Module pool 

Memory pools for BFS and graph traversal allocations

Provides reusable memory pools to reduce allocation overhead during repeated BFS/DFS operations in minimum cut algorithms.

Overview

Graph algorithms like BFS perform many allocations:

• Queue for vertices to visit
• HashSet/BitSet for visited vertices
• Vec for collecting results

By reusing these structures, we avoid repeated allocation/deallocation overhead, which can be significant for algorithms that perform many traversals.

Thread Safety

The pools use thread-local storage for zero-contention access. Each thread gets its own pool of resources.

Example

use ruvector_mincut::pool::BfsPool;

// Acquire resources from the pool
let mut resources = BfsPool::acquire(256);

// Use resources for BFS
resources.queue.push_back(0);
resources.visited.insert(0);

// Clear and return to pool automatically when dropped
drop(resources);

Structs

BfsPool

Pool for BFS memory allocation

BfsResources

BFS resources acquired from the pool

CompactBfsResources

Compact bitset pool for small graphs

DistanceBfsResources

Pool for distance-annotated BFS