Blazingly fast and type-safe memory comparison utility buffer views and array buffers.
npm install memcmp
Blazingly fast and type-safe memory comparison utility for buffer views and array buffers.
- High Performance
Outperforms popular libraries like dequal, are-deeply-equal, lodash.isEqual and node.isDeepStrictEqual.
- Broad Support
Handles DataView, ArrayBuffer, SharedArrayBuffer, Int8Array, Uint8Array, Uint8ClampedArray, Int16Array, Uint16Array, Int32Array, Uint32Array, Float16Array, Float32Array, Float64Array, BigInt64Array and BigUint64Array.
- Zero Dependencies
Tiny footprint and no external dependencies.
- Cross-Realm Safe
Works reliably across different JS realms (workers, vm contexts, etc).
- Type-Safe
Fully typed with TypeScript declarations.
``bashnpm / node
npm install memcmp
Usage
ts
import memcmp from 'memcmp';type MemcmpInput =
| ArrayBuffer
| SharedArrayBuffer
| TypedArrays
| DataView;
type TypedArrays =
| Int8Array
| Uint8Array
| Uint8ClampedArray
| Int16Array
| Uint16Array
| Int32Array
| Uint32Array
| Float16Array
| Float32Array
| Float64Array
| BigInt64Array
| BigUint64Array;
/**
* Compares two memory buffers or views for binary equality.
*
* Accepts only
ArrayBuffer, SharedArrayBuffer, TypedArray or DataView.
* Optimized for both small and large memory regions, using Buffer.compare if available.
*
* @param target - The first input to compare
* @param source - The second input to compare
* @returns true if both inputs are binary-equal, false otherwise
*/function memcmp(target: MemcmpInput, source: MemcmpInput): boolean
`Example
ts
import memcmp from 'memcmp';const target = new Uint8Array([1, 2, 3, 4]);
const source = new Uint8Array([1, 2, 3, 4]);
console.log(memcmp(target, source)); //true
`Benchmarks
is optimized for raw memory performance across runtimes like Node, Bun, Deno and browsers. It dynamically leverages platform features where available, such as Buffer.compare and gracefully falls back to portable logic elsewhere.$3
memcmp conditionally uses Buffer.compare for large memory regions only when:- The platform exposes
Buffer.compare (e.g. Node, Bun).- And the runtime is not Deno (Deno exposes Buffer, but compare method is slow, equal to basic loop).
When
Buffer.compare is not available or intentionally bypassed (e.g. in Deno, browsers, workers), memcmp falls back to a carefully optimized cross-platform implementation using Int32 byte-aligned traversal for deterministic and consistent results.To fairly reflect this dual-path logic, benchmarks are divided into:
- Benchmark #1: Node or Bun environments using native
Buffer.compare.- Benchmark #2: Deno or browser-like environments using fallback logic.
$3
- Memory Sizes: 16, 512, 4096, 16386 bytes.
- Libraries:
memcmp, dequal, are-deeply-equal, lodash.isEqual, node.isDeepStrictEqual.- Benchmarked with mitata.
$3
> Executed on Node with native Buffer.compare.
#### ArrayBuffer
| Library | 16 | 512 | 4096 | 16386 | Speed Range |
| :--- | :--- | :--- | :--- | :--- | :--- |
| memcmp | 47.36 ns | 115.55 ns | 159.72 ns | 323.23 ns | 1.00x (baseline) |
| dequal | 65.53 ns | 414.22 ns | 3.06 µs | 11.27 µs | 1.38x-34.85x slower |
| are-deeply-equal | 102.55 ns | 380.72 ns | 2.45 µs | 9.33 µs | 2.17x-28.88x slower |
| node.isDeepStrictEqual | 680.80 ns | 551.08 ns | 554.11 ns | 775.81 ns | 14.38x-2.40x slower |
| lodash.isEqual | 42.74 µs | 54.58 µs | 66.62 µs | 107.18 µs | 902.58x-331.59x slower |
Full benchmark results with hardware counters
`console
clk: ~3.81 GHz
cpu: AMD Ryzen 5 3600 6-Core Processor
runtime: node 24.4.1 (x64-linux)benchmark avg (min … max) p75 / p99 (min … top 1%)
------------------------------------------- -------------------------------
• Array Buffer [size=16]
------------------------------------------- -------------------------------
memcmp 47.36 ns/iter 46.83 ns █
(42.82 ns … 197.24 ns) 89.91 ns █
( 90.28 b … 264.21 b) 208.19 b ▅██▂▂▂▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁
4.00 ipc ( 96.79% cache) 0.01 branch misses
189.33 cycles 757.63 instructions 6.90 c-refs 0.22 c-misses
dequal 65.53 ns/iter 64.46 ns █
(62.54 ns … 166.54 ns) 108.72 ns █▆
( 18.60 b … 303.27 b) 208.12 b ██▂▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
3.17 ipc ( 97.42% cache) 0.01 branch misses
250.22 cycles 794.27 instructions 8.86 c-refs 0.23 c-misses
are-deeply-equal 102.55 ns/iter 100.47 ns ▆█
(97.68 ns … 240.58 ns) 161.72 ns ██
(198.46 b … 520.28 b) 392.21 b ██▂▁▁▁▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁
3.00 ipc ( 96.64% cache) 0.02 branch misses
393.93 cycles 1.18k instructions 13.01 c-refs 0.44 c-misses
lodash.isEqual 42.74 µs/iter 43.06 µs ██ █
(41.99 µs … 44.47 µs) 43.49 µs ██▅ ▅▅ ▅█ ▅
( 6.63 kb … 6.70 kb) 6.64 kb ███▁▁▁▁▁▁▁██▁██▁▁▁▁▁█
2.45 ipc ( 98.54% cache) 324.56 branch misses
164.24k cycles 402.85k instructions 18.58k c-refs 272.11 c-misses
node.isDeepStrictEqual 680.80 ns/iter 680.00 ns █
(580.00 ns … 222.94 µs) 1.01 µs ▇ █▂
(904.00 b … 270.20 kb) 917.63 b ▁▅█▇██▃▃▂▁▁▂▁▁▁▁▁▁▁▁▁
1.56 ipc ( 99.68% cache) 30.85 branch misses
4.01k cycles 6.24k instructions 799.47 c-refs 2.52 c-misses
