Latency and bandwidth-minimizing failure detectors

Failure detectors are fundamental building blocks in distributed systems. Multi-node failure detectors, where the detector is tasked with monitoring N other nodes, play a critical role in overlay networks and peer-to-peer systems. In such networks, failures need to be detected quickly and with low overhead. Achieving these properties simultaneously poses a difficult tradeoff between detection latency and resource consumption. In this paper, we examine this central tradeoff, formalize it as an optimization problem and analytically derive the optimal closed form formulas for multi-node failure detectors. We provide two variants of the optimal solution for optimality metrics appropriate for two different deployment scenarios. √ s-LM is a latency-minimizing optimal failure detector that achieves the lowest average failure detection latency given a fixed bandwidth constraint for system maintenance. √ s-BM is a bandwidth-minimizing optimal failure detector that meets a desired detec...

• Type: Conference
• Year: 2007
• Where: EUROSYS
• Authors: Kelvin C. W. So, Emin Gün Sirer
• Keywords: EUROSYS 2007, Failure Detection Latency, Multi-node Failure Detectors, Optimal Failure Detector, System Software