🤖 AI Summary
In distributed systems, dynamic bucket scaling causes massive key remapping under conventional modulo-based hashing, triggering resource request spikes; meanwhile, mainstream consistent hashing relies on floating-point arithmetic or specialized hash families, limiting performance and engineering practicality. This paper proposes JumpBackHash: a purely integer-domain consistent hashing algorithm that employs a jump-backtracking strategy with bitwise operations to achieve deterministic probe sequences—eliminating floating-point computations and dependence on non-standard hash functions. JumpBackHash achieves theoretically optimal remapping rates while delivering 3.2× higher throughput than state-of-the-art alternatives. It uniquely balances stability, computational efficiency, and deployment simplicity, offering a highly practical and scalable hashing solution for dynamic scaling scenarios.
📝 Abstract
Distributed data processing and storage systems require efficient methods to distribute keys across buckets. While simple and fast, the traditional modulo‐based mapping is unstable when the number of buckets changes, leading to spikes in system resource utilization, such as network or database requests. Consistent hash algorithms minimize remappings but are either significantly slower, require floating‐point arithmetic, or are based on a family of hash functions rarely available in standard libraries. This work introduces JumpBackHash, a consistent hash algorithm that overcomes those shortcomings.