At the ETH Shanghai Web 3.0 Developer Summit on May 20, 2022, Ethereum founder Vitalik said that the first major event after the merge of Ethereum could be the implementation of EIP-4844, which is known as Proto-danksharding, and that Proto-danksharding is another series of upgrades — Danksharding‘s first step in another series of upgrades. After the successful completion of the merger on September 15, subsequent upgrades, including EIP-4844, are finally on the agenda.
The Shanghai upgrade is the first and most important upgrade after the Ethereum merge and will include key items such as unlocking stake ETH, which is expected to be formalized possibly during the year 2023. On November 24, 2022, the Ethereum developers said in a community meeting that eight EIPs, including EIP-4844, were included in the consideration of the Shanghai upgrade. At the same time, Vitalik also tweeted again, reiterating that the implementation of EIP-4844 will be the key start of a massive reduction in Layer 2 fees for Ethereum, making it affordable for users to use on-chain applications directly, without having to rely on other CeFi intermediaries.
So, what exactly is EIP-4844, and what does it mean for the Ethereum ecosystem as a whole? This article will take you through these questions in depth.
The Merge completed in September 2022 only solved the energy consumption problem of Ethereum by migrating to a PoS mechanism, but it did not solve the problem of the insufficient processing capacity of the Ethereum network; what can truly provide higher processing performance for Ethereum is sharding, which will be deployed to the network later. After completing the Merge phase, Ethereum has now entered the Surge phase, which will significantly improve scalability by utilizing sharding technology.
source: Vitalik’s Twitter
To finally solve the contradictions caused by the blockchain impossibility trinity, Ethereum has planned an expansion plan centered on Sharding and Rollup technologies, after the expansion is completed, the transaction processing capacity of Ethereum will be greatly improved, and the current problem of high fees will be solved. Sharding technology is a Layer 1 scaling solution, while Beacon Chain will play the role of communicating with all major sharding chains. The Ethereum Sharding chain is isomorphic and is connected to various compositions Layer 2 in addition to the entire “Beacon Chain + Sharding Chain” structure.
In this future system, Beacon Chain is at the core, providing the Proof of Stake for the whole of Ethereum, forming the consensus layer, and synchronizing with other shard crosslinks. The beacon Chain acts as the heart of the network, coordinating the entire network through a 12-second Slot and an epoch of 32-time slots. In addition, Beacon Chain also provides uniform security for all major sharding chains by generating random numbers to assign verifiers to sharding chains. For each time slot, Beacon Chain randomly selects a node from the verifiers to become block proposers, and for the sharding chain, each time slot also generates a block proposer. For each time slot, Beacon Chain also distributes the verifiers evenly among the time slots, and then randomly selects at least 128 nodes from each group of time slot verifiers to become committees to witness the blocks and validate the Beacon Chain blocks and shard blocks.
Sharding technology is derived from database sharding technology, which currently refers to data sharding.
The Ethernet beacon chain is planned to first connect 64 fractional chains, which, conservatively estimated, will have 64 times the processing power of Ethernet 1.0, while in the initial planning, the number of fractional chains could be up to 1024.
Among them, each big Sharding chain is relatively independent, and Ethereum addresses will be divided into multiple classes according to their beginning letters. Each Sharding will handle transactions from the same class of addresses. Therefore, each node will only need to store a portion of all transactions on the Ethereum network and verify only that portion for which it is responsible.
Figure: Schematic diagram of Ethereum Sharding
If a node needs to use the data stored on other nodes, it can communicate and coordinate through Beacon Chain, mainly to get the Root Hash of other Sharding, but to avoid a large burden on Beacon Chain, this communication is carried out once every 6 minutes. At the same time, Beacon Chain assigns validators to the Sharding chain by generating random numbers, so that Beacon Chain provides uniform security for each large Sharding chain.
Sharding technology can significantly improve performance without sacrificing security.
Database Sharding is the current sharding technology used by Ethereum, in which each sharding chain only provides data capacity and is not responsible for processing transactions or running smart contracts. Layer 2 will handle the latter two, particularly the major Rollup projects.
A Rollup is the aggregation of a batch of transaction information into a transaction data calling, which is then periodically packaged into blocks to be submitted to each major sharding chain (Layer 1) to complete the record. According to the current plan, each sharding can provide approximately 250kB of dedicated data capacity for Rollup, and 64 shardings can achieve 16MB capacity. The more sharding will be used in the future, the more data capacity Rollup can use, and the lower the cost of operation will be.
