React Native Performance Experiments

Various performance experiments with React Native over a swipeable card example. The same example is implemented multiple times with a pan responder, direct component manipulation, native container and Animated library.

Full article: https://medium.com/@talkol/performance-limitations-of-react-native-and-how-to-overcome-them-947630d7f440

Running the demo

• npm install
• Open ios/example.xcodeproj in Xcode
• It's recommended to run on a real device since the simulator doesn't provide authentic results
• You may have to set your credentials under
  • Target example - General - Identity - Team
  • Target example - Build Settings - Code Signing Identity
• Don't forget in AppDelegate.m to unremark the jsCodeLocation in OPTION 2 if you want to avoid the packager and upload your bundle to the device

Simulation of stress conditions

Since this app is overly simplified, performance will probably be equally good in all 4 implementaitons. In order to feel the differences between the implementations, the following stress conditions can be simulated:

• Simulate heavy render function - a computationally-intensive block is added to the render function

• Simulate busy bridge every 1 sec - a burst of data is sent over the bridge every 1 second