summary
memcmp
1.38x faster than dequal
2.17x faster than are-deeply-equal
14.38x faster than node.isDeepStrictEqual
902.58x faster than lodash.isEqual
• Array Buffer [size=512]
------------------------------------------- -------------------------------
memcmp 115.55 ns/iter 115.08 ns ▃█
(105.75 ns … 275.50 ns) 163.61 ns ██
( 25.77 b … 320.19 b) 208.12 b ▁▄██▇▂▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁
2.93 ipc ( 95.87% cache) 0.01 branch misses
446.36 cycles 1.31k instructions 7.25 c-refs 0.30 c-misses
dequal 414.22 ns/iter 419.86 ns █▆
(398.16 ns … 608.65 ns) 548.91 ns ██
( 30.13 b … 366.95 b) 208.05 b ██▅█▆▄▁▁▁▂▁▁▁▁▁▁▁▁▁▁▁
5.15 ipc ( 96.77% cache) 1.02 branch misses
1.67k cycles 8.61k instructions 9.12 c-refs 0.29 c-misses
are-deeply-equal 380.72 ns/iter 376.24 ns █
(371.01 ns … 558.68 ns) 498.12 ns █▂
(160.70 b … 526.31 b) 392.18 b ██▂▃▃▁▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁
5.15 ipc ( 96.19% cache) 1.02 branch misses
1.55k cycles 8.00k instructions 13.29 c-refs 0.51 c-misses
lodash.isEqual 54.58 µs/iter 54.84 µs █ █
(52.99 µs … 56.61 µs) 55.71 µs ▅ ▅ █▅█▅ ▅ ▅ ▅
( 6.63 kb … 6.69 kb) 6.64 kb █▁▁▁▁▁█▁████▁▁█▁▁█▁▁█
2.77 ipc ( 98.07% cache) 340.71 branch misses
209.99k cycles 582.22k instructions 19.16k c-refs 369.58 c-misses
node.isDeepStrictEqual 551.08 ns/iter 556.10 ns █
(503.74 ns … 729.40 ns) 658.34 ns ▇█
(351.75 b … 679.33 b) 544.11 b ▃▃▂▂███▇▃▂▂▄▃▂▂▁▁▁▁▁▁
2.46 ipc ( 99.48% cache) 0.10 branch misses
2.11k cycles 5.19k instructions 173.57 c-refs 0.91 c-misses
summary
memcmp
3.29x faster than are-deeply-equal
3.58x faster than dequal
4.77x faster than node.isDeepStrictEqual
472.37x faster than lodash.isEqual
• Array Buffer [size=4096]
------------------------------------------- -------------------------------
memcmp 159.72 ns/iter 159.07 ns █▃
(149.84 ns … 236.87 ns) 211.93 ns ▃██
( 33.96 b … 330.37 b) 208.12 b ▂███▅▂▁▁▁▁▁▁▁▁▁▁▁▁▂▂▁
2.92 ipc ( 99.30% cache) 0.02 branch misses
610.80 cycles 1.78k instructions 46.94 c-refs 0.33 c-misses
dequal 3.06 µs/iter 3.02 µs █
(2.98 µs … 3.50 µs) 3.46 µs █▆
(199.49 b … 208.45 b) 207.88 b ██▄▁▂▁▁▁▁▃▂▂▁▁▂▁▂▂▂▁▂
5.49 ipc ( 94.61% cache) 1.04 branch misses
11.86k cycles 65.08k instructions 11.00 c-refs 0.59 c-misses
are-deeply-equal 2.45 µs/iter 2.51 µs █
(2.35 µs … 2.73 µs) 2.65 µs ▄█
(383.61 b … 392.54 b) 392.01 b ██▃▃▃▃▂▁▂▇▇▇▅▅▁▁▁▂▅▂▂
5.84 ipc ( 98.07% cache) 1.05 branch misses
9.82k cycles 57.30k instructions 43.20 c-refs 0.83 c-misses
lodash.isEqual 66.62 µs/iter 66.04 µs █
(58.88 µs … 197.96 µs) 93.82 µs █
( 1.71 kb … 555.18 kb) 8.06 kb ▂▁▂▆█▂▂▂▁▁▁▁▁▁▁▁▁▁▁▁▁
2.97 ipc ( 94.73% cache) 454.23 branch misses
271.93k cycles 807.25k instructions 22.39k c-refs 1.18k c-misses
node.isDeepStrictEqual 554.11 ns/iter 555.84 ns ▇█
(522.44 ns … 747.77 ns) 674.82 ns ▅██▂
(217.23 b … 736.80 b) 544.25 b ▂▅████▄▂▁▂▂▂▂▁▁▁▁▁▁▁▁
2.51 ipc ( 99.54% cache) 0.10 branch misses
2.25k cycles 5.66k instructions 219.20 c-refs 1.01 c-misses
summary
memcmp
3.47x faster than node.isDeepStrictEqual
15.33x faster than are-deeply-equal
19.16x faster than dequal
417.13x faster than lodash.isEqual
• Array Buffer [size=16384]
------------------------------------------- -------------------------------
memcmp 323.23 ns/iter 323.69 ns ▄█
(314.48 ns … 406.29 ns) 371.75 ns ██▅
(112.00 b … 318.11 b) 208.10 b ▆████▅▂▁▁▁▁▁▂▂▁▁▁▁▁▁▂
2.61 ipc ( 99.90% cache) 1.02 branch misses
1.31k cycles 3.41k instructions 594.53 c-refs 0.60 c-misses
dequal 11.27 µs/iter 11.36 µs ██
(11.11 µs … 11.56 µs) 11.50 µs ██ ▅▅ ▅ ▅ ▅ ▅ ▅
(202.26 b … 208.45 b) 207.80 b ██▁██▁█▁█▁▁▁▁█▁█▁▁▁▁█
5.60 ipc ( 99.71% cache) 1.10 branch misses
46.18k cycles 258.68k instructions 411.99 c-refs 1.19 c-misses
are-deeply-equal 9.33 µs/iter 9.30 µs █
(9.18 µs … 10.14 µs) 9.59 µs ▅█ ▅
(383.61 b … 392.54 b) 391.50 b ▇██▇▁▇█▁▁▇▁▇▁▁▁▁▁▁▁▁▇
5.96 ipc ( 99.56% cache) 1.12 branch misses
37.97k cycles 226.32k instructions 493.48 c-refs 2.17 c-misses
lodash.isEqual 107.18 µs/iter 108.53 µs █
(96.11 µs … 327.83 µs) 169.19 µs █▆
( 1.45 kb … 345.82 kb) 8.19 kb ▅███▄▃▁▁▂▂▁▁▁▁▁▁▁▁▁▁▁
3.55 ipc ( 94.50% cache) 472.58 branch misses
426.17k cycles 1.51M instructions 23.05k c-refs 1.27k c-misses
node.isDeepStrictEqual 775.81 ns/iter 769.47 ns █
(737.93 ns … 1.08 µs) 970.47 ns ▆█
(383.70 b … 695.94 b) 544.09 b ▄██▃▂▄▃▃▁▁▂▁▂▁▁▁▁▁▁▁▁
2.35 ipc ( 99.84% cache) 1.11 branch misses
3.11k cycles 7.30k instructions 817.94 c-refs 1.32 c-misses
summary
memcmp
2.4x faster than node.isDeepStrictEqual
28.88x faster than are-deeply-equal
34.85x faster than dequal
331.59x faster than lodash.isEqual
`#### Typed Array
| Library | 16 | 512 | 4096 | 16386 | Speed Range |
| :--- | :--- | :--- | :--- | :--- | :--- |
| memcmp | 15.87 ns | 65.88 ns | 109.10 ns | 303.78 ns | 1.00x (baseline) |
| dequal | 19.13 ns | 364.92 ns | 2.97 µs | 11.93 µs | 1.21x-39.26x slower |
| are-deeply-equal | 45.53 ns | 330.07 ns | 2.49 µs | 9.71 µs | 2.87x-31.97x slower |
| node.isDeepStrictEqual | 590.29 ns | 618.08 ns | 638.77 ns | 927.69 ns | 37.19x-3.05x slower |
| lodash.isEqual | 21.17 µs | 29.18 µs | 44.35 µs | 85.50 µs | 1333.48x-281.45x slower |
