Problem
Research questions and friction points this paper is trying to address.
geo-replication
consensus
fast path
latency optimization
KCensus: Synthesizing Latency-Optimal Consensus Fast Paths (Extended Version)
🤖 AI Summary
This work addresses the limitation that existing fast-path protocols for strongly consistent geo-replication predominantly rely on manual design, rendering them ill-suited for heterogeneous network topologies and dynamic workloads. We propose a novel theoretical framework grounded in a knowledge dissemination perspective that reformulates fast-path protocol design as a tractable mathematical optimization problem, thereby enabling the automatic synthesis of optimal protocols tailored to specific deployment environments. Building upon this framework, we develop a geo-replicated key-value store and validate it through cross-region deployments on AWS. Experimental results demonstrate that the synthesized system reduces average latency by up to 16% compared to state-of-the-art protocols. Ultimately, this study achieves both the automated design of consensus protocols and substantial performance improvements in geo-distributed systems.
📝 Abstract
Strongly consistent geo-replication often relies on fast paths to reduce latency in the common case of no failures or contention. Existing fast-path schemes, however, are ad hoc and restrictive: each corresponds to a point in a broad design space shaped by network topology, workload, and latency objective, so no single scheme works best across settings. This paper looks at fast-path schemes from a new perspective, as mechanisms that spread knowledge about proposals. With this view, we identify a fundamental condition on the spread of knowledge for a fast-path scheme to work. We then introduce KCensus, a framework that turns this condition into an optimization problem, synthesizing new fast-path schemes that are optimal for a given setting. We use KCensus to build a geo-replicated key-value store and evaluate it across AWS regions. Our system outperforms competing protocols, with up to 16% lower average latency.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Fast-path consensus
Knowledge spread
Geo-replication
Latency optimization
KCensus
🔎 Similar Papers
No similar papers found.
