bloom-filter

High-performance, memory-efficient probabilistic data structure for membership testing.

Part of the Ocyrus suite.

✨ Features

• Memory Efficient: Test membership in a set of millions of items with just a few megabytes of memory.
• Configurable: Choose your expected item count and target false-positive rate.
• Blazing Fast: $O(k)$ lookup speed (where $k$ is number of hash functions).
• Zero-Allocation Lookup: Hashing and bit manipulation performed without temporary object creation.

🚀 Performance

• has (lookup): ~4.0M ops/sec
• add (insertion): ~1.8M ops/sec

📦 Installation

npm install @ocyrusjs/bloom-filter

🛠️ Usage

import { BloomFilter } from '@ocyrusjs/bloom-filter';

// Create a filter for 10,000 items with a 1% false positive rate
const filter = new BloomFilter(10000, 0.01);

filter.add('user_123');

// Guaranteed to be true if added
filter.has('user_123'); // true

// May return true (false positive) or false
filter.has('user_456');

License

MIT © Nishith Patel