Full benchmark results with hardware counters
console
clk: ~3.76 GHz
cpu: AMD Ryzen 5 3600 6-Core Processor
runtime: node 24.4.1 (x64-linux)benchmark avg (min … max) p75 / p99 (min … top 1%)
------------------------------------------- -------------------------------
• Typed Array [size=16]
------------------------------------------- -------------------------------
memcmp 15.87 ns/iter 15.80 ns █
(14.12 ns … 122.86 ns) 21.52 ns █
( 0.09 b … 47.09 b) 0.18 b ▁▂▁█▇▆▂▁▁▁▁▂▂▁▁▁▁▁▁▁▁
5.24 ipc ( 93.11% cache) 0.01 branch misses
60.41 cycles 316.60 instructions 0.03 c-refs 0.00 c-misses
dequal 19.13 ns/iter 19.06 ns █
(17.77 ns … 70.42 ns) 28.14 ns █
( 0.09 b … 52.21 b) 0.19 b ▃▇█▂▃▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
5.13 ipc ( 94.35% cache) 0.01 branch misses
74.99 cycles 384.94 instructions 0.03 c-refs 0.00 c-misses
are-deeply-equal 45.53 ns/iter 44.34 ns █
(42.63 ns … 190.40 ns) 91.57 ns █▂
( 34.80 b … 312.22 b) 184.22 b ██▁▂▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
4.02 ipc ( 96.68% cache) 0.01 branch misses
173.15 cycles 695.42 instructions 6.12 c-refs 0.20 c-misses
lodash.isEqual 21.17 µs/iter 21.95 µs █ █
(19.74 µs … 22.24 µs) 22.13 µs █ ▅ ▅ ▅ ▅ ▅▅ █▅
( 4.02 kb … 4.08 kb) 4.03 kb █▁▁█▁█▁▁▁▁█▁▁█▁▁██▁██
2.36 ipc ( 99.13% cache) 146.51 branch misses
82.09k cycles 194.13k instructions 10.99k c-refs 95.82 c-misses
node.isDeepStrictEqual 590.29 ns/iter 599.56 ns █
(555.68 ns … 897.92 ns) 799.37 ns █ ▄
(328.24 b … 623.91 b) 480.16 b ▇████▅▄▂▁▁▁▁▁▁▁▁▁▁▁▁▁
2.41 ipc ( 99.61% cache) 0.10 branch misses
2.37k cycles 5.70k instructions 220.82 c-refs 0.87 c-misses
summary
memcmp
1.21x faster than dequal
2.87x faster than are-deeply-equal
37.19x faster than node.isDeepStrictEqual
1333.48x faster than lodash.isEqual
• Typed Array [size=512]
------------------------------------------- -------------------------------
memcmp 65.88 ns/iter 66.63 ns █
(60.45 ns … 238.08 ns) 109.20 ns █
( 0.09 b … 91.70 b) 0.39 b ██▇█▂▁▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁
3.21 ipc ( 93.41% cache) 0.01 branch misses
267.11 cycles 857.64 instructions 0.16 c-refs 0.01 c-misses
dequal 364.92 ns/iter 359.87 ns █▄
(355.05 ns … 549.41 ns) 483.36 ns ██
( 0.09 b … 155.21 b) 0.75 b ██▂▃▂▁▁▁▃▁▁▁▁▁▁▁▁▁▁▁▁
5.53 ipc ( 84.14% cache) 1.01 branch misses
1.48k cycles 8.20k instructions 0.13 c-refs 0.02 c-misses
are-deeply-equal 330.07 ns/iter 325.21 ns █
(320.24 ns … 531.38 ns) 441.34 ns ██
( 95.45 b … 325.30 b) 184.60 b ██▂▂▄▁▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁
5.58 ipc ( 96.06% cache) 1.02 branch misses
1.35k cycles 7.52k instructions 6.33 c-refs 0.25 c-misses
lodash.isEqual 29.18 µs/iter 29.21 µs ██ █ █
(28.74 µs … 30.16 µs) 29.90 µs ██ █ ▅▅ ▅ █
( 4.02 kb … 4.08 kb) 4.04 kb ██▁█▁██▁█▁▁▁▁▁▁▁▁▁▁▁█
2.83 ipc ( 98.88% cache) 170.85 branch misses
119.05k cycles 336.81k instructions 11.48k c-refs 128.34 c-misses
node.isDeepStrictEqual 618.08 ns/iter 616.86 ns █
(575.00 ns … 973.21 ns) 779.53 ns ▂█▄
(370.90 b … 631.91 b) 480.11 b ▂▃███▃▂▃▂▁▁▂▁▁▁▁▁▁▁▁▁
2.42 ipc ( 99.61% cache) 0.10 branch misses
2.37k cycles 5.74k instructions 232.74 c-refs 0.92 c-misses
summary
memcmp
5.01x faster than are-deeply-equal
5.54x faster than dequal
9.38x faster than node.isDeepStrictEqual
443x faster than lodash.isEqual
• Typed Array [size=4096]
------------------------------------------- -------------------------------
memcmp 109.10 ns/iter 108.96 ns ▅█
(105.91 ns … 165.86 ns) 135.53 ns ██
( 0.09 b … 112.14 b) 0.43 b ▁██▅▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
3.15 ipc ( 96.77% cache) 0.01 branch misses
421.25 cycles 1.33k instructions 0.39 c-refs 0.01 c-misses
dequal 2.97 µs/iter 2.96 µs █▃
(2.93 µs … 3.29 µs) 3.19 µs ▅██
( 0.09 b … 0.42 b) 0.10 b ███▄▁▁▁▂▁▂▁▁▁▁▁▁▂▁▁▁▂
5.62 ipc ( 84.95% cache) 1.03 branch misses
11.50k cycles 64.67k instructions 1.09 c-refs 0.16 c-misses
are-deeply-equal 2.49 µs/iter 2.48 µs █
(2.43 µs … 3.09 µs) 2.89 µs █ ▇
(175.46 b … 184.48 b) 183.98 b █▅█▄▂▂▁▁▁▁▁▁▁▁▁▁▁▂▂▁▂
5.95 ipc ( 94.93% cache) 1.04 branch misses
9.55k cycles 56.81k instructions 8.36 c-refs 0.42 c-misses
lodash.isEqual 44.35 µs/iter 43.87 µs █
(40.57 µs … 278.48 µs) 63.09 µs █
( 4.23 kb … 388.64 kb) 4.95 kb ▂▂▁█▁▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁▁
3.21 ipc ( 95.69% cache) 232.17 branch misses
172.24k cycles 552.05k instructions 13.74k c-refs 591.47 c-misses
node.isDeepStrictEqual 638.77 ns/iter 647.46 ns █ ▃
(607.35 ns … 848.14 ns) 716.97 ns █ ▃▇█▃
(363.08 b … 606.23 b) 480.10 b ▇██▅▄████▇▂▂▂▁▅▄▁▁▂▁▁
2.47 ipc ( 99.66% cache) 0.09 branch misses
2.52k cycles 6.22k instructions 245.60 c-refs 0.84 c-misses
summary
memcmp
5.85x faster than node.isDeepStrictEqual
22.78x faster than are-deeply-equal
27.2x faster than dequal
406.5x faster than lodash.isEqual
• Typed Array [size=16384]
------------------------------------------- -------------------------------
memcmp 303.78 ns/iter 295.27 ns █
(292.76 ns … 493.91 ns) 455.10 ns █
( 0.09 b … 133.27 b) 0.60 b █▂▁▁▁▁▁▁▂▂▁▁▁▁▁▁▁▁▁▁▁
2.58 ipc ( 99.99% cache) 1.01 branch misses
1.15k cycles 2.97k instructions 557.46 c-refs 0.03 c-misses
dequal 11.93 µs/iter 11.89 µs ▂ █
