发布时间:2024-12-23 05:36:01
```go func sum(nums []int) int { total := 0 for _, num := range nums { total += num } return total } ```
传统的循环求和方法是简单直接的,但在处理大规模数据时可能会存在性能问题。为了解决这个问题,我们可以考虑使用并发来加速计算过程。```go func concurrentSum(nums []int) int { n := len(nums) numParts := runtime.NumCPU() partSize := n / numParts sumCh := make(chan int, numParts) for i := 0; i < numParts; i++ { start := i * partSize end := start + partSize if i == numParts-1 { end = n } go func(nums []int, sumCh chan<- int) { total := 0 for _, num := range nums { total += num } sumCh <- total }(nums[start:end], sumCh) } total := 0 for i := 0; i < numParts; i++ { total += <-sumCh } close(sumCh) return total } ```
上述代码中,我们使用了runtime.NumCPU()函数来获取当前系统的CPU核心数,然后根据核心数将原始数据拆分成多个部分。 创建了相应数量的goroutine后,每个goroutine都会对其所负责的部分进行求和,并将结果发送到sumCh通道中。主协程通过从sumCh通道接收每个部分的结果,然后将其累加得到最终的求和结果。最后,我们需要关闭sumCh通道以释放资源。```go func main() { nums := generateNums(10000000) start := time.Now() total := sum(nums) elapsed := time.Since(start) fmt.Println("Sequential sum:", total) fmt.Println("Time taken (sequential):", elapsed) start = time.Now() total = concurrentSum(nums) elapsed = time.Since(start) fmt.Println("Concurrent sum:", total) fmt.Println("Time taken (concurrent):", elapsed) } func generateNums(n int) []int { nums := make([]int, n) for i := 0; i < n; i++ { nums[i] = rand.Intn(100) } return nums } ```
运行以上代码后,我们可以看到输出的结果类似如下:``` Sequential sum: 50159873 Time taken (sequential): 6.499843041s Concurrent sum: 50159873 Time taken (concurrent): 2.263267034s ```
根据测试结果可以发现,并发求和明显比传统的循环求和更快,尤其是在处理大量数据时。并发求和的运行时间几乎缩短了3倍,这证明了Golang在并发编程方面的强大能力。