NEAR DA data analysis: Arbitrum would have saved $16.2M in 10 weeks
Thread1,117 words
Market since$NEAR−23%since this pieceHow it played out ↓Data case for NEAR DA: rollups spent 9,000 to 16,000 ETH monthly on DA, EIP-4844 helps only partly and temporarily, and NEAR's sharded Nightshade design would have cost an Arbitrum-stack rollup $87.62 versus over $16M on Ethereum in 10 weeks.
Key claims
Rollups spent 9,000-16,000 ETH per month on DA, about $42M at 12k ETH
NEAR DA offers 4 MB/s per shard across 4 shards
Pre-4844 calldata was 80-90% of rollup fees
Arbitrum would have saved USD 16.2M in 10 weeks if they were posting transaction data to @NEARProtocol instead of Ethereum. ✨
NEAR Protocol will be a major data layer for modular chains.
A data-driven research 🧵 on NEAR DA.
📊👇
1/ The modular thesis will reshape Web3.
Separating the core functions into distinct, interchangeable layers, it gives rise to a future built on scalable execution layers that are optimized at every level of the tech stack to serve specific use cases or verticals.
2/ While modularity allows for all sorts of network configurations, rollups are probably the most prominent modular network design.
In a nutshell, rollups batch transactions for off-chain processing, only periodically verifying their validity on-chain in aggregate.
3/ Depending on state validation mechanism and where the rollup posts state data, there are more granular distinctions.
From an Ethereum-centric perspective this leaves us with the terminology outlined in tweet 4.
For an overview over of the landscape, check the visual below.
4/
✦ Validity Rollups: Data & verify state on L1 (validity proofs)
✦ Optimistic Rollups: Data & verify state on L1 (fraud proofs)
✦ Validiums: Data off-chain & verify state on L1 (validity proofs)
✦ Optimiums: Data off-chain & verify state on L1 (fraud proofs)
5/ So how can @NEARProtocol serve these highly scalable, rollup-based scaling solutions?
✦ NEAR Fast Finality Layer: Verify state quicker & cheaper
✦ NEAR DA: Post data much more efficiently
NEAR is a complement to Ethereum. Not a competitor.
x.com/zerokn0wledge_/status/1767758763984109787
6/ But let's dive into the data. 📊🔍
We won't dive into the fast finality layer today. Instead, our data-driven analysis will take us down the data availability rabbit hole.
Ready, anon?
7/ The capacity of the base layer, to handle data that is published by the rollups built on it, is the main scalability bottleneck for rollup-based scaling solutions.
Pre-EIP4844, the cost of call data (used to publish data to L1), accounted for 80-90% of total rollup fees.
8/ In the past 5 months, rollups on Ethereum, have consistently spent a total of 9'000-16'000 ETH per month on data availability.
Assuming an average of 12k ETH per month, this amounts to $42M in DA cost.
9/ Why do rollups pay so much for this? 🤔
Rollups need their state data to be available, in order to compute and verify its state.
If data is not available, no one can verify the chain anymore and the rollup shuts down or is at least prone to attacks.
10/ Ethereum is one of the most secure and battle-tested networks out there.
However, this comes at a price. 💸
For many use cases, the before-mentioned Validium and Optimium designs are hence much more feasible.
11/ By posting data somewhere else, these network designs can significantly reduce transaction cost and increase throughput.
However, while centralized off-chain committees provided an alternative that was easy to implement quickly, they introduce significant trust assumptions.
12/ That's why decentralized, modular DA layers that are secured by globally distributed validator sets and crypto-economic security mechanisms are key.
They allow for the same scalability and cost improvements while maintaining security and decentralization.
13/ Wait, doesn't EIP-4844 solve the problem anyhow?
Well, it does partially. And temporarily.
The introduction of data blobs that remain outside of EVM execution and only stored for a few weeks (to avoid state bloat), definitely reduces the cost for rollups to post data.
14/ The increase in data space that is exclusively available to rollups separates fee markets for general execution on L1 & data publication by L2s.
However, it's only a one-off increase.
Over time, blob fees will increase as the number of rollups & the amount of data grow.
15/ Long-term and for highly cost-sensitive use cases we need more scalable DA layers that can work in conjunction with Ethereum.
The most scalable solution both in terms of cost and throughput out there right now is @NEARProtocol.
16/ What makes @NEARProtocol so scalable as a DA layer?
NEAR DA leverages an important part of NEAR’s consensus mechanism, known as Nightshade, which parallelizes the network into multiple shards (essentially, multiple parallel blockchains).
17/ Each shard on NEAR produces a small portion of a block, which is called a chunk.
These chunks are aggregated to produce blocks.
All of this happens entirely at the protocol level and so is invisible to users and developers.
18/ When a chunk producer processes a receipt, there is consensus around the receipt.
Once the chunk has been processed and included in the block, the receipt is no longer required for consensus and can be pruned from the blockchain’s state.
19/ Hence, NEAR doesn’t slow down its consensus with more data than it requires, yet any user of NEAR DA would have ample time to query transaction data.
20/ This means scalable and cost-effective data availability for any rollup, especially to those with high transaction volume.
One segment of chains that definitely requires this is specialized gaming networks. But even AI/ML use cases and beyond require scalable data.
21/ With the move towards stateless validation, @NEARProtocol will further decrease the hardware requirements of certain types of validators (chunk validators).
It moves the state into memory, which will allow for more shards and increase the decentralization in the system.
22/ This means that the number of shards will grow, increasing the overall throughput.
While NEAR is already fast, with 4 MB/s per shard with 4 shards, scaling towards n shards means that rollups or any other project using NEAR will not have to compete for blockspace.
23/ This is a major contrast to the scaling limits of both monolithic systems and makes NEAR DA future-proof.
As the NEAR Protocol starts increasing the number of shards to n, the amount of data that has to be stored for a single shard decreases.
24/ Already today, the cost savings that NEAR DA provides are substantial.
Arbitrum would have paid $87.62 instead of >$16M and we're only talking 10 weeks of data here.
Over the same period, NEAR DA cost is also just a fraction of Celestia.
25/ In conclusion, @NEARProtocol provides a very powerful DA solution that will definitely find adoption across the modular stack.
Cheap and scalable DA with battle-tested security and built on future-proof technology.
26/ Integrations with rollup SDKs/stacks such as the @0xPolygon CDK, or the @arbitrum Orbit Stack and Rollup-as-a-Service providers (e.g. @alt_layer) continue to expand.
Soon, there will be an entire ecosystem of modular chains built on @NEARProtocol.
Don't fade, anon. 👀
That's a wrap for today! I hope you've found this thread helpful! 🫡
Make sure to follow me @zerokn0wledge_ and @NEARProtocol to stay ahead of the curve!
Like/Repost of the below post is also highly appreciated! 🙏👇
13 more images on X
How the covered assets moved · since 22 Mar 2024
Prices as of 23 Oct 2026This data piece reports the numbers without making a call on the token, so price moves since publication are shown as market context.
Covered in this piece
Since then
Zero has written 48 more pieces on NEAR Protocol and $NEAR since this one.
- Quantum and post-quantum map: PQ crypto assets by readinessOn NEAR
- I Built a Research Desk Staffed Entirely by AI Agents. It Works.Covers $NEAR
- New vs returning wallets across Ethereum, Solana, NEAR, Polygon and SuiCovers $NEAR
- NEAR DCML explained: confidential, verifiable and monetizable AI via TEEsCovers $NEAR
Daily closes in USD from DefiLlama and Yahoo Finance. Only a bullish or bearish thesis is read as a call; explainers, maps and mentions are context. Not financial advice.