Performance Analysis Tool

Sky Engine is performance analysis tool based on web services. This tool analyzes your application performance and provide hint performance improvement for opensource developer using at GitHub.

Key Features

• integrates with GitHub and Travis-CI
• automatically analyze performance and key point with improvement for application when commit to repository
• will be supported to ruby, python, javascript(node.js or phantom.js), java, and another languages
• shows the result to analyzing performance improvement within each functions at runtime

How To Build

1. Build: execute # mvn test package if maven is installed or # ./mvnw test package in SkyEngine source root directory
2. Enjoy!

To-do

• Refactoring
  • do refactoring service classes
  • change jsp files to freemarker
  • detach sky pom.xml from spring-boot-parent pom.xml
  • add sky-server unit-test for javascript files using angularjs
  • create distributed sky-worker servers using zookeeper or message queueing (use apache curator framework)
  • network communication using thrift replace with pure RESTFul protocol
• Bug fixes
  • missing data as profiling log
  • don't show profile result with error while running sky-worker
• New features
  • Dashboard
    • create comparison chart between profiled commits
    • create chart with each Profile and Method
  • add line number and filename in profiled Method of database
  • add performance analysis for program written by Ruby language
  • add feature to manage sky-worker server clusters (add, remove, information, etc...)
  • create bash script for running analysis to sky engine in travis ci servers
• Backlogs
  • how to profile hadoop mapreduce
  • think about simplfy for worker
  • attach module to encryption with jasypt (http://www.jasypt.com)