In a recent blog post named “Possible futures for the Ethereum protocol, part 2: The Surge,” Vitalik Buterin, co-founder of Ethereum, presented an ambitious plan to boost Ethereum’s transaction throughput to over 100,000 transactions per second (TPS) on both Layer 1 (L1) and Layer 2 (L2) solutions. This initiative, referred to as “The Surge,” aims to improve scalability while maintaining decentralization and security.
Buterin began by revisiting Ethereum’s early scaling approaches, which included sharding and Layer 2 protocols like state channels and Plasma. He mentioned that Ethereum’s original roadmap from 2015 included two strategies for scaling, specifically referring to sharding—which allows each node to validate a subset of transactions—similarly to how peer-to-peer protocols like BitTorrent function.
At the same time, Layer 2 solutions were being developed to alleviate computational and data demands from the main chain while benefiting from Ethereum’s security framework. Rollups emerged in 2019 as a vital Layer 2 enhancement, requiring a considerable amount of data bandwidth. Buterin noted, “By 2019, sharding research resolved the challenge of verifying ‘data availability’ on a large scale, leading to a convergence of strategies into what is now the rollup-centric approach that continues to guide Ethereum’s scaling efforts.”
Ethereum Roadmap: The Surge
The Surge has set several critical objectives, including exceeding 100,000 TPS on both L1 and L2, maintaining the decentralization and integrity of L1, ensuring some L2 solutions fully inherit Ethereum’s essential properties such as trustlessness, openness, and resilience against censorship, along with maximizing compatibility across L2s for a cohesive Ethereum experience.
A key method to realize these goals is through Data Availability Sampling (DAS). Currently, the data bandwidth for Ethereum’s L1 limits rollup TPS to approximately 174. To overcome this constraint, Ethereum intends to adopt PeerDAS, which employs one-dimensional sampling for efficient data availability verification.
Buterin mentioned, “Our medium-term objective is to reach 16 MB per slot, which combined with enhancements in rollup data compression would potentially yield around 58,000 TPS.” He also indicated that in the future, two-dimensional sampling might be adopted to improve efficiency, albeit at the cost of added complexity. “We need more research to find the optimal form of 2D DAS and ensure its safety,” he noted.
Data compression methods are vital for minimizing transaction data size. These methods involve utilizing BLS signatures for signature aggregation, substituting addresses with pointers to historical data, and creating custom serialization for transaction amounts. Buterin suggested that most currency values could be efficiently represented using a tailored decimal floating point format or a dictionary of frequently used values.
Generalized Plasma is another crucial element of The Surge. Plasma facilitates off-chain transactions while maintaining on-chain security guarantees. By integrating SNARKs (Succinct Non-interactive Arguments of Knowledge), Plasma can become even stronger and more versatile. “Even if you can only protect part of the assets […] this significantly enhances the current capabilities of ultra-scalable EVM, which is a validium,” he remarked.
Buterin highlighted the necessity of advancing L2 proof systems. Currently, many rollups do not operate fully trustlessly and depend on security councils that can override proof mechanisms. He emphasized the importance of achieving “Stage 2” rollups, which are entirely trustless and secure. This involves formal verification, a process that employs mathematical techniques to confirm that proof systems conform to EVM specifications.
“We can establish a formally verified SNARK prover for a minimal virtual machine,” he explained. Implementing diverse proof systems or “multi-provers” can also enhance redundancy and security. “If all proof systems reach a consensus, the security council cannot interfere,” Buterin pointed out.
Fostering cross-L2 interoperability remains a primary focus. A significant challenge is ensuring a seamless interaction for users across the L2 ecosystem. Buterin proposed improvements such as implementing chain-specific addresses that indicate the chain identifier for easier cross-L2 transactions, standardized payment requests for secure payment demands across various chains, and creating protocols like ERC-7683 and RIP-7755 to facilitate asset transfers and gas payments.
Additionally, he advocated for the integration of light clients and keystore wallets, which would empower users to verify chains independently of RPC providers and manage keys efficiently across multiple networks. “Our success in resolving this issue will be a testament to our ability to unite as a community,” Buterin asserted.
While optimizing L2 is crucial, enhancing L1 remains essential for Ethereum’s security and economic stability. Buterin discussed strategies, including raising the gas limit, reducing the cost of certain operations through proposals like EOF (EVM Object Format), and investigating native rollups. “An important question for any L1 scaling strategy is determining the ultimate vision for what should remain on L1 and what should transition to L2,” he remarked, underscoring the necessity for a balance that preserves Ethereum’s foundational strengths.
In conclusion, Buterin stated, “Our current mission is to finalize the rollup-centric roadmap and tackle these challenges while ensuring the resilience and decentralization that make Ethereum L1 unique.”
As of now, ETH is priced at $2,625.