Cost sensitive moving target consensus

Consensus is a fundamental approach to implementing fault-tolerant services through replication. It is well known that there exists a tradeoff between the cost and the resilience. For instance, Crash Fault Tolerant (CFT) protocols have a low cost but can only handle crash failures while Byzantine Fault Tolerant (BFT) protocols handle arbitrary failures but have a higher cost. Hybrid protocols enjoy the benefits of both high performance without failures and high resiliency under failures by switching among different subprotocols. However, it is challenging to determine which subprotocols should be used. We propose a moving target approach to switch among protocols according to the existing system and network vulnerability. At the core of our approach is a formalized cost model that evaluates the vulnerability and performance of consensus protocols based on real-time Intrusion Detection System (IDS) signals. Based on the evaluation results, we demonstrate that a safe, cheap, and unpredictable protocol is always used and a high IDS error rate can be tolerated.

I. Introduction

Consensus is a generic technique that implements faulttolerant services through replication. It is critical to achieve both reliability and availability in various online services and cloud computing applications, including Google's Chubby [6], Amazon Web Services [1], and VMware's vSphere [2], [3]. Depending on the types of failures we aim to tolerate, various protocols are designed with different security guarantees and performance characteristics. It is in general known that there exists a tradeoff between the resiliency and the cost of the consensus protocols. Namely, a low cost protocol with low redundancy and high performance can only handle limited types of failures. For instance, Crash Fault Tolerant (CFT) protocols are less redundant but can only tolerate crash fail-ures. In comparison, Byzantine Fault Tolerant (BFT) protocols handle arbitrary failures but are very expensive, which may be an overkill to use most of the time.