In Search for a Linear Byzantine Agreement

The long-standing byzantine agreement problem gets more attention in recent years due to the increasing demand for scalable geo-replicated Byzantine state machine replication (SMR) systems (e.g., Blockchains). To date, the key bottleneck of such systems is the communication cost of the byzantine agreement they employ as a building block, which motivates many researchers to search for low-communication byzantine agreement protocols. The conventional approach is to design deterministic protocols in the eventually synchronous communication model that are optimized to reduce the communication cost after the global stabilization time (GST). In this paper, we challenge the conventional approach and argue it is not the best fit for scalable SMR systems since it might induce an unbounded communication cost during asynchronous periods before GST, which we prove to be inherent. Instead, we forgo eventual synchrony and propose a different approach that hopes for the best (synchrony) but prepares for the worst (asynchrony). Accordingly, we design an optimistic protocol that first tries to reach an agreement via an efficient deterministic algorithm that relies on synchrony for termination, and then, only if an agreement was not reached due to asynchrony, the protocol uses a randomized asynchronous algorithm for fallback that guarantees termination with probability $1$. Although randomized asynchronous algorithms are considered to be costly, we design our solution to pay this cost only when an equivalent cost has already been paid while unsuccessfully trying the synchronous protocol. Moreover, we formally prove that our protocol achieves optimal communication complexity under all network conditions and failure scenarios.