CavityRank：针对布谷鸟过滤器的零额外字节残差路由
CavityRank: Zero-Extra-Byte Residual Routing for Cuckoo Filters
浏览论文内容
中文总结 AI 辅助
CavityRank 为布谷鸟过滤器设计的零额外字节残差路由，通过利用压缩桶资源编码秩，在插入性能上优于随机踢出与深度10 BFS，缩小了预言机残差间隙，处于实用设计区间。
中文摘要 AI 辅助
接近容量时，布谷鸟过滤器（Cuckoo Filter）即便仍存在合法放置位置，也可能拒绝插入：表结构仍可行，但有界策略无法找到增广路径。随机踢出（Random Kick-out）使每一步开销低廉，但无持久方向；广度优先搜索（BFS）通过扩展前沿并维护表级状态恢复方向。CavityRank 利用四槽压缩桶中已有的资源：查找仅观察指纹多重集，因此查询等价的 lane 顺序可编码两个比较位，无需加宽 64 位桶或改变双桶查询；这些位构成四级有序残差秩。插入操作遵循最小秩边，重定位后从其出边重新编码每个修改的桶，传播压缩字实际实现的秩。精确的容量四定向预言机将结构不可行性与有界搜索损失区分开。在配对的 4096 桶 XOR16 梯中，CR2 缩小了 Random CF 86.47% 的预言机间隙，而 CavityRank 仅剩余该原始间隙的 1.39%。规范平局的 CR2 与 CavityRank 消融实验分离出第二个隐式位，其分别缩小 CR2 残差间隙的 89.95% 和 90.69%；65536 桶时对应缩小比例为 83.95% 和 84.27%。在 64 MiB 负载 97.75% 的独立压缩实现研究中，每次插入的逻辑读取为 42.53，而显式标签为 62.67、深度 10 BFS 为 355.74，且每桶零额外字节、无表级工作空间。因此，CavityRank 处于无引导驱逐与前沿搜索之间的实用设计点。
英文摘要
Near capacity, a cuckoo filter may reject an insertion even though a legal placement still exists: the table remains structurally feasible, but a bounded policy fails to find an augmenting path. Random kick-out keeps each step cheap but leaves no persistent direction; breadth-first search recovers direction by expanding a frontier and maintaining table-scaled state. CavityRank exploits a resource already present in four-slot packed buckets. Lookup observes only the fingerprint multiset, so query-equivalent lane orders can encode two comparison bits without widening the 64-bit bucket or changing the two-bucket query. The bits form a four-level ordinal residual rank. Insertion follows a minimum-rank edge and re-encodes each modified bucket from its outgoing edges after relocation, propagating the rank actually realized by the packed word. An exact capacity-four orientation oracle separates structural infeasibility from bounded-search loss. In a paired 4,096-bucket XOR16 ladder, CR2 closes 86.47% of Random CF's oracle gap and CavityRank leaves 1.39% of that original gap. A canonical-tie CR2-versus-CavityRank ablation isolates the second implicit bit, which closes 89.95% and 90.69% of CR2's residual gap; the corresponding closures at 65,536 buckets are 83.95% and 84.27%. A separate packed implementation study at 64 MiB and 97.75% load records 42.53 logical reads per insertion, versus 62.67 for explicit labels and 355.74 for depth-10 BFS, with zero extra bytes per bucket and no table-scaled workspace. CavityRank therefore occupies a practical design point between unguided eviction and frontier search.
