1.阿里巴巴代码规范的要求

2.使用阿里巴巴插件扫描时

3. 源码

3.1 当初始化不设置大小时

Map<Integer, BigDecimal> staffDidiMap = new HashMap<>();

3.2 当初始化设置大小时

Map<Integer, BigDecimal> staffDidiMap = new HashMap<>(16);

@Test
	public void caffineTest() {
		System.out.println(1 << 30);
	}
result: 1073741824

---

this.threshold = tableSizeFor(initialCapacity);

@Test
	public void caffineTest() {
		int cap = 16;
		int n = cap - 1;
		n |= n >>> 1;
		n |= n >>> 2;
		n |= n >>> 4;
		n |= n >>> 8;
		n |= n >>> 16;
		int result = (n < 0) ? 1 : (n >= 1073741824) ? 1073741824 : n + 1;
		System.out.println(result);
	}
result: 16

// TODO 源码部分待完善

4. 测试

@SpringBootTest
public class CaffineTest {

	@Test
	public void caffineTest() throws InterruptedException {
		test1();
		test2();
		test3();
		test4();
		test5();
	}

	void test1() throws InterruptedException {
		int aHundredMillion = 10000000;
		Map<Integer, Integer> map = new HashMap<>();
		long s1 = System.currentTimeMillis();
		for (int i = 0; i < aHundredMillion; i++) {
			map.put(i, i);
		}
		long s2 = System.currentTimeMillis();
		System.out.println("未初始化容量，耗时 ： " + (s2 - s1));
		map.clear();
		Thread.sleep(5000);
	}

	void test2() throws InterruptedException {
		int aHundredMillion = 10000000;
		Map<Integer, Integer> map1 = new HashMap<>(aHundredMillion / 2);
		long s5 = System.currentTimeMillis();
		for (int i = 0; i < aHundredMillion; i++) {
			map1.put(i, i);
		}
		long s6 = System.currentTimeMillis();
		System.out.println("初始化容量5000000，耗时 ： " + (s6 - s5));
		map1.clear();
		Thread.sleep(5000);
	}

	void test3() throws InterruptedException {
		int aHundredMillion = 10000000;
		Map<Integer, Integer> map2 = new HashMap<>(aHundredMillion);
		long s3 = System.currentTimeMillis();
		for (int i = 0; i < aHundredMillion; i++) {
			map2.put(i, i);
		}
		long s4 = System.currentTimeMillis();
		System.out.println("初始化容量为10000000，耗时 ： " + (s4 - s3));
		map2.clear();
		Thread.sleep(5000);
	}

	void test4() throws InterruptedException {
		int aHundredMillion = 10000000;
		// 按照阿里巴巴规范   initialCapacity = (需要存储的元素个数 / 负载因子) + 1。注意负载因子（即 loader factor）默认 为 0.75
		int initial = (int) (aHundredMillion / 0.75F + 1F);
		Map<Integer, Integer> map3 = new HashMap<>(initial);
		long s7 = System.currentTimeMillis();
		for (int i = 0; i < aHundredMillion; i++) {
			map3.put(i, i);
		}
		long s8 = System.currentTimeMillis();
		System.out.println("初始化容量为" + initial + "耗时" + (s8 - s7));
		map3.clear();
		Thread.sleep(5000);
	}

	void test5() throws InterruptedException {
		int aHundredMillion = 10000000;
		// 初始化16
		Map<Integer, Integer> map4 = new HashMap<>(16);
		long s9 = System.currentTimeMillis();
		for (int i = 0; i < aHundredMillion; i++) {
			map4.put(i, i);
		}
		long s10 = System.currentTimeMillis();
		System.out.println("初始化容量为" + 16 + "耗时" + (s10 - s9));
		map4.clear();
		Thread.sleep(5000);
	}
}

结果：

每次向map中插入一千万条数据。
方法一：未初始化耗时3995
方法二：初始化五百万耗时2345
方法三：初始化为一千万耗时261
方法四：按照阿里的规范初始化，耗时192
方法五：初始化16，耗时182

！！！
多次测试过后，可以得出结论性能最差的是未初始化，最高效且稳定的是按照阿里巴巴规范initialCapacity = (需要存储的元素个数 / 负载因子) + 1。注意负载因子（即 loader factor）默认 为 0.75，然后是16最后是按照多大赋多大

附录

1.为什么阿里巴巴建议HashMap初始化时需要指定容量大小？