What is the relationship between Danksharding and Proto-Danksharding (EIP-4844)?
Alright, no problem. Many people get this confused, so I'll explain it to you in an easy-to-understand way.
What is the Relationship Between Danksharding and Proto-Danksharding (EIP-4844)?
Simply put, Proto-Danksharding (also known as EIP-4844) is the 'precursor' or 'first phase' of the full Danksharding implementation.
You can imagine their relationship like building a highway:
- The ultimate goal (full Danksharding): Constructing a superhighway with 64 lanes to completely eliminate traffic congestion.
- The first step (Proto-Danksharding / EIP-4844): Building a temporary, extremely low-cost 'data-specific lane' alongside the existing road, and simultaneously establishing all the necessary plans, interfaces, and toll booths (transaction formats) required for future expansion to 64 lanes.
Let's dive into the details below.
1. What's the Problem? – Why Ethereum Needs Scaling
We all know that the Ethereum mainnet (Layer 1) is expensive and congested, which led to the emergence of many Layer 2 solutions (e.g., Arbitrum, Optimism). These are like 'buses' diverting traffic from the main road, bundling many transactions together and then submitting them to the mainnet in one go to share costs.
However, the problem is that these L2 'buses' have to store their bundled transaction data in the mainnet's CALLDATA. While this is cheaper than directly transacting on the mainnet, it's still expensive and, crucially, it's permanent storage, occupying valuable mainnet space. Most of L2's costs are spent on this.
2. The Ultimate Dream: Full Danksharding
Danksharding is the ultimate scaling solution for Ethereum. Its vision is very ambitious:
- Goal: To significantly increase the amount of data Ethereum can handle, specifically for Layer 2s.
- Method: As per the analogy above, it involves a complete overhaul of Ethereum, introducing a massive, entirely new data space divided into 64 'shards,' equivalent to having 64 data lanes simultaneously. This will reduce L2 costs to a negligible level.
- Challenge: This undertaking is highly complex, requiring extensive modifications to Ethereum's underlying infrastructure, making a single-step implementation impossible.
3. The Pragmatic First Step: Proto-Danksharding (EIP-4844)
Since the ultimate goal is difficult to achieve in one go, the first step must be taken. EIP-4844 is that step, and it's particularly clever because it tackles the most challenging work first.
It introduces an entirely new concept: Blob (Binary Large Object).
- What is a Blob? You can think of it as an 'attached data packet.' EIP-4844 creates a new transaction type that can carry one or more Blobs.
- What are the Benefits of Blobs?
- Cheaper: Blob data has its own independent fee market. It doesn't compete with regular Ethereum transactions for resources, making its cost very low.
- Temporary: Blob data does not need to be permanently stored on Ethereum. It only needs to be retained on-chain for a short period (e.g., one month) to ensure L2 security during that time. Afterwards, it is deleted, which greatly reduces the burden on the mainnet. It's like 'read-once, delete-after-reading' data, significantly lowering storage costs.
Thus, the role of EIP-4844 is to first open 1 dedicated Blob lane on the path to the 64-lane superhighway. Although it's just one lane, it already significantly alleviates L2 cost pressures. More importantly, it lays down all the 'infrastructure' needed for future expansion to 64 lanes, such as:
- New transaction format (transactions carrying Blobs)
- Cryptographic tools for validating Blob data (KZG commitments)
- New fee market
Summary of Their Relationship
| Feature | Proto-Danksharding (EIP-4844) | Full Danksharding |
|---|---|---|
| Core Goal | Reduce data costs for L2s, prepare for the future | Greatly expand Ethereum's data availability |
| Implementation Method | Introduces new transaction types and temporary "Blob" data space | Expands Blob space to 64 shards |
| Data Capacity | Each block can carry a small number of Blobs (initial target ~0.5 MB) | Each block can carry a large number of Blobs (target ~32 MB) |
| Relationship with L2 | L2s can already move data from CALLDATA to cheaper Blobs, reducing costs by 10-100x | L2 costs will become extremely low, almost free |
| Current Status | Live with the Cancun upgrade | Future upgrade target, highly complex to implement |
So, you can understand it this way:
EIP-4844 (Proto-Danksharding) has already built the engine, chassis, steering wheel, and operating system for the 'supercar' that is Danksharding. Currently, it's running with just one seat, but it's already significantly boosting speed. For the future full Danksharding, the task will be to expand the body of the car and add the remaining 63 seats, a process that will be considerably easier.