a payer and a payee as long as there is a route between other channels that connects the two. This system makes the channels ideal for micropayments, but larger transactions may need to take place on the main blockchain if there are not enough funds in the channel.
Depiction of a Payment Channel Network
The only purpose of the RDN token sold in the ICO is as a currency compatible with the Raiden Network.
This solution is by far the most basic of the scaling solutions analyzed. Even though there is a niche use case for micropayments, it is doubtful that this solution will be the solution of choice to scale the entire Ethereum blockchain.
There are plenty of projects vying to be the Layer 2 scaling solution for Ethereum, but determining which will succeed is difficult. If the capability of the technology was the only factor, it may be an easy choice, but one has to also determine adoption as well. For example, if some random competitor offered the cheapest and fastest solution, it would mean nothing without anyone to transact with on the network. OMG Network, which has the first-mover advantage and momentum coming from their integration with Bitfinex and Tether, seems to be in an admirable position, but also can only scale transfers, still has fees involved, and has the drawbacks of the Plasma Framework.
Loopring, which offers faster and cheaper transactions through zkRollups, is mainly focusing on payments and their decentralized exchange features, and less scaling of the entire Etheruem network. Matic Network is the only option covered which is capable of scaling both payments and dApps, with dApps already building on their network and a working mainnet. However, this technical superiority does not equal an automatic victory for the platform, as the project is still in its infancy with a very recent mainnet launch and little real-world testing. Raiden Network, with its specific use-case and limitations, does not have the capability to be used by the entire network, but still could be adopted by some merchants and consumers as a quick and cheap option to make small transactions.
One thing is for certain: whichever Layer 2 solution gets adopted by Reddit as the champion of their Bake-Off will have a massive advantage and is on the path of becoming the scaling solution for Ethereum.