(11.85 µs … 12.28 µs) 12.07 µs █ █
( 0.10 b … 0.41 b) 0.12 b █▁█▁▇▇▁▁▁▁▁▁▁▁▁▁▁▁▁▁▇
5.66 ipc ( 99.83% cache) 1.09 branch misses
45.62k cycles 258.27k instructions 288.99 c-refs 0.50 c-misses
are-deeply-equal 9.71 µs/iter 9.73 µs █
(9.63 µs … 9.90 µs) 9.86 µs ██
(178.89 b … 184.48 b) 183.88 b ███▁███▁▁██▁▁▁▁█▁▁▁▁█
6.00 ipc ( 99.64% cache) 1.10 branch misses
37.62k cycles 225.84k instructions 427.11 c-refs 1.54 c-misses
lodash.isEqual 85.50 µs/iter 84.14 µs █
(80.64 µs … 294.13 µs) 133.06 µs █
( 4.41 kb … 414.78 kb) 5.54 kb ▇█▃▃▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
3.83 ipc ( 93.84% cache) 233.22 branch misses
330.10k cycles 1.26M instructions 14.86k c-refs 915.34 c-misses
node.isDeepStrictEqual 927.69 ns/iter 927.93 ns █▅▅
(897.04 ns … 1.11 µs) 1.06 µs ███
(156.77 b … 795.08 b) 480.09 b ▇███▄▄▃▄▃▇▃▂▂▂▁▂▂▁▁▁▂
2.20 ipc ( 99.87% cache) 1.11 branch misses
3.56k cycles 7.85k instructions 976.88 c-refs 1.31 c-misses
summary
memcmp
3.05x faster than node.isDeepStrictEqual
31.97x faster than are-deeply-equal
39.26x faster than dequal
281.45x faster than lodash.isEqual
`#### Data View
| Library | 16 | 512 | 4096 | 16386 | Speed Range |
| :--- | :--- | :--- | :--- | :--- | :--- |
| memcmp | 8.76 ns | 121.58 ns | 161.15 ns | 350.18 ns | 1.00x (baseline) |
| dequal | 19.02 ns | 368.03 ns | 2.98 µs | 11.20 µs | 2.17x-31.98x slower |
| are-deeply-equal | 94.91 ns | 381.76 ns | 2.47 µs | 9.36 µs | 10.83x-26.73x slower |
| node.isDeepStrictEqual | 683.29 ns | 530.72 ns | 602.79 ns | 790.50 ns | 78.00x-2.26x slower |
| lodash.isEqual | 46.55 µs | 54.49 µs | 68.09 µs | 105.71 µs | 5314.08x-301.87x slower |
Full benchmark results with hardware counters
console
clk: ~3.75 GHz
cpu: AMD Ryzen 5 3600 6-Core Processor
runtime: node 24.4.1 (x64-linux)benchmark avg (min … max) p75 / p99 (min … top 1%)
------------------------------------------- -------------------------------
• Data View [size=16]
------------------------------------------- -------------------------------
memcmp 8.76 ns/iter 8.62 ns █
(8.14 ns … 63.38 ns) 11.35 ns █▃
( 0.09 b … 34.83 b) 0.15 b ▁▂██▃▃▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁
4.75 ipc ( 96.93% cache) 0.01 branch misses
35.69 cycles 169.38 instructions 0.02 c-refs 0.00 c-misses
dequal 19.02 ns/iter 18.34 ns █
(18.27 ns … 65.38 ns) 28.65 ns █
( 0.09 b … 52.20 b) 0.18 b █▁▃▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▂▁▁
4.68 ipc ( 94.36% cache) 0.01 branch misses
73.17 cycles 342.68 instructions 0.03 c-refs 0.00 c-misses
are-deeply-equal 94.91 ns/iter 94.63 ns █
(84.98 ns … 261.74 ns) 141.62 ns ▇█▂
(210.01 b … 520.28 b) 392.21 b ▁████▂▁▁▁▁▁▁▁▁▁▁▁▁▁▂▁
3.37 ipc ( 96.60% cache) 0.02 branch misses
366.35 cycles 1.24k instructions 12.98 c-refs 0.44 c-misses
lodash.isEqual 46.55 µs/iter 46.28 µs ▂█
(29.71 µs … 322.04 µs) 64.18 µs ███
( 1.70 kb … 644.69 kb) 7.89 kb ▁▁▁▁▂▁▁▁███▂▂▂▂▂▁▁▁▁▁
2.40 ipc ( 96.38% cache) 430.61 branch misses
181.36k cycles 435.36k instructions 21.10k c-refs 764.03 c-misses
node.isDeepStrictEqual 683.29 ns/iter 680.00 ns █
(590.00 ns … 210.21 µs) 1.01 µs █
(840.00 b … 192.94 kb) 853.06 b ▁▆██▆▆▄▂▁▁▁▁▁▂▂▂▁▁▁▁▁
1.57 ipc ( 99.73% cache) 30.94 branch misses
4.10k cycles 6.43k instructions 817.60 c-refs 2.19 c-misses
summary
memcmp
2.17x faster than dequal
10.83x faster than are-deeply-equal
78x faster than node.isDeepStrictEqual
5314.08x faster than lodash.isEqual
• Data View [size=512]
------------------------------------------- -------------------------------
memcmp 121.58 ns/iter 120.58 ns █
(114.60 ns … 317.36 ns) 174.88 ns ██
( 22.56 b … 312.18 b) 208.11 b ▂██▄▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
2.83 ipc ( 96.66% cache) 0.01 branch misses
494.63 cycles 1.40k instructions 9.25 c-refs 0.31 c-misses
dequal 368.03 ns/iter 359.54 ns █
(354.30 ns … 551.65 ns) 522.78 ns █
( 0.09 b … 142.55 b) 0.72 b █▇▃▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▂
4.76 ipc ( 80.50% cache) 1.01 branch misses
1.51k cycles 7.17k instructions 0.14 c-refs 0.03 c-misses
are-deeply-equal 381.76 ns/iter 374.52 ns █
(365.35 ns … 577.26 ns) 570.95 ns █▆
( 89.60 b … 526.31 b) 392.18 b ██▅▂▁▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▂
5.17 ipc ( 95.95% cache) 1.02 branch misses
1.56k cycles 8.05k instructions 13.39 c-refs 0.54 c-misses
lodash.isEqual 54.49 µs/iter 55.18 µs █
(51.89 µs … 58.24 µs) 58.08 µs █ █
( 6.61 kb … 6.67 kb) 6.62 kb ██▁▁█▁█▁▁█▁██▁▁▁▁▁▁▁█
2.78 ipc ( 98.48% cache) 368.87 branch misses
210.80k cycles 585.67k instructions 19.45k c-refs 294.90 c-misses
node.isDeepStrictEqual 530.72 ns/iter 531.26 ns ▂█▆
(501.46 ns … 710.89 ns) 676.25 ns ███
(293.74 b … 615.29 b) 480.12 b ▄████▅▂▃▁▃▂▁▂▁▁▁▁▂▁▁▂
2.46 ipc ( 99.37% cache) 0.10 branch misses
2.16k cycles 5.33k instructions 141.99 c-refs 0.89 c-misses
summary
memcmp
3.03x faster than dequal
3.14x faster than are-deeply-equal
4.37x faster than node.isDeepStrictEqual
448.15x faster than lodash.isEqual
• Data View [size=4096]
------------------------------------------- -------------------------------
memcmp 161.15 ns/iter 159.42 ns ▅█
(153.08 ns … 239.88 ns) 208.12 ns ██
