The design of next generation proof-of-stake Ethereum brings various security and performance challenges. We devise attacks and security analyses to support ongoing Ethereum research and development efforts.
One might desire an ideal blockchain protocol to simultaneously provide liveness under dynamic participation, safety under temporary network partitions, and accountability to punish adversarial behavior. Yet, various impossibility results show that no single ledger can have all these properties, each of which is a desideratum for a global consensus layer such as Ethereum. Supported by a gift from the Ethereum foundation, we propose problem formulations and protocol constructions that reconcile these desiderata using multiple nested ledgers. This provides a formal framework to reason about the design landscape for next generation proof-of-stake Ethereum. We also analyze the security of and devise attacks on existing protocols such as Gasper, the Ethereum 2 beacon chain protocol, to highlight attack vectors and spur security enhancements of Ethereum.
David Tse
Room 264, Packard Building
Electrical Engineering Department
350 Jane Stanford Way
Stanford, CA 94305-9515
dntse@stanford.edu
Helen Niu
Room 310, Packard Building
Electrical Engineering Department
350 Jane Stanford Way
Stanford, CA 94305-9515
Tel: (650) 723-8121
Fax: (650) 723-9251
helen.niu@stanford.edu
