Is the DA war coming to an end? Deconstructing PeerDAS: How can it help Ethereum reclaim "data sovereignty"?

Written by: imToken

By the end of 2025, the Ethereum community quietly welcomed the completion of the Fusaka upgrade.

Looking back over the past year, although discussions about underlying technical upgrades have gradually faded from the market spotlight, many on-chain users have undoubtedly noticed a significant change: Ethereum L2s have become increasingly cheaper.

Now, whether it's a simple transfer or a complex DeFi operation, on-chain interactions often require only a few cents in gas fees, sometimes even negligible. Behind this, the Dencun upgrade and the Blob mechanism have certainly played a major role. At the same time, with the official activation of the core feature PeerDAS (Peer Data Availability Sampling) in the Fusaka upgrade, Ethereum is also bidding farewell to the era of "full download" data verification.

It can be said that what truly determines whether Ethereum can sustainably support large-scale applications in the long run is not just the Blob itself, but more importantly, the next step represented by PeerDAS.

I. What is PeerDAS?

To understand the revolutionary significance of PeerDAS, we cannot just discuss the concept in isolation; we must first look back at a key milestone in Ethereum's scaling journey, namely the Dencun upgrade in March 2024.

At that time, EIP-4844 introduced a transaction model carrying Blobs (embedding large amounts of transaction data into blobs), allowing L2s to no longer rely on the expensive calldata storage mechanism, but instead use temporary Blob storage.

This change directly reduced the cost of Rollups to a fraction of what it was before, ensuring that L2 platforms could offer cheaper and faster transactions without compromising Ethereum's security and decentralization, and allowed users to enjoy the benefits of the "low gas fee era."

However, although Blobs are very useful, the number of Blobs that each block on the Ethereum mainnet can carry has a hard upper limit (usually 3-6), for a very practical reason: physical bandwidth and hard drive space are limited.

Under the traditional verification model, every validator in the network—whether a server operated by a professional institution or an ordinary home computer—still had to download and propagate the complete Blob data in order to confirm its validity.

This led to a dilemma:

• If you increase the number of Blobs (for scaling): the data volume surges, home nodes' bandwidth will be maxed out, hard drives will be overloaded, forcing them offline, and the network will quickly become centralized, eventually turning into a giant chain that only large data centers can run;
• If you limit the number of Blobs (for decentralization): the throughput of L2s is capped, unable to meet the explosive growth in future demand.

Simply put, Blobs are just the first step, solving the problem of "where to store the data." When the data volume is small, everything is fine, but if the number of Rollups continues to increase in the future, and each Rollup frequently submits data, Blob capacity keeps expanding, then the bandwidth and storage pressure on nodes will become a new centralization risk.

If the traditional full download model continues, and bandwidth pressure cannot be solved, Ethereum's scaling journey will hit a brick wall at the limits of physical bandwidth. PeerDAS is the key to unlocking this deadlock.

In a nutshell, PeerDAS is essentially a brand-new data verification architecture that breaks the iron rule that verification must involve full downloads, allowing Blob expansion to surpass current physical throughput levels (for example, jumping from 6 Blobs/block to 48 or even more).

II. Blob solves "where to store," PeerDAS solves "how to store"

As mentioned above, Blobs took the first step in scaling, solving the problem of "where to store the data" (moving from expensive Calldata to temporary Blob space), while PeerDAS aims to solve the problem of "how to store more efficiently."

The core issue it addresses is how to prevent nodes' physical bandwidth from being overwhelmed as data volume grows exponentially. The idea is straightforward: based on probability and distributed collaboration, "not everyone needs to store all the data, yet we can still confirm with high probability that the data truly exists."

This is evident from the full name of PeerDAS: "Peer Data Availability Sampling."

This concept may sound obscure, but we can use a simple analogy to understand this paradigm shift. For example, the previous full verification was like a library acquiring a multi-thousand-page "Encyclopedia Britannica" (Blob data), and to prevent loss, every librarian (node) was required to make a full copy as a backup.

This meant that only those with enough money and time (large bandwidth/hard drives) could be librarians, especially as the "Encyclopedia Britannica" (Blob data) kept expanding, with more and more content. Over time, ordinary people would be eliminated, and decentralization would disappear.

Now, with PeerDAS sampling and the introduction of technologies like Erasure Coding, it's as if the book can be torn into countless fragments and mathematically encoded and expanded. Each librarian no longer needs to keep the whole book, but only needs to randomly keep a few pages.

Even during verification, no one needs to present the entire book. In theory, as long as the network can gather any 50% of the fragments (regardless of whether people hold page 10 or page 100), we can use mathematical algorithms to instantly reconstruct the entire book with 100% certainty.

This is the magic of PeerDAS—shifting the burden of downloading data from individual nodes to a collaborative network of thousands of nodes across the entire network.

Source: @Maaztwts

Just from an intuitive data perspective, before the Fusaka upgrade, the number of Blobs was stuck in the single digits (3-6). With the implementation of PeerDAS, this upper limit is directly broken, allowing the Blob target to jump from 6 to 48 or even more.

When users initiate a transaction on Arbitrum or Optimism and the data is packaged and sent back to the mainnet, there is no longer a need to broadcast the complete data packet across the entire network. This allows Ethereum to achieve scaling without linearly increasing node costs.

Objectively speaking, Blob + PeerDAS is the complete DA (Data Availability) solution. From a roadmap perspective, this is also the key transition for Ethereum from Proto-Danksharding to full Danksharding.

III. The New On-Chain Normal in the Post-Fusaka Era

As is well known, in the past two years, third-party modular DA layers such as Celestia once gained huge market space due to the high cost of Ethereum mainnet. Their narrative logic was built on the premise that Ethereum's native data storage was expensive.

Now, with Blobs and the latest PeerDAS, Ethereum has become both cheap and extremely secure: the cost for L2s to publish data to L1 has been slashed by more than half, and Ethereum also has the largest validator set in the entire network, with security far surpassing third-party chains.

Objectively, this is a dimensionality reduction blow to third-party DA solutions like Celestia, marking Ethereum's reclaiming of data availability sovereignty and greatly squeezing their room for survival.

You might ask, all of this sounds very low-level—what does it have to do with my wallet, transfers, or DeFi?

The connection is actually very direct. If PeerDAS can be successfully implemented, it means that L2 data costs can remain low in the long term, Rollups won't be forced to raise fees due to DA cost rebounds, on-chain applications can confidently design high-frequency interactions, and wallets and DApps no longer have to constantly compromise between "features vs. cost"....

In other words, the reason we can use cheap L2s today is thanks to Blobs, and if we can continue to afford them in the future, it will be due to the silent contribution of PeerDAS.

This is also why, in Ethereum's scaling roadmap, PeerDAS, though low-key, has always been seen as an indispensable step. Essentially, this is the best form of technology in the author's eyes—"beneficial yet unnoticed, but hard to live without if lost," making you unaware of its existence.

Ultimately, PeerDAS proves that blockchain can, through ingenious mathematical design (such as data sampling), support Web2-level massive data without excessively sacrificing the vision of decentralization.

With this, Ethereum's data highway has been fully paved. What kind of vehicles will run on this road next is a question for the application layer to answer.

Let's wait and see.