Block-based consensus: Synchronous

https://gyazo.com/0ac0a1bf2915fdabff4206982443037a

DFINITY

Timo Hanke, Mahnush Movahedi and Dominic William

Ittai Abraham, Dahlia Malkhi, Kartik Nayak, and Ling Ren (VMware Research)

18/ It turned out we could design blockchain protocols that are 1. far faster than those using traditional consensus, 2. scale to any number of participants as they should and 3. continue to make progress during asynchrony

T-H. Hubert, Chan Rafael and Pass Elaine Shi

Stability-favoring proposer rotation policy geared towards performance-demanding scenarios

In HotStuff, liveness relies on having a rotating proposer (i.e., the “democracy-favoring approach”)

Doubly-pipelined BFT: a generalization of the pipelined-BFT for performance

Possible to be adopted by ThunderCore

Ittai Abraham, Dahlia Malkhi, Kartik Nayak, Ling Ren, and Maofan Yin

Optimistic Responsiveness

PiLi requires a latency of at least 40∆-65∆ (65∆ for the earliest block and 40∆ for the last committed block). In comparison, our solution requires 2∆ + 9δ latency where δ denotes the actual network delay

S&P'20

Ittai Abraham, Srinivas Devadas, Danny Dolev, Kartik Nayak, and Ling Ren

Ittai Abraham, Kartik Nayak, Ling Ren, and Zhuolun Xiang

1∆-SMR: A new synchronous BFT SMR, n=2f+1, where each decision has optimal good-case latency of just 1∆ (twice as fast as Sync Hotstuff)

2∆-MLF: Optimal good-case latency under Mobile Link Failures

1∆-MSF: Optimal good-case latency under Mobile Sluggish Faults

Atsuki Momose, Jason Paul Cruz, and Yuichi Kaji

Ittai Abraham and Kartik Nayak and Ling Ren and Nibesh Shrestha

CCS'20

Hybrid

Rafael Pass Elaine Shi (CornellTech/Cornell, IC3)

Improve latency in the optimistic case (3/4 of the computing power is controlled by honest players) with and a special player called the “accelerator”

Otherwise, fall back on slow chain