Awesome
isomorphic-ui-benchmarks
This repo includes multiple benchmarks for various UI libraries. Each benchmark is designed to measure rendering performance (on the server and in the browser) and the time that it takes to update the DOM (client-side only).
Current results
Search results
<img width="1335" alt="Search Results Benchmarks" src="https://user-images.githubusercontent.com/4985201/67251108-ca344780-f422-11e9-9ca0-cce6ed1aad6b.png">Color picker
<img width="1323" alt="Color Picker Benchmarks" src="https://user-images.githubusercontent.com/4985201/67251074-a96bf200-f422-11e9-9bee-a15124f4b10d.png">Below are the results of a run on October 21, 2019
Environment:
- Node.js v10.16.1
- MacBook Pro (13-inch, 2017)
- Processor: 3.5 GHz Intel Core i7
- Memory: 16 GB 2133 MHz LPDDR3
- macOS Mojave: 10.14.6
- Google Chrome Version 77.0.3865.120 (Official Build) (64-bit)
Server-side
Warming up...
Warmup complete.
Running "search-results"...
Running benchmark "marko"...
marko x 6,399 ops/sec ±2.71% (84 runs sampled)
Running benchmark "preact"...
preact x 746 ops/sec ±2.88% (81 runs sampled)
Running benchmark "react"...
react x 765 ops/sec ±5.02% (72 runs sampled)
Running benchmark "vue"...
vue x 2,657 ops/sec ±4.41% (60 runs sampled)
Running benchmark "inferno"...
inferno x 3,014 ops/sec ±1.78% (87 runs sampled)
Fastest is marko
--------------
Warming up...
Warmup complete.
Running "color-picker"...
Running benchmark "marko"...
marko x 24,540 ops/sec ±1.48% (86 runs sampled)
Running benchmark "preact"...
preact x 4,587 ops/sec ±1.81% (85 runs sampled)
Running benchmark "react"...
react x 4,300 ops/sec ±4.72% (72 runs sampled)
Running benchmark "vue"...
vue x 9,120 ops/sec ±5.56% (70 runs sampled)
Running benchmark "inferno"...
inferno x 21,453 ops/sec ±2.12% (84 runs sampled)
Fastest is marko
--------------
DONE!
~/marko-js/isomorphic-ui-benchmarks (master)> node -v
v10.16.1
Client-side
Search results
Google Chrome
Warming up...
Warmup complete.
Running "search-results"...
Running benchmark "marko"...
marko x 175 ops/sec ±1.84% (53 runs sampled)
Running benchmark "preact"...
preact x 132 ops/sec ±1.66% (48 runs sampled)
Running benchmark "react"...
react x 210 ops/sec ±1.36% (53 runs sampled)
Running benchmark "vue"...
vue x 142 ops/sec ±1.31% (52 runs sampled)
Running benchmark "inferno"...
inferno x 239 ops/sec ±1.24% (55 runs sampled)
Fastest is inferno
Color picker
Google Chrome
Warming up...
Warmup complete.
Running "color-picker"...
Running benchmark "marko"...
marko x 6,008 ops/sec ±1.66% (34 runs sampled)
Running benchmark "preact"...
preact x 6,435 ops/sec ±0.96% (59 runs sampled)
Running benchmark "react"...
react x 7,358 ops/sec ±1.43% (58 runs sampled)
Running benchmark "vue"...
vue x 4,291 ops/sec ±1.96% (55 runs sampled)
Running benchmark "inferno"...
inferno x 17,078 ops/sec ±2.17% (60 runs sampled)
Fastest is inferno
Additional details
Included libraries
The following UI libraries are currently included:
Included benchmarks
This repo currently includes the following benchmarks
Search Results
This benchmark measures the time it takes to render pages of search results. Each page includes 100 search result items. Every iteration renders an entirely new set of search results. As a result of rendering new search results for every cycle, a significant number of DOM nodes must be updated.
Color Picker
This benchmark measures the time it takes to cycle through a selected color. The selected color index changes every cycle. When the selected color index changes two things happen:
- The new selected color is highlighted
- The old selected color is unhighlighted
- The selected color is shown at the end
Compared to the search results benchmark, there are a relatively small number of changes to the DOM for every cycle.
Running the benchmarks
Install
git clone https://github.com/marko-js/isomorphic-ui-benchmarks.git
cd isomorphic-ui-benchmarks
npm install
npm run build # Build client-side JS bundles
Run server-side benchmarks
npm run benchmark
Run client-side benchmarks
Start
npm start
Open http://localhost:8080/ in your browser and choose a benchmark to run.
Contributions and Feedback
If you see any problems or have any suggestions please let us know. Every effort was made to be as fair and accurate as possible, but mistakes do happen. If you find a problem please open a Github issue to discuss.