nano-memoize

Faster than fast, smaller than micro ... nano-memoizer.

Introduction

The devs caiogondim and planttheidea have produced great memoizers. We analyzed their code to see if we could build something faster than fast-memoize and smaller than micro-memoize while adding back some of the functionality of moize removed in micro-memoize. We think we have done it ... but credit to them ... we just merged the best ideas in both and eliminated excess code.

The minified/gzipped size is 887 bytes for nano-memoize vs 959 bytes for micro-memoize. And, nano-memoize has slightly more functionality.

The speed tests are below. nano-memoize is the fastest in all cases. For single argument functions is it comparable to, but slightly and probably un-importantly faster than, fast-memoize. For multiple argument functions it is comparable to, but slightly and probably un-importantly faster than, micro-memoize.

We have found that benchmarks can vary dramatically from O/S to O/S or Node version to Node version. These tests were run on a Windows 10 64bit 2.4ghz machine with 8GB RAM and Node v9.4.0.

Functions with a single primitive parameter...

Name	Ops / sec	Relative margin of error	Sample size
nano-memoize	152,526,010	� 2.58%	80
fast-memoize	147,683,192	� 2.90%	85
micro-memoize	22,682,348	� 3.69%	75
iMemoized	22,292,411	� 4.47%	72
lodash	20,937,311	� 1.94%	88
moize	16,296,876	� 4.77%	74
memoizee	9,651,118	� 3.07%	86
underscore	9,266,277	� 2.66%	75
lru-memoize	6,676,849	� 2.93%	87
addy-osmani	3,899,834	� 2.27%	86
memoizerific	3,753,347	� 2.33%	86
ramda	493,665	� 1.77%	88

Functions with a single object parameter...

Name	Ops / sec	Relative margin of error	Sample size
nano-memoize	53,741,011	� 2.06%	85
fast-memoize	51,041,370	� 2.40%	82
micro-memoize	22,638,078	� 3.96%	77
lodash	22,187,376	� 1.72%	83
moize	19,446,817	� 3.32%	81
underscore	13,643,959	� 3.17%	81
iMemoized	11,926,976	� 5.90%	80
memoizee	8,010,016	� 1.99%	83
lru-memoize	5,709,156	� 1.89%	89
memoizerific	3,817,781	� 1.46%	90
addy-osmani	3,699,956	� 3.30%	85
ramda	793,756	� 1.92%	87

Functions with multiple parameters that contain only primitives...

Name	Ops / sec	Relative margin of error	Sample size
nano-memoize	18,408,074	� 1.24%	85
micro-memoize	17,310,593	� 1.11%	85
moize	14,457,697	� 0.79%	90
memoizee	7,723,320	� 0.54%	94
iMemoized	5,934,041	� 1.03%	90
lru-memoize	5,388,273	� 0.52%	94
memoizerific	3,206,479	� 0.30%	93
addy-osmani	2,397,744	� 0.48%	93
fast-memoize	899,483	� 0.37%	91

Functions with multiple parameters that contain objects...

Name	Ops / sec	Relative margin of error	Sample size
nano-memoize	13,869,690	� 1.25%	86
micro-memoize	13,192,239	� 3.13%	78
moize	10,895,627	� 3.33%	71
memoizee	5,794,981	� 0.83%	92
lru-memoize	5,148,065	� 0.44%	92
memoizerific	3,206,713	� 0.86%	93
addy-osmani	996,705	� 0.45%	92
fast-memoize	699,597	� 1.17%	91

Deep equals ...

Name	Ops / sec	Relative margin of error	Sample size
nano-memoize deep equals (lodash isEqual)	18,024,422	� 1.78%	83
micro-memoize deep equals (lodash isEqual)	17,219,476	� 0.69%	86
nano-memoize deep equals (fast-equals deepEqual)	14,732,731	� 3.10%	85
micro-memoize deep equals (fast-equals deepEqual)	8,785,408	� 11.28%	51
micro-memoize deep equals (hash-it isEqual)	5,744,080	� 10.69%	48

We were puzzled about the multiple argument performance on fast-memoize given its stated goal of being the "fastest possible". We discovered that the default caching and serialization approach used by fast-memoize only performs well for single argument functions for two reasons:

1. It uses JSON.stringify to create a key for an entire argument list. This can be slow.

2. Because a single key is generated for all arguments when perhaps only the first argument differs in a call, a lot of extra work is done. The moize and micro-memoize approach adopted by nano-memoize is far faster for multiple arguments.

Usage

npm install nano-memoize

use the code in the browser directory for the browser

Since most devs are running a build pipeline, the code is not transpiled, although it is browserified

API

The API is a subset of the moize API.

const memoized = micromemoize(sum(a,b) => a + b);
memoized(1,2); // 3
memoized(1,2); // pulled from cache

memoized(function,options) returns function

The shape of options is:

{
  maxArgs: number, // only use the provided maxArgs for cache look-up, useful for ignoring final callback arguments
  maxAge: number, // number of milliseconds to cache a result
  serializer: function, // the serializer/key generator to use for single argument functions (multi-argument functions do not use a serializer)
  equals: function, // the equals function to use for multi-argument functions, e.g. deepEquals for objects (single-argument functions use serializer not equals)
  vargs: boolean // forces the use of multi-argument paradigm, auto set if function has a spread argument or uses `arguments` in its body.
}

# Release History (reverse chronological order)

2018-04-13 v1.0.0 Code style improvements.

2018-02-07 v0.1.2 Documentation updates

2018-02-07 v0.1.1 Documentationand benchmark updates

2018-02-01 v0.1.0 Documentation updates. 50 byte decrease.

2018-01-27 v0.0.7b BETA Documentation updates.

2018-01-27 v0.0.6b BETA Minor size and speed improvements.

2018-01-27 v0.0.5b BETA Fixed edge case where multi-arg key may be shorter than current args.

2018-01-27 v0.0.4b BETA Fixed benchmarks. Removed maxSize. More unit tests. Fixed maxAge.

2018-01-27 v0.0.3b BETA More unit tests. Documentation. Benchmark code in repository not yet running.

2018-01-24 v0.0.2a ALPHA Minor speed enhancements. Benchmark code in repository not yet running.

2018=01-24 v0.0.1a ALPHA First public release. Benchmark code in repository not yet running.