Skip to content

A sophisticated approach to bloom filter for NodeJs supporting deletions and serialization

Notifications You must be signed in to change notification settings

raghav-grover/node-bloom-filter

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Node Bloom filter

Node bloom filter is a sophisticated Bloom filter available for NodeJs which can be optionally transformed to a counting bloom filter and can thus perform delete operations as well.

  • Supports deletions
  • Can yield a desired false positive rate(probability) when given the total number of elements in the sets
  • Supports an optimisation flag to calculate the number of hashes and bits to use internally
  • Serialization to an Arraybuffer, which can be converted to JSON/UTF-8/base64 and is left to the user :)
  • Dynamically resizes the underlying Arraybuffer to handle the counter overflows when used as a counting bloom filter.
  • Implements double hashing technique to generate multiple hashes using murmur3
  • Uses bit manipulations to optimise on memory usage
  • When implemented as a counting bloom filter, it can support 4,8,16,32 bits per counters to handle overflows
  • Extends Events to notify when the underlying Arraybuffer is resized

Getting Started

To install: npm install node_bloom_filter

Usage

Bloom Filter

To create a general implementation, pass the following in the constructor:

let bFilter = new bloomFilter({
    nHash:10,
    nBits:1024*8
});

If no options are passed, the filter created by default can support 569 members with a false rate of upto 0.001

To create the optimized implementation, pass an object in the constructor:

let bFilter = new bloomFilter({
    optimize: true,/*This automatically calculates number of hashes and bits to be used internally*/
    falsePositiveRate: 0.001,/*Desired false positive rate,less rate will use more memory internally*/
    isCounting: false,// False by default,
    nElements:10000
});

To create a counting bloom Filter that supports deletions as well, just pass

let bFilter = new bloomFilter({
    optimize: true,/*This automatically calculates number of hashes and bits to be used internally*/
    falsePositiveRate: 0.001,/*Desired false positive rate,less rate will use more memory internally*/
    isCounting: true,// False by default,
    nElements:10000
});

Add an element

bFilter.add('hello world');

Test for membership

if(bFilter.has('hello world')===true){
    console.log('Present, but may be a false positive');
}else{
    console.log('Not present definitely');
}

Delete

This is available only for couting bloom filters

bFilter.delete('hello world');

Serialization and Deserialization

let bFilterArrayBuffer=bFilter.serialize();
let newFilter=new bloomFilter({},bFilterArrayBuffer);
newFilter.add('hello, i am the best bloom filter');

On resize

bFilter.on('resized',(bitsPerCounter)=>{
    console.log('bits per counter',bitsPerCounter);
});

Test

To test: npm run test

Acknowledgements

  • Hat tip to anyone who uses and contributes to the code....
  • Purav Aggarwal :)

About

A sophisticated approach to bloom filter for NodeJs supporting deletions and serialization

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published