Pantera Partners: Comparing the Advantages and Disadvantages of Arbitrum with Other Seed Contenders in the L2 Scaling Race

This article was published on ChainNews, author: Paul Veradittakit, partner at the blockchain investment firm Pantera Capital, translated by: Perry Wang.

At the beginning of 2019, Pantera led the seed round financing for Offchain Labs, which built the Arbitrum protocol, an Ethereum DApp scaling solution using Optimistic Rollups technology. It seems that Mark Cuban (owner of the NBA's Dallas Mavericks) has recently made additional investments!

Last Friday, the Arbitrum mainnet, which is open only to developers, went live, supported by the Uniswap community, which will "deploy Uniswap v3 on Arbitrum." You can read more about how Arbitrum works and how it compares to other Layer 2 (L2) solutions on Ethereum in the main text below.

• Due to Ethereum's design philosophy of prioritizing security and trust over efficiency, the massive adoption of Ethereum and the surge in transaction throughput have led to extremely low network efficiency. Crypto users on Ethereum DApps have therefore faced a poor user experience, often having to pay exorbitant transaction fees and dealing with extremely slow transaction times. This has resulted in a growing demand for L2 solutions built on Ethereum, which have found effective ways to package transaction data and periodically publish it to the Ethereum network, addressing many of the inefficiencies of the original Ethereum network.
• Offchain Labs recently announced the deployment of Arbitrum One, the mainnet beta of its L2 scaling technology Arbitrum, which uses the Optimistic Rollups mechanism.
• Arbitrum is built around a Rollup chain that executes all transactions submitted to the Arbitrum network and records its data in an inbox smart contract on Ethereum.
• To reflect these transactions on Ethereum, validators can aggregate the data of multiple transactions from the inbox into a concise summary called an "assertion," which is then published to the Ethereum mainnet. These assertions are optimistically assumed to be valid unless challenged by another validator within a set waiting period; incorrect assertions or challenges will incur penalties of forfeited funds.
• This mechanism effectively reduces the number of individual transactions that the Ethereum network must fully validate, allowing Arbitrum to provide significantly improved transaction speeds and lower transaction costs.
• Arbitrum is also designed to be very developer-friendly, fully compatible with the Ethereum Virtual Machine (EVM), and natively supports third-party tools such as Etherscan, Hardhat, and Truffle. These advantages, along with Arbitrum's significant scalability improvements, make the protocol a popular seed candidate for attracting developers building high-throughput DApps.
• The two competing technologies supporting L2 protocols are sidechains and ZK Rollups.
• (The sidechain used by Polygon) is essentially a completely independent blockchain that periodically "syncs state" with Ethereum by publishing block headers to Ethereum. Since these sidechains only receive a small fraction of Ethereum's transaction volume, they can significantly improve cost efficiency and speed. However, because sidechains operate completely independently of the underlying blockchain, they are vulnerable to various attacks, affecting developers' confidence in building on sidechains.
• (The ZK Rollups used by StarkWare) are similar to Optimistic Rollup technology: they execute transactions off-chain (on a Rollup chain), record all transaction data, and periodically publish assertions aggregating multiple transactions to the Ethereum mainnet. However, unlike Optimistic Rollups, they do not always optimistically assume that assertions are valid. Validators must also publish accompanying validity proofs, known as ZK-STARKs. Unfortunately, ZK-STARK technology is still immature, difficult to generalize, and often has very slow computational builds. The excellent security guarantees of this technology make it a long-term candidate for Ethereum scaling solutions, but in its current state, it remains difficult to use and handle.
• Optimistic Rollups are somewhat in between sidechains and ZK Rollups, retaining many of the security guarantees of ZK-STARKs through well-developed, usable infrastructure.
• Within a week of the Arbitrum developer mainnet beta launch, over 250 projects have landed, clearly demonstrating the insane demand for Ethereum scaling infrastructure. During this beta period, the Arbitrum team will retain the ability to pause or upgrade the network if necessary to eliminate any last-minute bugs or inefficiencies. Once all projects can build and test their applications on Arbitrum, the protocol plans to open to end-users and begin transitioning to full decentralization.
• Ultimately, as cryptocurrencies and DeFi continue to experience explosive growth, Arbitrum offers one of the most promising scaling solutions for developers looking to build powerful, user-friendly, and efficient DApps on Ethereum.

What is Arbitrum?

