Skip to content

cs-au-dk/streamliner

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Streamliner

This is the source code for the prototype implementation of the technique presented in Eliminating Abstraction Overhead of Java Stream Pipelines using Ahead-of-Time Program Optimization.

Build instructions

The project uses maven for dependency management. It is compiled with

mvn compile

Unit tests can be run with

mvn test

Most of the tests should run successfully, but it is expected that some 16-17 tests will fail.

Reproduce RQ1 results

The misc/jmh.sh script will optimise the micro benchmarks specified in dk.casa.streamliner.asm.TransformASM , and use JMH to benchmark the programs.

Reproduce RQ2 results

Clone and build the benchmarked projects with the RQ2/clone_and_build.py script.

Run the analysis and optimisation on the projects with the command:

env MAVEN_OPTS="-Xmx20g -Xss512m --add-opens java.base/java.lang.module=ALL-UNNAMED" mvn compile exec:java -Dexec.mainClass=dk.casa.streamliner.asm.RQ2.Experiment

Which pre-analysis to use is chosen with a command-line argument. For instance, to use the SPARK-powered analysis, add -Dexec.args="spark" to the command. The "wala" analysis is similarly available.

Additional options can be specified for SPARK by passing them as a second argument: -Dexec.args="spark cs-demand:true,on-fly-cg:true". A list of available options can be found here.

Be aware that these analyses contain bugs and issues that cause many type queries to fail. Some stats on query results will be shown when these analyses are used.

Further use

See dk.casa.streamliner.asm.OptimizeSkeleton for a guide on how to apply the tool in more contexts.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published