FULL PRODUCT VERSION :
Java(TM) SE Runtime Environment (build 1.8.0_45-b14)
Java HotSpot(TM) 64-Bit Server VM (build 25.45-b02, mixed mode)

ADDITIONAL OS VERSION INFORMATION :
Darwin 13.4.0 Darwin Kernel Version 13.4.0: Wed Mar 18 16:20:14 PDT 2015; root:xnu-2422.115.14~1/RELEASE_X86_64 x86_64

A DESCRIPTION OF THE PROBLEM :
Parallel reduction produces the wrong result. Reducing 100,200,300,400,500 by addition with a seed of 50 should give 1550. In fact, sequential reduction gives this result, but parallel reduction results in 1750.

This is reduction via java.util.Stream

    T reduce(T identity, BinaryOperator<T> accumulator);

Failing Unit test with Junit and Hamcrest

import static org.hamcrest.Matchers.*;
import static org.junit.Assert.assertThat;

assertThat(Stream.of(1,2,3,4,5).map(i->i*100).reduce( 50,(acc,next) -> acc+next),equalTo(Stream.of(1,2,3,4,5).map(i->i*100).parallel().reduce( 50,(acc,next) -> acc+next)));

ADDITIONAL REGRESSION INFORMATION:
Java(TM) SE Runtime Environment (build 1.8.0_45-b14)
Java HotSpot(TM) 64-Bit Server VM (build 25.45-b02, mixed mode)

STEPS TO FOLLOW TO REPRODUCE THE PROBLEM :
Failing Unit test with Junit and Hamcrest

import static org.hamcrest.Matchers.*;
import static org.junit.Assert.assertThat;

assertThat(Stream.of(1,2,3,4,5).map(i->i*100).reduce( 50,(acc,next) -> acc+next),equalTo(Stream.of(1,2,3,4,5).map(i->i*100).parallel().reduce( 50,(acc,next) -> acc+next)));

EXPECTED VERSUS ACTUAL BEHAVIOR :
EXPECTED -
1550 and passing test
ACTUAL -
1750 and failing test

REPRODUCIBILITY :
This bug can be reproduced always.

---------- BEGIN SOURCE ----------
import static org.hamcrest.Matchers.*;
import static org.junit.Assert.assertThat;

@Test
public void testMapReduceSeed2(){
assertThat(Stream.of(1,2,3,4,5).map(i->i*100).reduce( 50,(acc,next) -> acc+next),equalTo(Stream.of(1,2,3,4,5).map(i->i*100).parallel().reduce( 50,(acc,next) -> acc+next)));

  }
---------- END SOURCE ----------