Ethereum Struggles With Data-Heavy Blocks After Fusaka Upgrade, Research Finds

Ethereum is showing signs of strain when handling data-heavy blocks weeks after its December Fusaka upgrade, raising concerns about whether the network is ready to support higher data throughput from layer 2 blockchains, according to new research from MigaLabs.

The Fusaka hard fork was designed to expand Ethereum’s data capacity by allowing layer 2 networks to submit more “blobs,” a form of temporary data used primarily by rollups to post transaction information to the main chain.

The change was widely seen as a step toward cheaper and more scalable layer 2 activity.

Ethereum Blocks With Higher Blob Counts Face Higher Miss Rates

However, an empirical analysis published by MigaLabs suggests that blocks carrying higher blob counts are significantly more likely to be missed by the network.

The research draws on data collected since October 2025 and examines network behavior before and after Fusaka, as well as two subsequent Blob-Parameter-Only (BPO) updates that raised blob limits further.

MigaLabs, which has previously collaborated with Lido DAO and the Cambridge Centre for Alternative Finance, found that Ethereum is not coming close to using the expanded capacity.

Despite increases to the target blob count, most recently raised to 14, the median number of blobs per slot has actually fallen since the first BPO update.

High blob counts of 16 or more remain rare, appearing only a few hundred times out of more than 750,000 observed slots.

More troubling is what happens when blob counts do spike. The study shows that missed-slot rates rise sharply once blocks contain 16 or more blobs.

Tomorrow: Fusaka

Ethereum’s second major upgrade this year.

→ Feature highlight: PeerDAS – Unlocking up to 8x data throughput. For rollups, this means cheaper blob fees and more space to grow.

Learn more. https://t.co/3TOda5KjY2 pic.twitter.com/sEfeiTamy9

— Ethereum (@ethereum) December 2, 2025

While the baseline miss rate for slots with up to 15 blobs hovers around 0.5%, miss rates at higher blob counts range from 0.77% to as high as 1.79.

At the maximum observed level of 21 blobs, the miss rate was more than three times the network average.

These blobs are primarily submitted by large layer 2 networks such as Arbitrum and Base, which rely on Ethereum’s data availability to operate securely.

If demand from these networks increases and high blob counts become more common, the elevated miss rates could compound and pose risks to overall network stability.

MigaLabs Urges Pause on Ethereum Blob Increases Amid Rising Miss Rates

MigaLabs cautioned that while the sample size for very high blob counts is still limited, the pattern is consistent across all observed data points.

In its conclusion, the firm recommended against any further increases to blob capacity until miss rates at higher blob levels return to baseline and real demand begins to approach existing limits.

As reported, the Ethereum Foundation has elevated post-quantum security to a core strategic focus, forming a dedicated Post Quantum team and committing $2 million to the effort.

Announced by Ethereum researcher Justin Drake, the initiative will be led by Thomas Coratger alongside Emile, a contributor to leanVM.