reed-solomon

reed-solomon

Reed-Solomon erasure coding in pure Javascript. A Javascript port of the JavaReedSolomon library released by Backblaze. For an introduction to erasure coding, see the post by Brian Beach on the Backblaze blog. Special thanks to Backblaze.

License

reed-solomon is licensed under the MIT License, which means that you can use it in your own projects for free. You can even use it in commercial projects.

Installation

npm install reed-solomon

Efficiency

Data redundancy is typically achieved through mirroring or replication at a cost of 3x the original data. With Reed-Solomon erasure codes, you can achieve better redundancy at a cost of only 1.5x the original data, for example. Various storage efficiencies of 1.4x and 1.18x are also possible. You can trade storage efficiency, redundancy and recovery time by fine-tuning the number of data shards and parity shards you use.

Performance

reed-solomon includes a Javascript binding as well as an optional native binding (and simple benchmark):

ReedSolomon(17, 3)

1,8 GHz Intel Core i5

Binding: Native

Encode: 400.00 MB/s
Decode: 369.57 MB/s

Binding: Javascript

Encode: 195.68 MB/s
Decode: 193.18 MB/s

Optional Native Binding

The native binding will be installed by default when installing reed-solomon, and the Javascript binding will be used if the native binding could not be compiled. To compile the native binding manually, install node-gyp globally:

sudo npm install node-gyp -g

Then build the binding from within the reed-solomon module directory:

cd node_modules/reed-solomon
node-gyp rebuild

Usage

Divide a single Buffer into an Array of fixed-size data shards, then use reed-solomon to compute as many parity shards as you need. If you lose some data shards or some parity shards (no more than the number of parity shards you added), you can use reed-solomon to reconstruct the missing data and parity shards.

Encoding

var ReedSolomon = require('reed-solomon');
var dataShards = 6;
var parityShards = 3;
var shardSize = 1024 * 1024;
var shards = [
  // Data shards (containing user data):
  <Buffer (shardSize) >,
  <Buffer (shardSize) >,
  <Buffer (shardSize) >,
  <Buffer (shardSize) >,
  <Buffer (shardSize) >,
  <Buffer (shardSize) >,
  // Parity shards:
  new Buffer(shardSize),
  new Buffer(shardSize),
  new Buffer(shardSize)
];
var rs = new ReedSolomon(dataShards, parityShards);
var offset = 0; // The offset of each shard within each buffer.
var size = shardSize; // The size of each shard within each buffer.
rs.encode(shards, offset, size);
// Parity shards now contain parity data.

Verifying Parity Shards

rs.isParityCorrect(shards, offset, size); // true/false

Decoding Corrupted Shards

// Corrupt a data shard:
shards[0] = new Buffer(shardSize);
// Corrupt a parity shard:
shards[shards.length - 1] = new Buffer(shardSize);
// We still have enough parity to corrupt another shard.

// Decode the corrupted data and parity shards:
var present = [
  false, // We indicate that shard 1/9 is corrupt. This is a data shard.
  true,
  true,
  true,
  true,
  true,
  true,
  true,
  false // We indicate that shard 9/9 is corrupt. This is a parity shard.
];
rs.decode(shards, offset, size, present);
// Shards 1 and 9 have been repaired.

Tests

reed-solomon ships with extensive tests, including a long-running fuzz test.

cd node-modules/reed-solomon
node test.js