All you need to know about the Cancun upgrade and EIP-4884

What is the Cancun Upgrade?

The Cancun upgrade is a significant technical upgrade in Ethereum’s blockchain development roadmap. Similar to the prior Shanghai upgrade, it is named after the city where the Ethereum Developer Conference was held, which is Cancun in Mexico.

Each Ethereum technical upgrade incorporates several critical Ethereum Improvement Proposals (EIPs) aimed at better resolving the blockchain trilemma to improve performance and user experience. The Cancun upgrade primarily focuses on upgrading Ethereum’s execution layer, while the protocol layer upgrade is known as Deneb. Developers also combine the two names into “Dencun.” Thus, the Cancun upgrade is also referred to as the Dencun upgrade.

Ethereum, as the most active public chain in the Web3 world, has the highest number of blockchain applications and protocols. Each underlying technical upgrade has a significant impact, attracting close attention from industry experts.

What Does the Cancun Upgrade Mean for Ethereum?

According to the Ethereum core developer call records released on September 14, the current accepted proposals for this upgrade include:

1. EIP-4884: Shard blob transactions
2. EIP-1153: Transient storage opcodes
3. EIP-4788: Beacon block roots in EVM
4. EIP-5656: MCOPY - Memory copy instruction
5. EIP-6780: Selfdestruct only in the same transaction
6. EIP-7516: BLOBBASEFEE opcode*
7. EIP-7044: Lock voluntary exit signature domain on Capella
8. EIP-7045: Increase max attestation inclusion slot
9. EIP-7514: Add max epoch churn limits*

The * indicates the newly added EIP for this meeting. The last three EIPs are protocol-level upgrades for Deneb.

Although the terminology may seem technical, these EIPs mainly achieve two things:

1. Optimize the staking technology brought by the Shanghai upgrade and improve the staking infrastructure
2. Reduce Gas Fees and improve the blockchain’s scalability, providing more support for rollup technology in Ethereum

After the Shanghai upgrade, the current focus of Ethereum’s development is on promoting scalability centered around Rollup. Therefore, in this upgrade, the proposal EIP-4844, which can promote the development of Rollup and reduce Gas Fees, has become the core of this upgrade.

When Will the Upgrade Take Place?

According to the latest developer meeting, the launch of the Dencun testnet Devnet-9 has been postponed by one week to Tuesday, September 26. There is also a suggestion for opening Devnet-10 for short-term testing. The end of the developer-oriented testnet is primarily determined by network stress test results. If developers are satisfied, it will move to the public testnet; otherwise, Devnet-10 will launch. The public testnet will include networks like Goerli and Sepolia. The official upgrade on the mainnet will only occur if the public testnet runs successfully.

Although the exact upgrade time is uncertain, most industry experts estimate that it will be completed between Q3 2023 to Q1 2024.

What is EIP-4844?

EIP-4844 (Proto-Danksharding) is a solution proposed by the Ethereum Foundation to reduce gas fees and increase throughput. The discussion surrounding EIP-4844 predates the Shanghai upgrade, but developers postponed its implementation to ensure better upgrade effects.

In the short and medium term, Rollup is likely the most reliable scalability solution for Ethereum. Transaction fees on Layer 1 (L1) have always been high, so efforts must be made to shift the entire ecosystem towards Rollups. Rollups can significantly lower Ethereum user costs: Optimism and Arbitrum, as well as other Layer2 solutions, often have fees approximately 3-8 times lower than the Ethereum mainnet. On the other hand, ZK Rollups have superior data compression capabilities and can avoid including signatures, resulting in fees about 40-100 times lower than the mainnet.

However, despite these advancements, costs remain too high for many users. The long-term solution to address Rollup’s inherent limitations has always been data sharding (Danksharding), but its full implementation and deployment will take significant time. Therefore, EIP-4844 [1] was chosen as a stop-gap solution.

What is Danksharding？

Danksharding is a new sharding design proposed for Ethereum, introduced by Dankrad at the end of 2021. Prior to this, the scalability solution being discussed was Sharding 1.0, which can be simply understood as grouping network validators to parallelize the computation of various chain transactions. This means that a single blockchain comprises multiple parallel “shard chains.” This is essentially a form of parallel computing or divide and conquer approach. The Ethereum Beacon Chain plan aims to initially connect 64 shard chains, with a processing capacity estimated to be 64 times that of Ethereum 1.0. In the initial plan, the number of shard chains could reach 1024. The challenge with this technology is that the network needs to frequently synchronize the state and data of each shard chain, which is not only technically complex but also places high demands on the nodes, requiring all nodes to complete the synchronization within a specific period. This process can introduce network latency and potential network security issues.

Dankrad later introduced a new sharding scheme that satisfies three features: centralized block production, decentralized verification, and censorship resistance. The scheme has three core innovations:

1. Data Availability Sampling (DAS): In blockchain, block producers (miners/validators) are required to disclose and provide the transaction data of the blocks they produce for full node validation. This process is referred to as Data Availability (DA). Danksharding utilizes a mathematical design that allows validation nodes to only check a portion of data fragments to obtain data availability. This greatly improves the performance of the nodes.
2. Proposer-Builder Seperation (PBS): Currently, Ethereum validators create and broadcast blocks. They bundle together transactions that they have heard about through the gossip network and package them into a block that is sent out to peers on the Ethereum network. PBS splits these tasks across multiple validators. Block builders become responsible for creating blocks and offering them to the block proposer in each slot. The block proposer cannot see the contents of the block, they simply choose the most profitable one, paying a fee to the block builder before sending the block to its peers.
3. Censorship-Resistance List (crList): Block producers must specify a list of all eligible transactions they see in the mempool; packers must prove they have seen this list and are required to include these transactions, preventing the intentional omission of certain legal transactions.

