golang中map和slice的查询速度比较，结果意想不到

最新推荐文章于 2025-06-01 10:43:45 发布

原创最新推荐文章于 2025-06-01 10:43:45 发布 · 1.4w 阅读

3 ·

CC 4.0 BY-SA版权

文章标签：

#golang #算法

golang 专栏收录该内容

20 篇文章

订阅专栏

直觉认为map查询速度快于slice，但通过代码测试发现，当数据量小时，slice的查询速度反而更快。这是因为Golang的map底层使用hash，而hash函数在小数据量时的开销超过了直接遍历slice。随着数据量增加，map的O(1)优势逐渐体现。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

看到标题，第一反应，map肯定秒杀slice啊，我当时也是这么想的，毕竟前者的查询复杂度是O(1)，后者是O(n)。

但是，事事难预料哦，用代码来说明问题：

先写两个测试函数


var testTimeSlice=[]string{"aa","bb","cc","dd","ee","aa","zz"}

var testTimeMap = map[string]bool{"aa": true, "bb": true, "cc": true, "dd": true, "ee": true, "ff": true, "zz": true}

//以上为第一组查询测试数据


var testTimeSlice2=[] string{"aa","bb","cc","dd","ee","aa","aa","bb","cc","dd","ee","aa","aa","bb","cc","dd","ee","aa","aa","bb","cc","dd","ee","aa","i","j", "l", "m", "n", "o", "p", "q", "k", "x", "y", "z",
    "1", "2", "3", "4", "5", "6", "7", "8", "9", "10","zz"}

var testTimeMap2 = map[string]bool{"aa": true, "bb": true, "cc": true, "dd": true, "ee": true, "ff": true, "qq": true,"ww": true, "rr": true, "tt": true, "zz": true, "uu": true, "ii": true,"oo": true, "pp": true, "lk": true, "kl": true, "jk": true, "kj": true,"hl": true, "lh": true, "fg": true, "gfdd": true, "df": true, "fd": true,
    "i": true, "j": true, "l": true, "m": true, "n": true, "o": true, "p": true, "q": true, "k": true, "x": true, "y": true, "z": true,
    "1": true, "2": true, "3": true, "4": true, "5": true, "6": true, "7": true, "8": true, "9": true, "10": true}

//以上为第二组查询测试数据


func testSlice(a []string)  {
    now:=time.Now()

    for j:=0; j < 100000; j++{
        for _,v:=range a{
            if v=="zz"{
                break
            }
        }
    }
    finish:=time.Since(now)
fmt.Println(finish1)
}


func testMap(a map[string]bool) {
    now:=time.Now()
    for j:=0; j < 100000; j++{
        if _, ok := a["zz"]; ok {
            continue
        }
    }
    finish2:=time.Since(now)
    fmt.Println("!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!")
    fmt.Println(finish2)
    fmt.Println("!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!")
}

第一个函数接受一个slice，然后进行十万次查询，第二个函数接受一个map，进行十万次查询。

当用第一组测试数据时，结果如下：

1.0011ms   //Slice查询耗时
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
6.0038ms   //Map查询耗时
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

用第二组测试数据时，结果如下：

8.0038ms    //Slice查询耗时
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
2.0019ms    //Map查询耗时
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

现在可以得出结论了，当数据量小的时候，slice的查询速度是比map快的，为什么呢，因为golang的map底层是用hash实现的，既然有hash，那就要做映射，那就有hash函数，这个hash函数的开销千万不要忘记，不要一看到map就只记着O(1)

数据集小，直接遍历查找的开销肯定比hash的开销小。

当数据量上去了，map的优势自然而然就出来了。