Streamlining Ethereum: Prototype Cuts Blob-Recovery Work by Up to 18x
Ethereum is refining its data availability architecture through a new prototype that drastically lowers the computational requirements for node operators. Recent simulations of a "reduced" networking design showed an 11–18× reduction in the work needed to reconstruct data blobs, which are essential for Layer-2 rollup transactions. By optimizing how nodes handle missing data, this proposal offers a more efficient alternative to current methods without requiring the full complexity of upcoming networking upgrades.
Optimizing Reconstruction via Duty Assignment
The proposed system, detailed by researcher Csaba Kiraly, focuses on eliminating duplicated work across the network. Under the existing PeerDAS model, multiple "high-custody" nodes often reconstruct the same missing blob data simultaneously, wasting significant CPU resources. The new prototype assigns recovery duties to specific nodes first, allowing others to receive the finished data rather than rebuilding it themselves. Simulation data suggests that in a network of 1,000 nodes, the total computing cost could drop from 48.6 CPU-seconds to a mere 2.75 CPU-seconds. This represents a massive gain in efficiency for node operators using hardware like the Ryzen 9 series.
A Practical Path Toward RowDAS
This design serves as a strategic precursor to the full RowDAS proposal (EIP-8371), which aims to allow smaller nodes to collectively reconstruct data. By assigning duties within existing column-distribution channels, the prototype provides an immediate efficiency boost without needing the new row-networking channels required for the full EIP. It maintains a "backstop" where high-custody nodes retain their recovery roles, ensuring the network remains resilient even if assigned nodes fail. While the results are currently based on simulations using real cryptography, they highlight a clear opportunity to reduce processor strain. Further real-world testing on devnets will be necessary to confirm these performance gains before the design is fully integrated into Ethereum’s scaling roadmap.
