ep 24 – Of Data Availability & Danksharding

Scaling Ethereum requires addressing the data storage bottleneck for rollups, a challenge met by Danksharding and data availability sampling. By breaking large blocks into smaller, distributed fragments, the network avoids overloading individual validators while maintaining data integrity through erasure coding. Cryptographer Dan Boneh and research partner Lera Nikolaenko explain that this mechanism allows for efficient data reconstruction even when some nodes are offline. Their research proposes using bivariate polynomial interpolation to lower the reconstruction threshold from 75% to 25%, which would significantly improve system efficiency and reduce the number of samples required for verification. This approach treats data availability as a probabilistic check, akin to hitting a target with darts to ensure sufficient coverage, ultimately enabling faster, cheaper transactions and broader application utility for the Ethereum ecosystem.