( 31.98 b … 308.80 b) 208.11 b ▃██▅▂▂▁▁▁▁▁▂▂▁▁▁▁▁▁▁▁
3.07 ipc ( 96.39% cache) 0.01 branch misses
612.14 cycles 1.88k instructions 9.70 c-refs 0.35 c-misses
dequal 2.98 µs/iter 2.99 µs █
(2.93 µs … 3.44 µs) 3.43 µs █
( 0.09 b … 0.41 b) 0.10 b ███▃▂▁▁▁▂▁▁▁▁▁▁▁▂▁▁▁▂
4.90 ipc ( 82.40% cache) 1.03 branch misses
11.52k cycles 56.47k instructions 1.09 c-refs 0.19 c-misses
are-deeply-equal 2.47 µs/iter 2.50 µs █▅
(2.35 µs … 3.00 µs) 2.86 µs ██
(383.61 b … 392.54 b) 392.01 b █▆▄▃▃██▂▂▂▁▁▁▁▁▁▁▁▁▁▂
5.87 ipc ( 97.54% cache) 1.04 branch misses
9.77k cycles 57.35k instructions 30.88 c-refs 0.76 c-misses
lodash.isEqual 68.09 µs/iter 68.58 µs █
(58.92 µs … 307.73 µs) 97.11 µs ▂█
( 1.44 kb … 298.38 kb) 8.01 kb ▃▁▂███▆▂▂▁▁▂▁▁▁▁▁▁▁▁▁
2.98 ipc ( 95.40% cache) 484.65 branch misses
271.25k cycles 809.01k instructions 22.22k c-refs 1.02k c-misses
node.isDeepStrictEqual 602.79 ns/iter 605.03 ns ▂█
(561.62 ns … 848.40 ns) 757.89 ns ██
(327.73 b … 623.91 b) 480.11 b ▁▄██▇▇▃▂▂▃▂▁▂▁▁▁▁▁▁▂▁
2.52 ipc ( 99.39% cache) 0.10 branch misses
2.31k cycles 5.81k instructions 161.42 c-refs 0.99 c-misses
summary
memcmp
3.74x faster than node.isDeepStrictEqual
15.36x faster than are-deeply-equal
18.5x faster than dequal
422.55x faster than lodash.isEqual
• Data View [size=16384]
------------------------------------------- -------------------------------
memcmp 350.18 ns/iter 347.64 ns █
(342.47 ns … 469.80 ns) 405.69 ns █
( 77.99 b … 334.24 b) 208.10 b ▄█▇▂▁▁▂▂▂▁▁▁▁▁▁▁▁▁▁▁▁
2.59 ipc ( 99.90% cache) 1.02 branch misses
1.35k cycles 3.51k instructions 615.09 c-refs 0.62 c-misses
dequal 11.20 µs/iter 11.19 µs █
(11.11 µs … 11.78 µs) 11.23 µs █ █
( 0.10 b … 0.41 b) 0.12 b ███▁██▁▁▁▁▁▁▁██▁▁▁▁▁█
4.91 ipc ( 99.87% cache) 1.09 branch misses
45.92k cycles 225.50k instructions 307.93 c-refs 0.41 c-misses
are-deeply-equal 9.36 µs/iter 9.32 µs █
(9.22 µs … 10.07 µs) 9.57 µs ████ █
(383.61 b … 392.54 b) 391.50 b ████▁██▁▁▁▁█▁▁▁▁▁▁▁▁█
5.86 ipc ( 99.60% cache) 1.11 branch misses
38.62k cycles 226.37k instructions 488.52 c-refs 1.93 c-misses
lodash.isEqual 105.71 µs/iter 104.40 µs █
(96.45 µs … 278.26 µs) 157.99 µs █
( 6.65 kb … 239.14 kb) 7.76 kb ▃▆█▆▅▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
3.51 ipc ( 94.69% cache) 488.08 branch misses
431.88k cycles 1.52M instructions 23.50k c-refs 1.25k c-misses
node.isDeepStrictEqual 790.50 ns/iter 793.06 ns █▃
(761.07 ns … 976.25 ns) 861.84 ns ▃██
(274.62 b … 676.86 b) 480.03 b ▄▄▅████▄█▅▁▂▂▃▃▁▁▁▁▁▁
2.31 ipc ( 99.86% cache) 1.10 branch misses
3.23k cycles 7.44k instructions 847.95 c-refs 1.21 c-misses
summary
memcmp
2.26x faster than node.isDeepStrictEqual
26.73x faster than are-deeply-equal
31.98x faster than dequal
301.87x faster than lodash.isEqual
`$3
> Executed on Deno without native Buffer.compare.
#### Array Buffer
| Library | 16 | 512 | 4096 | 16386 | Speed Range |
| :--- | :--- | :--- | :--- | :--- | :--- |
| memcmp | 43.66 ns | 187.17 ns | 677.76 ns | 2.54 µs | 1.00x (baseline) |
| dequal | 47.27 ns | 416.88 ns | 2.90 µs | 11.89 µs | 1.08x-4.67x slower |
| are-deeply-equal | 83.19 ns | 402.53 ns | 2.54 µs | 9.28 µs | 1.91x-3.65x slower |
| lodash.isEqual | 58.53 µs | 63.03 µs | 76.30 µs | 109.57 µs | 1340.66x-43.06x slower |
Full benchmark results with hardware counters
console
clk: ~3.62 GHz
cpu: AMD Ryzen 5 3600 6-Core Processor
runtime: deno 2.4.2 (x86_64-unknown-linux-gnu)benchmark avg (min … max) p75 / p99 (min … top 1%)
------------------------------------------- -------------------------------
• Array Buffer [size=16]
------------------------------------------- -------------------------------
memcmp 43.66 ns/iter 42.14 ns █▄
(40.32 ns … 186.30 ns) 73.58 ns ██
( 73.90 b … 328.25 b) 208.15 b ██▄▂▁▁▁▁▁▁▁▂▂▂▁▁▁▁▁▁▁
4.13 ipc ( 95.82% cache) 0.02 branch misses
174.95 cycles 722.42 instructions 7.05 c-refs 0.29 c-misses
dequal 47.27 ns/iter 45.46 ns █
(42.07 ns … 161.52 ns) 126.82 ns █
( 65.72 b … 304.18 b) 208.18 b ██▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
4.01 ipc ( 96.27% cache) 0.02 branch misses
189.94 cycles 760.74 instructions 7.04 c-refs 0.26 c-misses
are-deeply-equal 83.19 ns/iter 81.37 ns ▄█
(74.71 ns … 255.96 ns) 135.68 ns ▇██
(202.30 b … 544.27 b) 392.15 b ███▃▂▂▄▃▁▁▁▂▂▂▂▁▁▁▁▁▁
3.53 ipc ( 96.18% cache) 0.03 branch misses
325.91 cycles 1.15k instructions 13.23 c-refs 0.50 c-misses
lodash.isEqual 58.53 µs/iter 60.78 µs █
(41.17 µs … 273.85 µs) 84.11 µs █ ▂
( 1.21 kb … 697.31 kb) 10.51 kb ▁▁▂▁▁▃▃█▂█▅▂▂▁▁▁▁▁▁▁▁
2.58 ipc ( 92.73% cache) 500.00 branch misses
225.66k cycles 581.16k instructions 23.53k c-refs 1.71k c-misses
summary
memcmp
1.08x faster than dequal
1.91x faster than are-deeply-equal
1340.66x faster than lodash.isEqual
• Array Buffer [size=512]
------------------------------------------- -------------------------------
memcmp 187.17 ns/iter 180.95 ns █
(177.14 ns … 374.42 ns) 362.40 ns █
(298.73 b … 563.28 b) 416.15 b █▄▁▁▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁
3.95 ipc ( 96.36% cache) 0.04 branch misses
709.94 cycles 2.81k instructions 14.06 c-refs 0.51 c-misses
dequal 416.88 ns/iter 407.96 ns █
(403.97 ns … 604.45 ns) 587.49 ns █
