WOC: Dual-Path Weighted Object Consensus Made Efficient
By: Tanisha Fonseca, Gengrui Zhang
Modern distributed systems face a critical challenge: existing consensus protocols optimize for either node heterogeneity or workload independence, but not both. For example, Cabinet leverages weighted quorums to handle node heterogeneity but serializes all operations through a global leader, limiting parallelism. EPaxos enables parallel execution for independent operations but treats all nodes uniformly, ignoring performance differences. To tackle this problem, we present WOC, a dual-path consensus protocol that dynamically routes operations into two paths based on their access patterns. Independent operations execute through a fast path that uses object-specific weighted quorums and completes in one network round-trip. Conflicting or shared objects route through a leader-coordinated slow path employing node-weighted consensus. Our evaluation demonstrates that WOC achieves up to 4X higher throughput than Cabinet for workloads with >70% independent objects, while maintaining equivalent performance under high contention.
Similar Papers
Cabinet: Dynamically Weighted Consensus Made Fast
Distributed, Parallel, and Cluster Computing
Makes computer groups work faster, even with slow parts.
A Fast-Converging Decentralized Approach to the Weighted Minimum Vertex Cover Problem
Distributed, Parallel, and Cluster Computing
Helps networks find the best way to watch things.
Non-Ergodic Convergence Algorithms for Distributed Consensus and Coupling-Constrained Optimization
Optimization and Control
Helps power grids share electricity fairly and fast.