如何优化 Java 函数处理大数据时的性能？

import java.util.concurrent.ForkJoinPool;
import java.util.stream.IntStream;

public class ParallelSort {

    public static void main(String[] args) {
        int[] arr = ...; // 大数据数组

        // 并行归并排序
        ForkJoinPool pool = new ForkJoinPool();
        int[] sorted = pool.invoke(new MergeSort(arr));

        // 使用流式 API 打印排序后的数组
        IntStream.of(sorted).forEach(x -> System.out.print(x + " "));
    }

    static class MergeSort extends RecursiveAction {

        private int[] arr;

        public MergeSort(int[] arr) {
            this.arr = arr;
        }

        @Override
        protected void compute() {
            if (arr.length <= 1) {
                return;
            }

            int mid = arr.length / 2;
            int[] left = Arrays.copyOfRange(arr, 0, mid);
            int[] right = Arrays.copyOfRange(arr, mid, arr.length);
            invokeAll(new MergeSort(left), new MergeSort(right));
            merge(left, right);
        }

        private void merge(int[] left, int[] right) {
            // 合并排好序的左数组和右数组
            ...
        }
    }
}