The plan, finally named after Dankrad as Danksharding, is technically challenging to implement and thus needs to be rolled out in phases. EIP-4844 is mainly used to implement most logic and “scaffolding ”(such as transaction format and validation rules) required for a complete Danksharding specification.

In blockchain, transactions are typically packaged and recorded by blocks. However, the new transaction type introduced by EIP-4844, called Blob, differs from the blocks visible to the Ethereum Virtual Machine (EVM). Blob is only available for a short period of time and is not visible to the EVM. Blobs, which occur every 12 seconds, can store 1MB of data. This significantly increases the storage capacity compared to the average block size of Ethereum, which is about 90 KB, allowing for more transactions to be accommodated. Additionally, Blobs exist on the Ethereum consensus layer rather than the computation-intensive execution layer. Because Blobs are not visible to the EVM and do not belong to the execution layer, their cost is extremely low, greatly reducing transaction fees. [2]

So why do many people say it has a significant impact on Layer2?

This is because Layer2 operates based on the Rollup technology, which essentially executes a bundle of transactions outside the Ethereum mainnet. After execution, the execution results and transaction data are compressed and sent back to L1 for others to verify the correctness of the transaction results. Obviously, if others cannot read the data, verification cannot be completed. Therefore, it is crucial for others to be able to access the original transaction data, which is also known as “data availability.”

However, due to the current architecture of Ethereum, the data transmitted from L2 to L1 is stored in the Calldata of the transactions. Calldata was originally designed as a parameter for smart contract function calls and is data that all nodes must synchronously download. If Calldata becomes bloated, it will cause a high load on Ethereum network nodes, making the cost of Calldata expensive. This is the main factor contributing to the current cost of L2. [3]

Blob addresses this issue by designing a separate data type for the data transmitted from L2, separating it from L1’s Calldata. This data type must only be accessible and downloadable by relevant parties within a certain time frame, without requiring full network synchronization.

Impact of EIP-4844

Lower Gas Fee

This reduces the transaction cost on the network, especially for Layer2 solutions.

Preparing for Sharding

The upgrade serves as a transition for future data sharding technology on Ethereum.

Strengthening Core Competitiveness

With more Layer1 solutions appearing, reducing Layer2 and mainnet transaction fees helps Ethereum maintain or gain more market share.

Enhanced User Experience

More transactions can be accommodated after the upgrade, and the fees are much lower.

Introduction of a Fee Market

Since Blob is different from previous storage methods, it will bring a new fee market independent of L1 Gas fees.

Projects with Potential Impact

Layer2: Arbitrum, Optimism, zkSync, etc.

As mentioned above, Layer2 has to pay high Gas Fees to store its data in the calldata field for verification. After the EIP-4844 upgrade, the costs can be significantly reduced, thereby increasing profits. Additionally, the further reduction in Layer2 fees will also promote on-chain activity on Layer2, contributing to its ecosystem’s prosperity.

Layer2 DeFi

Layer2-based DeFi projects like GMX and RDNT, which emphasize capital efficiency and transaction speed, can allow users to “do more with less money” after the Cancun upgrade, better preparing them to accommodate more users.

DA Track

As Blob data can only be saved for a short period, solving the issue of historical data retrieval may stimulate new services and start-ups, such as DA scaling solutions specifically designed for Layer2.

This foundational upgrade not only impacts the above tracks but will also affect NFTs, cross-chain bridges, and other areas.

Future Steps

According to the Ethereum roadmap presented by ETH Chinese, the Cancun upgrade is the core of The Surge, whose main goal is to promote rollup-centered scaling, achieving 100,000 TPS. It mainly consists of two phases:

1. Preliminary Rollup Scaling: EIP-4844 introduces a new transaction type to Ethereum that carries transient Blob data, reducing rollup overhead by 10-100 times. It will also employ preliminary OP Rollup fraud proofs and ZK-EVMs for initial scaling.
2. Complete Rollup Scaling: Building on the previous foundation, the focus will be optimizing data availability (DA), such as data availability sampling clients, P2P design, etc.

The Cancun upgrade is merely the first step in preliminary scaling, laying the groundwork for the second phase of complete scaling.

Ethereum Development Roadmap（Source: ETH）

After The Surge, there will be four more phases: The Scourge, The Verge, The Purge, and The Splurge.

Conclusion

As an important upgrade in the Surge phase, the Cancun upgrade will further optimize Layer2 experience and costs. This significantly enhances the competitiveness of Ethereum and EVM-based blockchains, strengthening Ethereum’s mainnet’s leading position. At the same time, because the costs for Layer2 have been further optimized, the profit margins in the Layer2 track could become larger, possibly leading to a new round of competition.