(152.53 b … 349.36 b) 208.63 b █▅▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▂
4.69 ipc ( 95.31% cache) 1.03 branch misses
1.63k cycles 7.62k instructions 7.70 c-refs 0.36 c-misses
are-deeply-equal 402.53 ns/iter 397.63 ns █
(386.85 ns … 543.70 ns) 520.15 ns █
(253.95 b … 520.76 b) 392.17 b ▁█▇▃▃▁▁▁▁▂▁▁▁▁▁▁▁▁▁▁▁
4.50 ipc ( 95.97% cache) 1.02 branch misses
1.55k cycles 6.97k instructions 13.39 c-refs 0.54 c-misses
lodash.isEqual 63.03 µs/iter 63.89 µs █
(59.26 µs … 64.70 µs) 64.37 µs ▅ ▅ ▅▅▅ ▅█▅▅▅
( 1.04 kb … 8.78 kb) 7.48 kb █▁▁▁▁▁▁▁▁▁█▁███▁█████
2.90 ipc ( 97.48% cache) 401.09 branch misses
242.84k cycles 703.04k instructions 21.12k c-refs 533.08 c-misses
summary
memcmp
2.15x faster than are-deeply-equal
2.23x faster than dequal
336.73x faster than lodash.isEqual
• Array Buffer [size=4096]
------------------------------------------- -------------------------------
memcmp 677.76 ns/iter 679.80 ns █
(665.55 ns … 810.88 ns) 779.01 ns █
(238.74 b … 585.10 b) 416.11 b ▄█▃▄▅▂▁▁▁▁▂▁▁▁▁▁▁▁▁▁▁
4.79 ipc ( 95.31% cache) 1.06 branch misses
2.78k cycles 13.34k instructions 15.26 c-refs 0.72 c-misses
dequal 2.90 µs/iter 2.84 µs █
(2.78 µs … 3.69 µs) 3.65 µs ▆█
(199.63 b … 208.16 b) 207.88 b ██▃▂▃▁▂▂▁▁▁▂▃▁▁▁▂▁▁▁▂
4.75 ipc ( 91.42% cache) 1.18 branch misses
12.02k cycles 57.10k instructions 12.06 c-refs 1.03 c-misses
are-deeply-equal 2.54 µs/iter 2.53 µs █
(2.46 µs … 2.99 µs) 2.94 µs █
(383.73 b … 392.23 b) 391.99 b █▄▄▅▂▂▂▁▁▂▁▁▂▁▁▂▁▂▃▁▂
5.04 ipc ( 93.25% cache) 1.18 branch misses
9.77k cycles 49.28k instructions 18.96 c-refs 1.28 c-misses
lodash.isEqual 76.30 µs/iter 75.56 µs █
(68.80 µs … 304.63 µs) 109.72 µs █
( 1.20 kb … 305.34 kb) 9.98 kb ▃▁▄█▃▁▃▂▁▁▁▁▁▁▁▁▁▁▁▁▁
3.05 ipc ( 93.06% cache) 510.16 branch misses
299.41k cycles 911.75k instructions 23.93k c-refs 1.66k c-misses
summary
memcmp
3.75x faster than are-deeply-equal
4.28x faster than dequal
112.58x faster than lodash.isEqual
• Array Buffer [size=16384]
------------------------------------------- -------------------------------
memcmp 2.54 µs/iter 2.53 µs ▇█
(2.51 µs … 2.94 µs) 2.80 µs ███
(407.66 b … 416.22 b) 415.94 b ███▁▂▂▁▁▂▁▁▂▁▁▁▁▁▁▂▁▂
4.98 ipc ( 99.70% cache) 1.17 branch misses
9.96k cycles 49.61k instructions 489.48 c-refs 1.46 c-misses
dequal 11.89 µs/iter 11.85 µs █
(11.82 µs … 12.41 µs) 11.86 µs █▅▅▅ ▅▅ ▅ ▅▅
(208.14 b … 208.16 b) 208.15 b ████▁▁██▁▁▁▁▁▁▁█▁▁▁██
4.86 ipc ( 99.40% cache) 1.47 branch misses
46.72k cycles 227.12k instructions 414.07 c-refs 2.50 c-misses
are-deeply-equal 9.28 µs/iter 9.24 µs █
(9.17 µs … 10.14 µs) 9.37 µs ▂▂█
(386.39 b … 392.23 b) 391.89 b ███▁▁▁▁▆▆▁▁▁▆▁▁▆▁▁▁▁▆
5.06 ipc ( 99.41% cache) 1.41 branch misses
38.33k cycles 194.08k instructions 505.51 c-refs 2.96 c-misses
lodash.isEqual 109.57 µs/iter 107.88 µs █
(100.22 µs … 242.30 µs) 177.47 µs █
( 1.26 kb … 319.86 kb) 10.52 kb ▃██▅▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
3.50 ipc ( 91.78% cache) 512.00 branch misses
449.07k cycles 1.57M instructions 25.66k c-refs 2.11k c-misses
summary
memcmp
3.65x faster than are-deeply-equal
4.67x faster than dequal
43.06x faster than lodash.isEqual
`#### Typed Array
| Library | 16 | 512 | 4096 | 16386 | Speed Range |
| :--- | :--- | :--- | :--- | :--- | :--- |
| memcmp | 15.17 ns | 147.83 ns | 685.25 ns | 2.48 µs | 1.00x (baseline) |
| dequal | 19.27 ns | 371.29 ns | 2.96 µs | 11.64 µs | 1.27x-4.70x slower |
| are-deeply-equal | 47.72 ns | 344.28 ns | 2.38 µs | 9.27 µs | 3.15x-3.74x slower |
| lodash.isEqual | 31.40 µs | 36.56 µs | 47.46 µs | 89.55 µs | 2070.25x-36.12x slower |
Full benchmark results with hardware counters
console
clk: ~3.52 GHz
cpu: AMD Ryzen 5 3600 6-Core Processor
runtime: deno 2.4.2 (x86_64-unknown-linux-gnu)benchmark avg (min … max) p75 / p99 (min … top 1%)
------------------------------------------- -------------------------------
• Typed Array [size=16]
------------------------------------------- -------------------------------
memcmp 15.17 ns/iter 15.37 ns █
(13.78 ns … 95.93 ns) 22.28 ns █
( 0.09 b … 56.57 b) 0.19 b ▃▂█▁▇▂▁▁▁▁▂▂▁▁▁▁▁▁▁▁▁
4.74 ipc ( 97.60% cache) 0.01 branch misses
60.59 cycles 287.10 instructions 0.09 c-refs 0.00 c-misses
dequal 19.27 ns/iter 18.73 ns █
(18.65 ns … 97.37 ns) 29.79 ns █
( 0.09 b … 60.25 b) 0.18 b █▄▃▁▁▁▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁
4.76 ipc ( 93.96% cache) 0.01 branch misses
74.67 cycles 355.50 instructions 0.10 c-refs 0.01 c-misses
are-deeply-equal 47.72 ns/iter 45.31 ns █
(42.35 ns … 184.89 ns) 129.82 ns ▄█
( 2.23 b … 360.23 b) 184.24 b ██▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
3.64 ipc ( 96.42% cache) 0.02 branch misses
181.14 cycles 659.43 instructions 6.30 c-refs 0.23 c-misses
lodash.isEqual 31.40 µs/iter 30.98 µs █▅
(25.66 µs … 246.84 µs) 44.51 µs ██
( 56.00 b … 705.63 kb) 6.42 kb ▁▁▂▄▂██▁▁▁▁▂▂▂▁▁▁▁▁▁▁
2.64 ipc ( 94.63% cache) 267.40 branch misses
124.77k cycles 329.84k instructions 14.11k c-refs 757.30 c-misses
summary
memcmp
1.27x faster than dequal
3.15x faster than are-deeply-equal
2070.25x faster than lodash.isEqual
• Typed Array [size=512]
------------------------------------------- -------------------------------
