Mathis/sharp
1
0
Fork 0
mirror of https://github.com/lovell/sharp.git synced 2025-07-09 10:30:15 +02:00
Code Issues Packages Projects Releases Wiki Activity

Docs: update performance test results

Browse Source
This commit is contained in:
Lovell Fuller 2020-12-21 21:00:44 +00:00
parent 0e62bde5c3
commit 774d78228e
1 changed files with 14 additions and 14 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
docs/performance.md
View File

@ -4,11 +4,11 @@ A test to benchmark the performance of this module relative to alternatives.
## The contenders
* [jimp](https://www.npmjs.com/package/jimp) v0.16.0 - Image processing in pure JavaScript. Provides bicubic interpolation.
* [mapnik](https://www.npmjs.org/package/mapnik) v4.5.2 - Whilst primarily a map renderer, Mapnik contains bitmap image utilities.
* [jimp](https://www.npmjs.com/package/jimp) v0.16.1 - Image processing in pure JavaScript. Provides bicubic interpolation.
* [mapnik](https://www.npmjs.org/package/mapnik) v4.5.5 - Whilst primarily a map renderer, Mapnik contains bitmap image utilities.
* [imagemagick](https://www.npmjs.com/package/imagemagick) v0.1.3 - Supports filesystem only and "*has been unmaintained for a long time*".
* [gm](https://www.npmjs.com/package/gm) v1.23.1 - Fully featured wrapper around GraphicsMagick's `gm` command line utility.
* sharp v0.26.0 / libvips v8.10.0 - Caching within libvips disabled to ensure a fair comparison.
* sharp v0.27.0 / libvips v8.10.5 - Caching within libvips disabled to ensure a fair comparison.
## The task
@ -18,22 +18,22 @@ then compress to JPEG at a "quality" setting of 80.
## Test environment
* AWS EC2 eu-west-1 [c5d.large](https://aws.amazon.com/ec2/instance-types/c5/) (2x Xeon Platinum 8124M CPU @ 3.00GHz)
* Ubuntu 20.04 (ami-0f1d11c92a9467c07)
* Node.js v14.8.0
* AWS EC2 eu-west-1 [c5d.large](https://aws.amazon.com/ec2/instance-types/c5/) (2x Xeon Platinum 8275CL CPU @ 3.00GHz)
* Ubuntu 20.10 (ami-046cdbcee95cdd75c)
* Node.js v14.15.3
## Results
| Module | Input | Output | Ops/sec | Speed-up |
| :----------------- | :----- | :----- | ------: | -------: |
| jimp | buffer | buffer | 0.75 | 1.0 |
| mapnik | buffer | buffer | 3.00 | 4.0 |
| gm | buffer | buffer | 4.12 | 5.5 |
| gm | file | file | 4.13 | 5.5 |
| imagemagick | file | file | 4.30 | 5.7 |
| sharp | stream | stream | 22.37 | 29.8 |
| sharp | file | file | 23.40 | 31.2 |
| sharp | buffer | buffer | 24.01 | 32.0 |
| jimp | buffer | buffer | 0.77 | 1.0 |
| mapnik | buffer | buffer | 3.39 | 4.4 |
| gm | buffer | buffer | 4.30 | 5.6 |
| gm | file | file | 4.33 | 5.6 |
| imagemagick | file | file | 4.39 | 5.7 |
| sharp | stream | stream | 23.81 | 30.9 |
| sharp | file | file | 25.09 | 32.6 |
| sharp | buffer | buffer | 25.60 | 33.2 |
Greater libvips performance can be expected with caching enabled (default)
and using 4+ core machines, especially those with larger L1/L2 CPU caches.