Stream.toList() vs Collectors.toList() 성능비교

일단 차이점을 찾아보니

• Stream.toList() return -> 변경할수 없는 List 이고 추가,삭제 불가
• Collectors.toList() return -> 일반적인 List 이고 추가,삭제 가능
• Collectors.toUnmodifiableList() return -> 변경할수 없는 List 이고 추가,삭제 불가

• 성능관련해서 테스트 해보았다.

@BenchmarkMode(Mode.All)
@Fork(1)
@State(Scope.Thread)
@Warmup(iterations = 20, time = 1, batchSize = 10000)
@Measurement(iterations = 20, time = 1, batchSize = 10000)
public class BenchmarkStreamToList {

    @Benchmark
    public List<Integer> streamToList() {
        return IntStream.range(1, 1000).boxed().toList();
    }

    @Benchmark
    public List<Integer> collectorsToList() {
        return IntStream.range(1, 1000).boxed().collect(Collectors.toList());
    }

    @Benchmark
    public List<Integer> streamToList() {
        return IntStream.range(1, 1000).boxed().toList();
    }

}

결과:

Benchmark                                                                                  Mode  Cnt   Score    Error  Units
BenchmarkStreamToList.collectorsToList                                                    thrpt   20  24.422 ±  0.268  ops/s
BenchmarkStreamToList.collectorsToUnmodifiableList                                        thrpt   20  22.784 ±  0.599  ops/s
BenchmarkStreamToList.streamToList                                                        thrpt   20  31.779 ±  1.732  ops/s
BenchmarkStreamToList.collectorsToList                                                     avgt   20   0.045 ±  0.006   s/op
BenchmarkStreamToList.collectorsToUnmodifiableList                                         avgt   20   0.062 ±  0.035   s/op
BenchmarkStreamToList.streamToList                                                         avgt   20   0.040 ±  0.028   s/op
BenchmarkStreamToList.collectorsToList                                                   sample  445   0.046 ±  0.002   s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.00                            sample        0.039            s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.50                            sample        0.041            s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.90                            sample        0.057            s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.95                            sample        0.073            s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.99                            sample        0.102            s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.999                           sample        0.150            s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.9999                          sample        0.150            s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p1.00                            sample        0.150            s/op
BenchmarkStreamToList.collectorsToUnmodifiableList                                       sample  460   0.044 ±  0.001   s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.00    sample        0.042            s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.50    sample        0.044            s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.90    sample        0.046            s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.95    sample        0.047            s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.99    sample        0.051            s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.999   sample        0.057            s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.9999  sample        0.057            s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p1.00    sample        0.057            s/op
BenchmarkStreamToList.streamToList                                                       sample  655   0.031 ±  0.001   s/op
BenchmarkStreamToList.streamToList:streamToList·p0.00                                    sample        0.030            s/op
BenchmarkStreamToList.streamToList:streamToList·p0.50                                    sample        0.031            s/op
BenchmarkStreamToList.streamToList:streamToList·p0.90                                    sample        0.032            s/op
BenchmarkStreamToList.streamToList:streamToList·p0.95                                    sample        0.033            s/op
BenchmarkStreamToList.streamToList:streamToList·p0.99                                    sample        0.035            s/op
BenchmarkStreamToList.streamToList:streamToList·p0.999                                   sample        0.037            s/op
BenchmarkStreamToList.streamToList:streamToList·p0.9999                                  sample        0.037            s/op
BenchmarkStreamToList.streamToList:streamToList·p1.00                                    sample        0.037            s/op
BenchmarkStreamToList.collectorsToList                                                       ss   20   0.043 ±  0.001   s/op
BenchmarkStreamToList.collectorsToUnmodifiableList                                           ss   20   0.045 ±  0.004   s/op
BenchmarkStreamToList.streamToList                                                           ss   20   0.031 ±  0.001   s/op

결론: Stream.toList()
성능은 모든 방면에서
Collectors.toList() 와 Collectors.toUnmodifiableList()를 앞지른다!

• 다음은 조금 더 큰 데이터를 세팅해보자!

@Benchmark
public List<Integer> streamToList() {
  return IntStream.range(1, 10000).boxed().toList();
}

@Benchmark
public List<Integer> collectorsToList() {
  return IntStream.range(1, 10000).boxed().collect(Collectors.toList());
}

@Benchmark
public List<Integer> streamToList() {
  return IntStream.range(1, 10000).boxed().toList();
}

결과:

Benchmark                                                                                  Mode  Cnt  Score   Error  Units
BenchmarkStreamToList.collectorsToList                                                    thrpt   20  2.186 ± 0.162  ops/s
BenchmarkStreamToList.collectorsToUnmodifiableList                                        thrpt   20  2.184 ± 0.042  ops/s
BenchmarkStreamToList.streamToList                                                        thrpt   20  3.538 ± 0.058  ops/s
BenchmarkStreamToList.collectorsToList                                                     avgt   20  0.426 ± 0.004   s/op
BenchmarkStreamToList.collectorsToUnmodifiableList                                         avgt   20  0.469 ± 0.016   s/op
BenchmarkStreamToList.streamToList                                                         avgt   20  0.293 ± 0.008   s/op
BenchmarkStreamToList.collectorsToList                                                   sample   58  0.448 ± 0.049   s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.00                            sample       0.414           s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.50                            sample       0.422           s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.90                            sample       0.458           s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.95                            sample       0.560           s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.99                            sample       1.160           s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.999                           sample       1.160           s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p0.9999                          sample       1.160           s/op
BenchmarkStreamToList.collectorsToList:collectorsToList·p1.00                            sample       1.160           s/op
BenchmarkStreamToList.collectorsToUnmodifiableList                                       sample   60  0.458 ± 0.004   s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.00    sample       0.447           s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.50    sample       0.455           s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.90    sample       0.471           s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.95    sample       0.482           s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.99    sample       0.492           s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.999   sample       0.492           s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p0.9999  sample       0.492           s/op
BenchmarkStreamToList.collectorsToUnmodifiableList:collectorsToUnmodifiableList·p1.00    sample       0.492           s/op
BenchmarkStreamToList.streamToList                                                       sample   78  0.293 ± 0.012   s/op
BenchmarkStreamToList.streamToList:streamToList·p0.00                                    sample       0.277           s/op
BenchmarkStreamToList.streamToList:streamToList·p0.50                                    sample       0.284           s/op
BenchmarkStreamToList.streamToList:streamToList·p0.90                                    sample       0.309           s/op
BenchmarkStreamToList.streamToList:streamToList·p0.95                                    sample       0.377           s/op
BenchmarkStreamToList.streamToList:streamToList·p0.99                                    sample       0.459           s/op
BenchmarkStreamToList.streamToList:streamToList·p0.999                                   sample       0.459           s/op
BenchmarkStreamToList.streamToList:streamToList·p0.9999                                  sample       0.459           s/op
BenchmarkStreamToList.streamToList:streamToList·p1.00                                    sample       0.459           s/op
BenchmarkStreamToList.collectorsToList                                                       ss   20  0.474 ± 0.133   s/op
BenchmarkStreamToList.collectorsToUnmodifiableList                                           ss   20  0.493 ± 0.099   s/op
BenchmarkStreamToList.streamToList                                                           ss   20  0.325 ± 0.056   s/op

결론: 데이터를 크게 셋팅해도 Stream.toList() 성능은 모든 방면에서 Collectors.toList() 와 Collectors.toUnmodifiableList()를 앞지른다!

괜히 테스트 했나 ????

끝!