memcmp 147.83 ns/iter 142.82 ns ▃▅█
(133.04 ns … 262.20 ns) 243.43 ns ███
( 65.86 b … 344.19 b) 208.31 b ███▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▃▂▁
4.19 ipc ( 95.88% cache) 0.04 branch misses
565.62 cycles 2.37k instructions 7.26 c-refs 0.30 c-misses
dequal 371.29 ns/iter 380.32 ns █
(356.25 ns … 512.22 ns) 449.80 ns █ ▅
( 0.09 b … 141.33 b) 0.47 b ██▃▂▁█▂▁▁▁▁▁▁▁▁▁▂▁▁▁▁
4.85 ipc ( 86.67% cache) 1.01 branch misses
1.48k cycles 7.18k instructions 0.21 c-refs 0.03 c-misses
are-deeply-equal 344.28 ns/iter 340.81 ns █
(337.71 ns … 514.44 ns) 450.07 ns █
(178.38 b … 376.50 b) 184.89 b █▇▂▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
4.89 ipc ( 95.85% cache) 1.02 branch misses
1.33k cycles 6.49k instructions 6.41 c-refs 0.27 c-misses
lodash.isEqual 36.56 µs/iter 36.68 µs █ █
(35.81 µs … 39.12 µs) 36.96 µs ▅▅█ ▅ ▅▅ █ ▅ ▅
( 5.22 kb … 5.26 kb) 5.23 kb ███▁█▁▁▁▁▁▁▁██▁█▁█▁▁█
2.98 ipc ( 98.56% cache) 197.05 branch misses
139.79k cycles 417.05k instructions 12.74k c-refs 184.14 c-misses
summary
memcmp
2.33x faster than are-deeply-equal
2.51x faster than dequal
247.34x faster than lodash.isEqual
• Typed Array [size=4096]
------------------------------------------- -------------------------------
memcmp 685.25 ns/iter 680.47 ns █
(670.06 ns … 864.32 ns) 824.19 ns █
( 38.69 b … 369.06 b) 208.07 b █▅▂▂▁▁▁▆▁▁▁▁▁▁▁▁▁▁▁▁▁
4.93 ipc ( 94.25% cache) 1.06 branch misses
2.62k cycles 12.90k instructions 8.07 c-refs 0.46 c-misses
dequal 2.96 µs/iter 2.99 µs ▆█
(2.81 µs … 3.28 µs) 3.11 µs ████
( 0.09 b … 0.10 b) 0.09 b ▅▃▁▃▃▅▃▅▅████▁█▃▁▁▁▁▃
4.93 ipc ( 87.45% cache) 1.22 branch misses
11.52k cycles 56.80k instructions 4.49 c-refs 0.56 c-misses
are-deeply-equal 2.38 µs/iter 2.35 µs █
(2.28 µs … 3.34 µs) 3.29 µs ▆█
(175.69 b … 184.19 b) 183.96 b ███▁▂▂▁▁▁▁▂▁▂▁▁▁▁▁▁▁▂
5.02 ipc ( 93.22% cache) 1.12 branch misses
9.72k cycles 48.80k instructions 10.37 c-refs 0.70 c-misses
lodash.isEqual 47.46 µs/iter 47.91 µs █
(42.50 µs … 278.67 µs) 67.54 µs █ █
( 4.07 kb … 333.79 kb) 6.21 kb ▃▄█▃█▃▂▁▁▂▁▂▂▁▁▁▁▁▁▁▁
3.24 ipc ( 94.57% cache) 266.60 branch misses
192.03k cycles 622.18k instructions 15.06k c-refs 818.11 c-misses
summary
memcmp
3.47x faster than are-deeply-equal
4.32x faster than dequal
69.26x faster than lodash.isEqual
• Typed Array [size=16384]
------------------------------------------- -------------------------------
memcmp 2.48 µs/iter 2.49 µs █
(2.46 µs … 2.52 µs) 2.51 µs █▇ ▄
(199.63 b … 208.16 b) 207.91 b ▅██▂▅█▅▁▂█▄▄▄▂▂▁▁▁▂▄▂
5.08 ipc ( 99.74% cache) 1.15 branch misses
9.67k cycles 49.17k instructions 410.25 c-refs 1.05 c-misses
dequal 11.64 µs/iter 11.76 µs █
(11.11 µs … 11.96 µs) 11.78 µs ███
( 0.10 b … 0.10 b) 0.10 b █▁▁▁▁█▁▁▁▁█▁▁▁▁▁▁▁███
4.94 ipc ( 99.37% cache) 1.70 branch misses
45.87k cycles 226.46k instructions 347.13 c-refs 2.19 c-misses
are-deeply-equal 9.27 µs/iter 9.17 µs ▄█
(9.10 µs … 10.06 µs) 10.01 µs ██
(178.33 b … 184.19 b) 183.84 b ██▁█▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▅
5.12 ipc ( 99.51% cache) 1.40 branch misses
37.93k cycles 194.04k instructions 453.74 c-refs 2.21 c-misses
lodash.isEqual 89.55 µs/iter 87.19 µs █
(83.66 µs … 386.22 µs) 153.08 µs █
( 5.66 kb … 483.73 kb) 6.96 kb ▆█▄▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
3.82 ipc ( 92.80% cache) 285.53 branch misses
341.30k cycles 1.30M instructions 16.76k c-refs 1.21k c-misses
summary
memcmp
3.74x faster than are-deeply-equal
4.7x faster than dequal
36.12x faster than lodash.isEqual
`#### Data View
| Library | 16 | 512 | 4096 | 16386 | Speed Range |
| :--- | :--- | :--- | :--- | :--- | :--- |
| memcmp | 9.10 ns | 92.41 ns | 628.69 ns | 2.84 µs | 1.00x (baseline) |
| dequal | 19.31 ns | 388.50 ns | 2.92 µs | 11.24 µs | 2.12x-3.96x slower |
| are-deeply-equal | 108.42 ns | 426.48 ns | 2.49 µs | 9.82 µs | 11.92x-3.46x slower |
| lodash.isEqual | 57.39 µs | 62.89 µs | 78.45 µs | 116.28 µs | 6309.52x-40.91x slower |
Full benchmark results with hardware counters
console
clk: ~3.49 GHz
cpu: AMD Ryzen 5 3600 6-Core Processor
runtime: deno 2.4.2 (x86_64-unknown-linux-gnu)benchmark avg (min … max) p75 / p99 (min … top 1%)
------------------------------------------- -------------------------------
• Data View [size=16]
------------------------------------------- -------------------------------
memcmp 9.10 ns/iter 8.75 ns █
(8.74 ns … 64.48 ns) 13.44 ns █
( 0.09 b … 59.81 b) 0.15 b █▂▃▁▂▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁▁
4.84 ipc ( 98.03% cache) 0.01 branch misses
35.09 cycles 169.83 instructions 0.08 c-refs 0.00 c-misses
dequal 19.31 ns/iter 18.81 ns █
(18.68 ns … 80.09 ns) 30.19 ns █
( 0.09 b … 64.63 b) 0.19 b █▄▃▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
4.61 ipc ( 95.91% cache) 0.01 branch misses
75.10 cycles 346.26 instructions 0.10 c-refs 0.00 c-misses
are-deeply-equal 108.42 ns/iter 103.04 ns █
(99.53 ns … 342.78 ns) 295.16 ns █
(156.49 b … 510.29 b) 392.18 b █▃▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
2.81 ipc ( 95.97% cache) 0.03 branch misses
432.75 cycles 1.22k instructions 13.48 c-refs 0.54 c-misses
lodash.isEqual 57.39 µs/iter 56.90 µs █