Since the introduction of Ethereum’s Sharding scaling solution in 2016, related Sharding technology solutions have evolved, with one of the most notable changes being the shift from full execution Sharding to simpler data Sharding. Danksharding is a scaling solution previously promoted by the Ethereum development group, named after Dankrad Feist, a group member, and aims to balance decentralization and security.
Vitalik and others believe that in the future, if no single Rollup technology can occupy the entire market and multiple Rollups coexist, then there are complex trading and arbitrage opportunities for multiple Rollups due to the existence of cross-domain miner extractable value opportunities, and eventually, a small number of people seeking MEV will control the entire blockchain produces blocks.
As centralized block production is unavoidable, the only solution is to separate the roles of block builder and proposer, which is known as Proposer-Builder Separation (PBS) block builder, to obtain the maximum profit, sorts the transaction pool from high to low in terms of gas fee. The block builder, to maximize profit, orders the contents of the block builder blocks in the transaction pool from highest to lowest gas fee, and then bids to the proposer (i.e., validator), who can only choose the highest bidder to create the block. However, in the previous sharding design, each major sharding and Beacon Chain was validated independently, which hampered the implementation of PBS.
Danksharding redesigned its system around the cross-domain MEV problem in pursuit of countering the trend of centralization caused by MEV. In Danksharding, Beacon Chain will contain all blocks, and the data will be unified and authenticated by a committee. The sharding system is greatly simplified by this new design, Layer1 and Rollup can be synchronized directly with each other, the data capacity provided by sharding to Rollup for processing is increased, and the latency confirmation problem is avoided.
Danksharding is expected to be the “optimal solution” for scaling Ethereum, but there are numerous technical and commercial challenges to its actual implementation. Therefore, a stop-gap plan that can temporarily alleviate performance problems by going live earlier is crucial.
Vitalik Buterin, the founder of Ethereum, shared his answers to frequently asked questions about EIP-4844, or Proto-danksharding, on Twitter on March 19, 2022. Vitalik and others propose using Proto-danksharding as an early Ethereum Sharding design to increase data space for the Rollup project while paving the way for future data sharding. The proposal is expected to reduce the performance requirements of the Ethereum network for storage and memory, and potentially reduce all Rollup fees to one percent of today’s levels.
source: [email protected]
In essence, EIP-4844, which includes Proto-danksharding, is a temporary solution. Before the introduction of full Database Sharding, which increases the data availability space per block of the Ethereum Beacon Chain to 16MB, EIP-4844 can increase the available space per block to a maximum of 2MB. The prefix “Proto-“ denotes “original” and “primitive,” whereas Proto-danksharding denotes the “initial version” before Danksharding goes live. Danksharding may take several years to be formally deployed due to its technical difficulty, whereas Proto-danksharding, which can go live as soon as possible, can solve the urgent problem of insufficient scalability of the Ethereum network before the formal deployment of Sharding. Danksharding implementation is expected to boost Ethereum scalability to new heights.
EVM stands for Ethereum virtual machine, whose role is a runtime environment for all Ethereum accounts and smart contracts live on the Ethereum chain. and deploys valuable computing resources in the network through Gas Fee consumption. In EVM, Call data is an important storage resource, and the cost of using Call Data accounts for a significant portion of the costs incurred by Layer 2 transactions. After EIP-4844, Layer 2 data no longer needs to be stored in the expensive Call Date, but in the unique form of Blob data, which is called a blob-carrying transaction.
Blob-carrying transactions are a new form of transaction previously proposed by Vitalik, which can significantly reduce Call data consumption and Rollup consumption. Blob-carrying transactions cannot be accessed by the EVM execution layer, but can only be inspected, but users and validators can download the blob data. For this reason, EIP-4844 is also called Shard Blob Transactions, and Rollup will use the space in these blobs to store compressed transaction data in them.
It should be noted, however, that Proto-danksharding only completes the network logic and specification for future data sharding (i.e. danksharding), and even after the Proto-danksharding update, the Ethereum network will not connect to any actual running sharding. In Proto-danksharding, all validators and users must still directly verify the availability of the complete data. As a result, to reduce the burden on consensus nodes, Ethereum will set a storage time limit for Blob data (probably between 30 and 90 days), and this historical data may be stored on BitTorrent or IPFS in the future.
According to the website eip4844.com, there are six major advantages of EIP-4844: Rollup-friendly, lower gas fee, forward compatibility, existing in beacon nodes, reduced hard disk usage, and shorter Blob data storage time. It is expected that in 2023, EIP-4844 will be officially deployed to the ETH system after the Shanghai hard fork upgrade. After the upgrade is completed, Layer2 projects such as Arbitrum and Optimism are expected to benefit.
At that time, EIP-4844 will likely become another major node in the history of Ethereum development, while opening the way for full sharding in the future.