This is a segment of the 0xresearch newsletter. Subscribe to read full editions.
Since Ethereum, blockchain networks have stylized themselves as “decentralized world computers”. But an important part of modern computer use is missing: memory.
Now, with the launch of his private test network this week, MIT-inspired Startup Optimum is taking a first step to repair that.
Founded by MIT-professor and renowned network coding researcher Muriel Médard, the optimum strives to build a powerful memory layer for block chains. Starting with data propagation and ultimately approaches, which comes down to decentralized RAM, the core innovation is based on random linear network coding (RLNC), a technique that Médard has refined for more than two decades.
In contrast to typical gossip-based peer-to-peer systems and junctions and superfluous packages transmit RLNC-used data packages that increase the efficiency and resilience of the bandwidth. In private tests and academic research, the technology has shown up to 20x improvements in propagation band and 2x profits in speed, with flexible considerations, depending on the application needs, according to the team.
The first tests with validator Node Operators are now underway, part of the broader go-to-market strategy of the team, Médard told Blockworks.
“We have drawn several mous with large junction operators,” said Médard. The private test network is only inviting and an opportunity for operators to validate the performance and to shape the product direction.
OptimP2P, the first product of the protocol, replaces traditional gossip protocols such as LIBP2P with a Pub-SUB (publishing-SubSubscribe) that functions more as a memory bus in traditional computer use. The technology could have clear benefits, such as faster block propagation, reduced mempool congestion and better performance for validators, RPC providers and even DEX operators.
Some of these improvements go directly to the bottom line of an operator. But in the end, optimum will also be aimed at small players. It is planning to introduce a utility to reward nodes – called “flex nodes” – for contributing to data planting and, ultimately, memory access, with rewards that are proportional to the useful work they do, regardless of the uptime or the importance.
“Almost all blockchain innovation has focused on calculation and implementation,” said Médard. “But without fast, reliable data access and distribution, block chains cannot function as real high-speed computing networks.”
The Blockchain -agnostic approach to Optimum has already attracted considerable support, the most recently a sperm of $ 11 million under the leadership of 1KX, and including investors such as robot ventures, CMT Digital, Spartan and Finality Capital.
This focus on practical performance improvements has attracted the attention of other infrastructure teams, including providers for data availability such as B are benefits.
Pabal Banerjee, co-founder of Besut, noted the potential relevance of RLNC for his route map.
“RLNC and its application to optimize P2P is interesting for us,” Banerjee told Blockworks, although the addition of it must be revised. “At this stage we explore multiple roads to optimize the network for 10 GB blocks based on the strategy described in our scale route map.”