Proto-Danksharding

Introduces a new transaction type that carries large data packets called blobs. Unlike calldata, which is stored permanently on-chain, blob data is only accessible for a short period (approximately 18 days) and is not processed by the Ethereum Virtual Machine (EVM). This separation of data availability from execution is the core innovation for cost reduction.

Lays the groundwork for Data Availability Sampling, a technique where light clients can cryptographically verify that data is available without downloading it all. In proto-danksharding, the infrastructure for blob storage and distribution is established, enabling nodes to sample small, random pieces of blob data to confirm its presence.

Creates a separate gas market for blob space, distinct from the existing gas for execution. This decouples the pricing of data availability from network congestion for standard transactions. Blob gas prices are managed by a dedicated fee mechanism (similar to EIP-1559) that targets a specific number of blobs per block, making costs more predictable for rollups.

Uses KZG polynomial commitments (a type of cryptographic proof) to commit to blob data. This allows for efficient verification that the data in a blob corresponds to the commitment published on-chain. KZG commitments enable the trust-minimized and scalable data verification required for future Danksharding.

The primary practical outcome is a 10-100x reduction in data publishing costs for Layer 2 rollups (like Optimism and Arbitrum). By providing a dedicated, low-cost data availability layer, proto-danksharding directly lowers transaction fees for end-users on these scaling solutions.

Serves as a stepping stone to full Danksharding. It implements many core components (blobs, KZG, separate gas) in a simplified, single-blob-per-block format. This allows the network to test the technology and economic mechanisms before scaling to 64+ blobs per block in the final sharded design.

The core mechanism is a new transaction type that carries large data blobs. These blobs are temporary, high-capacity data packets (up to ~128 KB each) that are not accessible to the Ethereum Virtual Machine (EVM) but are fully available to Layer 2 sequencers. This separation of execution and data availability is key to reducing gas costs for rollups.

By posting data to blobs instead of calldata, Layer 2s like Optimism, Arbitrum, and zkSync experience a dramatic reduction in data posting fees. Blob space is priced via a separate fee market, decoupling it from mainnet execution congestion. This directly translates to lower transaction fees for end-users on these scaling solutions.

A dedicated EIP-1559-style fee market for blob space manages supply and demand. It uses a base fee that adjusts per block and a priority fee (tip). Excess blob fees are burned, similar to base fee burn. This creates predictable pricing for rollups independent of regular transaction gas wars.

Blob data is not stored permanently on the Beacon Chain. Nodes are only required to store blobs for ~18 days (4096 Epochs), after which they can be pruned. Long-term data availability is ensured by rollups, data availability committees, or third-party services, maintaining the chain's scalability.

Blob data is committed to using KZG polynomial commitments. Each blob is accompanied by a KZG commitment and a proof, which allows nodes to verify the data's availability and correctness without downloading the entire blob. This cryptographic primitive is essential for the security and efficiency of the scheme.

Proto-Danksharding establishes the core architecture for full Danksharding. It introduces the blob transaction format, the fee market, and the principle of data availability sampling. Future upgrades will scale the system by increasing the number of blobs per block and implementing peer-to-peer sampling across the network.

The core mechanism is a new transaction type that carries large data packets called blobs. Unlike calldata, blob data is not accessible to the EVM and is not permanently stored on the Beacon Chain. It is held by consensus nodes for a short blob retention period (currently ~18 days) before being pruned, dramatically reducing the long-term storage burden and cost.

Each blob is accompanied by a KZG commitment, a cryptographic proof that commits to the blob's data. This allows nodes to verify that the data is available without downloading the entire blob. The system's security relies on data availability sampling (DAS), where light clients can probabilistically verify that blob data is published and retrievable.

A separate blob gas fee market is created, distinct from the standard execution gas market. This decouples the pricing of data availability from computation. Blob gas prices are determined by a dedicated target and limit system, targeting 3 blobs per block with a maximum of 6, allowing the fee to adjust based on demand for blob space.

This is a stepping stone. Key differences:

• Scale: EIP-4844 provides ~0.375 MB per block. Full Danksharding aims for ~16-32 MB.
• Architecture: EIP-4844 uses a single-dimensional fee market. Full Danksharding will implement proposer-builder separation (PBS) and a multi-dimensional market.
• Nodes: EIP-4844 requires all consensus nodes to store blobs temporarily. Full Danksharding distributes data via data availability sampling across a committee.

The primary beneficiary. Rollups like Optimism, Arbitrum, zkSync, and StarkNet post their transaction data (or proofs) as blobs instead of calldata. This reduces their data publishing costs by an estimated 10-100x, directly lowering transaction fees for end-users on these L2s.

Requires coordinated upgrades across the Ethereum stack:

• Consensus Layer (CL): Manages the blob sidecar, validates KZG commitments, and enforces the blob retention policy.
• Execution Layer (EL): Processes the new transaction type, handles the blob gas market, and passes commitments to the CL.
• Engine API: Extended with new methods like engine_newPayloadV3 to propagate blobs.

A new transaction type introduced by EIP-4844 that carries large data packets called blobs. These blobs are stored temporarily by consensus nodes but are not accessible to the EVM, separating data availability from execution. This design allows rollups to post data cheaply without permanently bloating Ethereum's execution layer state.

A critical technique for full Danksharding where light clients can verify data availability by randomly sampling small pieces of a blob. It ensures that the data for a shard block is published and accessible without any single node needing to download the entire block, enabling secure scaling.

Cryptographic commitments (using Kate-Zaverucha-Goldberg polynomial commitments) that act as a short proof for blob data. In Proto-Danksharding, each blob has a KZG commitment stored on-chain, allowing nodes to verify the correctness and availability of the blob data without storing it directly.

The end-state scaling vision for which Proto-Danksharding is a precursor. It aims to scale Ethereum's data layer to ~1.3 MB per slot across 64 data shards, enabled by Data Availability Sampling (DAS) and a distributed validator set. This would allow for massive throughput for rollups.

The official Ethereum Improvement Proposal number for Proto-Danksharding. It details the core specifications, including the new transaction format, blob validation rules, fee market for blobs, and the 4096-field-element blob size (~128 KB).

A separate fee market for blob-carrying transactions, distinct from standard EIP-1559 execution gas. Its price is adjusted based on demand for blob space, protecting block space for regular transactions and creating a predictable, low-cost environment for rollup data.