Understanding the Value of Parallelism
Rachel Bousfield, a tech lead at Offchain Labs, believes that the value of parallelism is often exaggerated. In an interview with Blockworks, she emphasized that the returns and fee reductions people anticipate do not always materialize in practice. Ethereum, for instance, is designed to ensure that running a node is affordable, allowing low-end computers to contribute effectively to the network by running applications or validators.
In contrast, other blockchains that enable parallelism also impose higher costs for running a node. Ryan Watkins, co-founder of Syncracy Capital, highlights that Solana, a network that supports parallelism, faces different trade-offs. Bousfield questions whether increasing demands and costs for Ethereum validators would truly result in improved performance, increased capacity, and lower fees, concluding that it may not be the most advisable decision for Ethereum.
The Benefits and Challenges of Parallelism
Bousfield recognizes that parallelism can enhance throughput and efficiency when multiple users engage in distinct crypto activities simultaneously. However, the reality is that most people tend to engage in similar activities, such as rushing to exploit price discrepancies on decentralized exchanges (DEXs) for maximum value extraction (MEV) arbitrage opportunities. This phenomenon, known as “contention,” leads to higher gas prices when multiple users compete to perform the same action.
Polygon Labs’ study reveals that parallelism is applicable to approximately 55% of transactions in most blocks on its network. However, even with perfect parallelism, capable of executing tasks instantaneously, it could only double Polygon’s capacity at best.
While Bousfield acknowledges that parallelism has its merits, she cautions against overestimating its potential as a panacea for blockchain scaling challenges.
Tackling Transaction Speed with Arbitrum Stylus
To address transaction speed and improve throughput, Bousfield points to Arbitrum Stylus. This solution simplifies data interpretation for hardware, eliminating time-consuming steps such as accuracy checks, branch enabling, and memory simulation prevalent in traditional Ethereum Virtual Machines (EVMs).
By removing these interpretive layers, Stylus achieves a significant speed boost of 10-100x on all compute workloads, offering a promising approach to enhancing scalability.