By Muriel Médard, Kishori Konwar, Nicolas Nicolaou

Blockchains power everything from simple payments to complex financial markets, from AI training to inferencing. As the use cases grow more demanding, existing methods for sharing data between nodes haven’t been able to keep up—networks get congested and slow down under heavy load. Developers with ambitious, latency-sensitive, large-scale ideas actively avoid using blockchains despite their benefits because of these scalability constraints. OptimumP2P fixes this by coding data into random linear equations. The result: faster, more reliable block and transaction delivery for everyone involved.

In peer-to-peer networks, nodes need to disseminate information to each other, a task known as gossip. Gossipsub is a widely-used algorithm to perform

OptimumP2P is coming. Currently in private testnet with a group of leading Ethereum validators and node operators, OptimumP2P boosts the speed and efficiency of communicating data between nodes. It leverages Random Linear Network Coding, a data coding technique our Co-founder Prof. Muriel Medard developed at MIT. RLNC is built specifically for achieving high throughput at scale on decentralized networks. To give a better understanding of how OptimumP2P works and its effects on blockchain ecosystems we’ll be walking through validator implementation, interaction with existing chain architecture, and specific benefits for each of the following groups:

In the first part of this series, we explored how packet loss accumulates in block and fountain codes—even in simple network topologies such as daisy chains—and how the recoding capability of RLNC effectively mitigates this issue. In this follow-up article, we examine hierarchical topologies, such as trees, which have been adopted by Web3 protocols such as Monad and Solana. We present a simple example demonstrating that, even in tree-based architectures, RLNC can outperform both Raptor and Reed–Solomon (RS) codes by achieving lower latency.

Random Linear Network Codes (RLNC) made their appearance in the early 2000s as an erasure coding solution for decentralized...

But as this "world computer" grows, it’s becoming increasingly clear that something fundamental is missing: a high-performance memory layer.