In the past week, Offchain Labs has deployed its Ethereum scaling solution, the Arbitrum protocol, to the Ethereum mainnet. As an L2 protocol, Arbitrum uses Optimistic Rollups technology to efficiently package smart contract execution data onto Ethereum, significantly improving transaction speeds and reducing transaction costs.

What is L2?

Ethereum is undoubtedly one of the most popular DApp blockchains today, with various DeFi protocols on Ethereum locking nearly 65 billion dollars in assets, and over 100,000 active users using Ethereum-based DApps daily. The Ethereum network focuses on optimizing verifiability and security, but faces a massive influx of transactions (nearly 1.5 million daily), leading to extremely slow transaction speeds and exorbitant Gas fees. For context, Ethereum users currently expect their transactions to take about 4 minutes to settle, with an average Gas fee of 1 dollar. For larger transactions, such as Uniswap trades, Gas fees can approach 8-9 dollars.

Ethereum Gas fee levels over the past year measured in Gwei, source: YCharts

To better handle the high throughput of DApps and DeFi protocols, several projects have launched L2 solutions built on Ethereum, referred to as the "first layer" L1 blockchain. L2 protocols find an effective way to package transaction data and periodically publish it to the Ethereum network at a high level. When transaction volumes are high, verifying each transaction directly on Ethereum can lead to sky-high fees and snail-like efficiency; L2 can achieve much greater performance improvements than native Ethereum because they effectively reduce the computational load of each transaction on the blockchain network.

How Does Arbitrum Improve Ethereum's Efficiency?

The key to the Arbitrum L2 solution is Optimistic Rollup. In a Rollup, transaction validators bundle multiple transactions together by executing them off-chain, then publish the transaction data and the concise account of the final state of the smart contract (called an "assertion") to the L1 blockchain.

Arbitrum specifically uses Optimistic Rollups, where assertions published to Ethereum are optimistically "assumed" to be true and correct unless challenged by other validators. Validators only publish assertions—without needing accompanying validity proofs—to the Ethereum network. To incentivize honest validation, validators must also stake collateral on Ethereum for a specific period. During this window, if another validator disputes the original validator's assertion, they can formally "challenge" the assertion on Ethereum and stake their own collateral. If the original validator's assertion is incorrect, they will lose their collateral. If the challenger is incorrect, they will lose their own collateral. If an assertion passes the designated time window without any successful challenges, it is "accepted" as true and merged with the final state of the Ethereum blockchain.

Assertion validation and challenges on Arbitrum, source: Arbitrum Whitepaper

By aggregating multiple transactions into a single interaction with Ethereum (publishing a single assertion), Arbitrum significantly reduces the total number of individual transactions that Ethereum must handle. All these transactions are technically still "published" to Ethereum (via assertions), but the network no longer needs to independently validate each transaction (and fully execute every step of each transaction on every node). Additionally, because assertions are designed to be compact, they typically occupy less space on the Ethereum blockchain, helping them to be packed into Ethereum blocks more quickly. Overall, these efficiency improvements lead to significant increases in transaction speed and cost, making it easier for DApps to handle large numbers of users and transactions.

From an architectural perspective, transactions on Arbitrum are submitted to the Arbitrum Rollup chain, which can be thought of as a sidechain for processing transaction execution off the Ethereum chain. Data from these transactions is recorded and then published to an open inbox smart contract on Ethereum. Validators (who can be anyone, as Arbitrum is completely permissionless) then aggregate the transaction data from the inbox into assertions and publish them to Ethereum. External users can restore the current state of the Arbitrum Rollup chain by simply executing transactions published to the inbox.

Arbitrum is also highly optimized for developer difficulty, including full EVM compatibility, integration with Etherscan, and native support for popular tools like Hardhat and Truffle. These features, combined with Arbitrum's significant scalability improvements, make the protocol a popular seed candidate for attracting developers building high-throughput DApps.

How Does Arbitrum Compare to Other Protocols in the L2 Space?

Given Ethereum's widespread popularity, several projects have launched L2 solutions to enhance the blockchain's scalability. In addition to Optimistic Rollup technology, two other high-profile technologies for L2 solutions are zk-Rollups and sidechains.

