A comparison between lodash.isequal, fast-deep-equal and object-hash for a specific usecase
Check the benchmarks at index.js
-
Clone this repo and
cdinto the project root -
Install dependencies:
yarn- Generate dataset. Change the sample size in
generate.js. Then run:
yarn generate- Run the tests:
yarn start-
lodash.isequal: Deep comparing objects to determine changed objects
-
fast-deep-equal: Deep comparing objects to determine changes objects
-
object-hash: Storing the hash value of objects and later comparing them with the new objects. The new objects undergo a md5 hash process and this hash value is compared against their last hash value
Tests ran only once for brevity:
| Sample size | Lodash.isequal (ms) | fast-deep-equal (ms) | object-hash (ms) |
|---|---|---|---|
| 10 | 3.73 (slowest) | 0.44 (fastest) | 3.55 |
| 100 | 5.97 | 1.95 (fastest) | 24.92 (slowest) |
| 1,000 | 126.77 | 113.87 (fastest) | 226.28 (slowest) |
| 5,000 | 2839.30 | 2620.96 (fastest) | 3249.85 (slowest) |
| 10,000 | 12350.28 | 12264.97 (fastest) | 12877.76 (slowest) |
| 20,000 | 55649.79 (slowest) | 55494.73 | 50748.38 (fastest) |
| Sample size | Full JSON (MB) | Hashed JSON (MB) |
|---|---|---|
| 10 | 0.033 | 0.0005 |
| 100 | 0.034 | 0.005 |
| 1,000 | 0.351 | 0.054 |
| 5,000 | 1.7 | 0.272 |
| 10,000 | 3.5 | 0.546 |
| 20,000 | 7 | 1.1 |