mtumilowicz / java11-stream-parallel-linkedlist-vs-arraylist

ArrayList vs LinkedList parallel streams performance comparison.

java11-stream-parallel-linkedlist-vs-arraylist

Reference: http://alexandrastech.blogspot.com/2016/04/java-8-parallel-streams-performance.html
Reference: https://www.amazon.com/Modern-Java-Action-functional-programming/dp/1617293563

preface

Source	Decomposability
ArrayList	Good
LinkedList	Poor
IntStream.range	Excellent
Stream.iterate	Poor
HashSet	Good
TreeSet	Good

ArrayList can be split much more efficiently than a LinkedList, because the first can be evenly divided without traversing it (implements RandomAccess), as it’s necessary to do with the latter.

The stream is internally splitted using Spliterator - please refer my other github projects:

• https://github.com/mtumilowicz/java11-spliterator
• https://github.com/mtumilowicz/java11-spliterator-forkjoin

project description

We provide basic comparison of parallel streams over ArrayList and LinkedList.

• compare
  • ArrayList

    var integers = new ArrayList<Integer>();
    
    setAll(integers);
    
    System.out.println(timeOfSumming(integers));
    

    time: 60
  • LinkedList

    var integers = new LinkedList<Integer>();
    
    setAll(integers);
    
    System.out.println(timeOfSumming(integers));
    

    time: 400
• summing in parallel

  private int parallelSum(List<Integer> values) {
      return values.parallelStream().mapToInt(i -> i).sum();
  }
  

• count time of summing

  private long timeOfSumming(List<Integer> integers) {
      long startTime = System.currentTimeMillis();
      parallelSum(integers);
      return System.currentTimeMillis() - startTime;
  }
  

• fill lists with random numbers

  private void setAll(List<Integer> integers) {
      Random random = new Random();
      IntStream.range(1, 10_000_000)
              .forEach(i -> integers.add(random.nextInt(500)));
  }