Polygon (formerly known as Matic Network) uses sidechains that are essentially completely independent blockchains for executing transactions, occasionally interacting with the L1 blockchain. Since sidechains only handle a small fraction of the total transaction volume of the L1 network, the likelihood of being "blocked" by high throughput is greatly reduced, allowing for lower transaction fees and faster transaction speeds. To ensure that the sidechain and the L1 blockchain are consistent, sidechains periodically publish their block headers to the L1 chain, ensuring that the "state" of both chains remains aligned. Sidechains improve efficiency by executing transactions outside of Ethereum, reducing the overall computational load on Ethereum. An unfortunate consequence of this is that sidechains often cannot provide the same security guarantees as the L1 blockchain, as they fundamentally require the L1 chain not to independently verify each transaction/block. Invalid state transition attacks become possible, where a majority of sidechain validators collude to artificially generate a block, steal funds from transaction participants, and publish the accompanying block header to Ethereum to complete the theft. Optimistic Rollups prevent such attacks by recording all transaction data in Ethereum's inbox. Anyone can execute transactions in the inbox to confirm the state of the Rollup chain and validate or challenge the validators' assertions. The main difference between sidechains and Optimistic Rollups is that sidechains publish snapshots of the blockchain's history to Ethereum, while Optimistic Rollups publish the entire blockchain history to Ethereum.

The ZK-Rollups technology used by StarkEx is architecturally similar to Optimistic Rollups. Both execute transactions on separate Rollup chains and periodically use assertions to record all data on Ethereum to verify Rollup transactions. The key difference is that in zk-Rollups, validators must publish validity proofs along with their assertions to Ethereum; unlike Optimistic Rollups, zk-Rollups do not optimistically assume that these assertions are valid. The prefix ZK comes from the fact that the accompanying proof is a ZK-STARK, or succinct transparent zero-knowledge proof. The specific technical details of ZK-STARKs are beyond the scope of this article, but simply put, ZK-STARKs are a cryptographic primitive that allows users to prove they "know" certain facts without including any revealing information about those facts in the proof itself. The main advantage of ZK-STARKs in scaling is that they are very compact, greatly enhancing the efficiency with which validators can publish assertions and related proofs to Ethereum. These proofs also ensure that the state of the Rollup chain is always synchronized with Ethereum; since all assertions must be provably valid, unlike Optimistic Rollups, there is no "challenge period" during which invalid assertions could be stored on Ethereum.

Despite their compact structure and strong security guarantees, the technology and abstractions for effectively building and computing ZK-STARKs are still quite underdeveloped, making this technology difficult to use and often highly use-case specific. Most existing implementations of ZK Rollups can only handle specific types of transactions, such as token transfers or atomic swaps. Additionally, their build times can be very slow; for example, building a verification architecture for a block with about 1000 transactions may take nearly 20 minutes on non-specialized hardware. As more specialized hardware (such as GPU verifiers) and cryptographic abstractions begin to emerge that cater specifically to ZK-STARKs, ZK Rollups may prove to be one of the most successful L2 scaling mechanisms. Until then, this technology remains quite limited in efficiency improvements and usability, making Optimistic Rollups a more promising solution in the short term.

What Are the Next Steps for Arbitrum?

The Arbitrum mainnet launched last week is specifically named "Arbitrum One" to clearly designate it as the flagship blockchain powered by Arbitrum technology. As the Arbitrum ecosystem continues to grow and mature, other Rollup chains (also using Arbitrum technology but possibly modified) may emerge, tailored for various use cases.

Arbitrum One is currently in the mainnet testing phase, and if necessary, the Arbitrum team will have the ability to pause and upgrade the blockchain. During this time, developers can experience the security and efficiency of Arbitrum and identify any last-minute vulnerabilities or bugs before the blockchain accepts massive throughput. Currently, Arbitrum One is open to all developers requesting access (with over 250 projects already), and once all projects can build and test their DApps on Arbitrum, access will be opened to end-users. The team's goal is to transition to full decentralization by the end of this summer.

A Few Thoughts

The recent surge of interest in the crypto space is a double-edged sword—bringing new users and sources of liquidity, but also leading to severe network congestion on Ethereum, long transaction times, and massive Gas fees. As Ethereum continues to gain traction now and in the future, building sustainable infrastructure that efficiently and securely powers higher-throughput DApps is more important than ever.

Arbitrum provides a powerful real-world example of using Optimistic Rollups as an L2 scaling mechanism, significantly improving the cost and time efficiency of Ethereum transactions while retaining several key security guarantees. A large number of projects (250 and counting) are already building on the Arbitrum developer mainnet, demonstrating the demand for this scalable infrastructure and the disappointment with current L2 players. As the Ethereum ecosystem continues to thrive, the Arbitrum L2 solution is giving rise to a new generation of highly scalable and efficient DApps, providing crypto users around the world with a significantly enhanced (cheaper and faster) experience.