(42.29 µs … 260.49 µs) 74.81 µs █
( 1.24 kb … 605.35 kb) 10.38 kb ▁▁▁▁▁▁▁▂▂█▂▂▂▂▁▁▁▁▁▁▁
2.61 ipc ( 94.16% cache) 488.41 branch misses
225.10k cycles 586.46k instructions 23.77k c-refs 1.39k c-misses
summary
memcmp
2.12x faster than dequal
11.92x faster than are-deeply-equal
6309.52x faster than lodash.isEqual
• Data View [size=512]
------------------------------------------- -------------------------------
memcmp 92.41 ns/iter 95.35 ns █
(76.12 ns … 163.05 ns) 136.33 ns ▃ ▆█▄
( 0.09 b … 128.13 b) 0.40 b ▃█▆▂▆███▂▁▁▁▁▁▁▁▁▁▁▁▃
5.55 ipc ( 94.83% cache) 0.01 branch misses
360.40 cycles 2.00k instructions 0.12 c-refs 0.01 c-misses
dequal 388.50 ns/iter 380.23 ns █
(377.51 ns … 631.96 ns) 516.44 ns █
( 0.09 b … 151.94 b) 0.75 b █▂▁▁▁▁▁▂▂▁▁▁▁▁▁▁▁▁▁▁▁
4.86 ipc ( 86.23% cache) 1.01 branch misses
1.48k cycles 7.17k instructions 0.21 c-refs 0.03 c-misses
are-deeply-equal 426.48 ns/iter 405.56 ns █
(399.72 ns … 617.48 ns) 594.59 ns █
( 58.54 b … 600.60 b) 392.17 b █▄▂▁▁▂▁▁▁▁▁▁▁▁▁▁▁▁▃▁▁
4.44 ipc ( 95.29% cache) 1.10 branch misses
1.59k cycles 7.08k instructions 14.77 c-refs 0.70 c-misses
lodash.isEqual 62.89 µs/iter 63.78 µs ██
(59.79 µs … 65.73 µs) 65.51 µs ▅▅▅ ▅ ▅ ██▅ ▅
( 8.75 kb … 8.75 kb) 8.75 kb ███▁▁▁▁▁█▁█▁▁███▁▁▁▁█
2.91 ipc ( 98.01% cache) 389.52 branch misses
245.71k cycles 714.28k instructions 21.75k c-refs 433.82 c-misses
summary
memcmp
4.2x faster than dequal
4.61x faster than are-deeply-equal
680.51x faster than lodash.isEqual
• Data View [size=4096]
------------------------------------------- -------------------------------
memcmp 628.69 ns/iter 636.38 ns █
(608.32 ns … 886.70 ns) 698.59 ns █
( 0.09 b … 161.03 b) 0.89 b ██▆▅▅▅▃▃▂▄▂▁▃▂▁▁▁▂▂▂▁
6.23 ipc ( 87.00% cache) 1.04 branch misses
2.45k cycles 15.29k instructions 1.06 c-refs 0.14 c-misses
dequal 2.92 µs/iter 2.94 µs █
(2.77 µs … 3.47 µs) 3.13 µs ██
( 0.09 b … 0.10 b) 0.09 b █▂▃▁▂▂▃▂▁██▁▂▁▁▁▁▁▁▁▂
4.89 ipc ( 86.04% cache) 1.15 branch misses
11.60k cycles 56.75k instructions 4.00 c-refs 0.56 c-misses
are-deeply-equal 2.49 µs/iter 2.42 µs █
(2.36 µs … 3.34 µs) 3.05 µs ██
(383.73 b … 392.23 b) 392.00 b ██▃▃▁▁▁▁▂▁▁▂▂▁▃▁▂▂▂▃▂
4.94 ipc ( 95.07% cache) 1.14 branch misses
10.00k cycles 49.43k instructions 21.76 c-refs 1.07 c-misses
lodash.isEqual 78.45 µs/iter 76.42 µs █
(65.24 µs … 312.25 µs) 127.63 µs █
( 1.18 kb … 273.32 kb) 10.08 kb ▂▁▂█▇▃▂▂▁▁▁▁▁▁▁▁▁▁▁▁▁
3.03 ipc ( 93.65% cache) 491.38 branch misses
303.49k cycles 920.74k instructions 25.24k c-refs 1.60k c-misses
summary
memcmp
3.96x faster than are-deeply-equal
4.64x faster than dequal
124.78x faster than lodash.isEqual
• Data View [size=16384]
------------------------------------------- -------------------------------
memcmp 2.84 µs/iter 2.86 µs ▂ █▂
(2.76 µs … 3.13 µs) 2.95 µs ▆▃ █▆██
( 0.09 b … 0.10 b) 0.09 b ▅▁▃▅▇██▅████▁▁▁▁▁▁▁▁▃
5.53 ipc ( 99.78% cache) 1.18 branch misses
10.99k cycles 60.77k instructions 272.00 c-refs 0.60 c-misses
dequal 11.24 µs/iter 11.23 µs █
(11.13 µs … 11.53 µs) 11.37 µs ██
( 0.10 b … 0.10 b) 0.10 b █▁██▁█▁██▁▁▁▁█▁▁▁▁▁▁█
4.93 ipc ( 99.46% cache) 1.80 branch misses
46.01k cycles 226.66k instructions 333.42 c-refs 1.79 c-misses
are-deeply-equal 9.82 µs/iter 10.19 µs █ █
(9.21 µs … 10.70 µs) 10.38 µs ▅█▅ ▅▅▅ ▅ ▅ ▅▅ █
(386.39 b … 392.23 b) 391.87 b ███▁▁███▁▁▁▁█▁█▁▁██▁█
5.06 ipc ( 99.34% cache) 1.54 branch misses
38.35k cycles 194.20k instructions 527.12 c-refs 3.50 c-misses
lodash.isEqual 116.28 µs/iter 115.10 µs █
(107.14 µs … 333.15 µs) 169.37 µs █
( 1.24 kb … 226.59 kb) 10.40 kb ▄▆█▅▅▂▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
3.52 ipc ( 92.38% cache) 505.71 branch misses
447.94k cycles 1.58M instructions 26.27k c-refs 2.00k c-misses
summary
memcmp
3.46x faster than are-deeply-equal
3.96x faster than dequal
40.91x faster than lodash.isEqual
`> [!NOTE]
>
is omitted here since it's unavailable in non-Node runtimes.$3
memcmp is designed to be both fast and portable. It adapts to the capabilities of the current runtime:- In Node/Bun, it taps into native memory comparison for best performance.
- In Deno and browsers, it ensures consistency without relying on non-standard globals.
As a result, it performs consistently faster than popular deep equality libraries across all tested sizes and buffer types.
$3
`bash
node
npm run benchmark:nodedeno
npm run benchmark:denobun
npm run benchmark:bun
`Make sure your environment matches the target runtime (Node, Deno, etc.) to reproduce the results under the intended logic path.
Build
This package uses rollup to generate clean and optimized ESM/CJS builds.
To build package from source code, run:
bash
npm run build
`This will generate the output in the
folder. Builds are handled via custom rollup config and exposed under appropriate exports in package.json.Testing
All tests are written in Vitest with native ESM support and zero transform overhead.
You can run the full suite with:
`bash
npm test
``If you find this project useful, you can support it with a one-time donation over PayPal. Thank you!
Feel free to open issues or submit pull requests on GitHub.
This project is licensed under the MIT License.