With the growing commercial interest in blockchains, permissioned approaches have received increasing attention. Unfortunately, the BFT consensus algorithms that are the backbone of most of these blockchains scale poorly and offer limited throughput. In fact, many state-of-the-art BFT consensus algorithms require a single leader process to receive and validate votes from a quorum of processes and then broadcast the result, which is inherently non-scalable. Recent approaches avoid this bottleneck by using dissemination/aggregation trees to propagate values and collect and validate votes. However, the use of trees increases the round latency, which ultimately limits the throughput for deeper trees. In this talk we present Kauri, a BFT communication abstraction that can sustain high throughput as the system size grows, leveraging a novel pipelining technique to perform scalable dissemination and aggregation on trees. Furthermore, when the number of faults is small (arguably the most common case in practice), our construction is able to recover from faults in an optimal number of reconfiguration steps. We have implemented and experimentally evaluated Kauri with up to 800 processes. Our the results show that Kauri outperforms the throughput of state-of-the-art permissioned blockchain protocols, such as HotStuff, by up to 58x. Interestingly, in many scenarios, such as when the leader's available bandwidth is the bottleneck, the parallelization provided by Kauri can also decrease the latency.
Please email for a
I am an Assistant Professor at Instituto Superior Técnico of the Universidade de Lisboa and a Researcher at INESC-ID in the Distributed Systems Group.
My current research interests lie on the areas Persistent Memory, and the oportunities and challenges it brings to programmers; blockchain scalability, security and performance; systems evaluation in particular in the topics of reproducibility and automation; as well as databases performance and consistency guarantees.