An Algorithm and Architecture Co-design for Accelerating Smart Contracts in Blockchain

What is it about?

Inefficient smart contract transaction execution severely limits blockchain‘s throughput and hinders the further application of blockchain. In this work, we achieve efficient multi-layer parallelism of transactions through an algorithm and architecture co-design.

Featured Image

Photo by Shubham Dhage on Unsplash

Photo by Shubham Dhage on Unsplash

Why is it important?

The beauty of blockchains supporting smart contracts is the high degree of decentralization and complete freedom, but this comes at the cost of performance, especially throughput [13]. For example, Ethereum, the global permissionless blockchain platform, has a throughput of only 13 transactions per second (TPS). While existing work improves the throughput by sacrificing partial decentralization or establishing new trust assumptions, this also falls far short of the requirements for real-time performance in large-scale digital scenarios. If the execution of the transactions themselves could be made more efficient, the blockchain throughput would be fundamentally mitigated without tradeoffs. Multi-layer serialization under consistency constraints is the root cause of inefficient transaction execution. In this work, we analyze the acceleration opportunities from multiple perspectives, inspiring us to explore a better algorithm and hardware architecture.