C++学习笔记

Astrologer__

已于 2022-03-21 16:18:45 修改

阅读量828

点赞数

文章标签： c++

于 2021-12-16 13:24:48 首次发布

本文链接：https://blog.csdn.net/qq_19905867/article/details/121972515

版权

冒泡排序
九九乘法表
水仙花数
猜数字
时间戳
7的倍数,含有7就拍桌子
goto
数组名
最大值
数组逆序
二维数组的数组名
成绩求和
两数相加函数
值传递,形参发生任何改变,都不会影响实参
函数的定义
函数的声明和定义
函数的分文件编写
指针的定义和使用
指针所占内存空间
空指针
野指针
const 修饰指针
指针和数组
值传递和地址传递
指针配合数组和函数
结构体定义和使用
结构体数组
结构体指针
结构体嵌套结构体
结构体做函数参数
结构体中const的使用场景
结构体案例1
结构体案例2
MPI并行——梯形积分
MPI并行——局部二分自适应区间加密积分
MPI并行——9点平滑（阻塞式通讯）
MPI并行——9点平滑（非阻塞式通讯）
MPI并行——Jacobi迭代解二维 Poisson 方程（待完善）
MPI并行——显式偏移读写

冒泡排序

void Bubble_Sort(int* const array, int array_size) {
	for (int i = 1; i < array_size; i++)
	{
		for (int j = 0; j < array_size - i; j++)
		{
			if (array[j] > array[j + 1])
			{
				int temp = array[j];
				array[j] = array[j + 1];
				array[j + 1] = temp;
			}
		}
	}
}

九九乘法表

#include<iostream>
using namespace std;

int main() {
	
	for (int i = 1; i < 10; i++) {
		for (int j = 1; j <=i; j++) {
		
			cout << j << "*" << i << "=" << i * j << "\t";
		}
		cout <<"\n" << endl;
	}

	system("pause");
	return 0;
}

水仙花数

#include<iostream>
#include<math.h>
using namespace std;

int main() {
	
	int a = 0, b = 0, c = 0;
	for (int i = 100; i < 1000; i++){
		a = i / 100;//百位
		b = i / 10 % 10;//十位
		c=i % 10;//个位
		if (pow(a, 3) + pow(b, 3) + pow(c, 3) == i) {
			cout << "水仙花数: " << i << endl;
		}
	}

	system("pause");
	return 0;
}

猜数字

#include<iostream>
#include<ctime>
using namespace std;

int main() {
	srand((unsigned int)time(NULL));//随机数种子
	int num = rand() % 100 + 1;//生成1~100的一个随机数
	int value;

	while (num) {
		cout << "输入一个数: ";
		cin >> value;

		if (value > num) {
			cout << "猜大了" << endl;
		}
		else if(value < num){
			cout << "猜小了" << endl;
		}
		else {
			cout << "猜对了!" << endl;
			break;
		}
	}

	system("pause");
	return 0;
}

时间戳

#include<iostream>
#include<ctime>
using namespace std;

int main() {

	cout << "时间戳: " << (unsigned int)time(NULL) << endl;

	system("pause");
	return 0;
}

7的倍数,含有7就拍桌子

#include<iostream>
using namespace std;

int main() {

	for (int i = 1; i <= 100; i++) {
		/*
		!(i%7)//7的倍数
		i/10==7//十位上有7
		i%10==7//个位上有7
		*/
		if (!(i % 7) || i / 10 == 7 || i % 10 == 7) {
			cout << "拍桌子(" << i << ")" << endl;
		}
		else {
			cout << i << endl;
		}
	}

	system("pause");
	return 0;
}

goto

#include<iostream>
using namespace std;

int main() {

	cout << "1.XXXXXX" << endl;

	cout << "2.XXXXXX" << endl;
	goto FLAG;
	cout << "3.XXXXXX" << endl;

	cout << "4.XXXXXX" << endl;
	FLAG:
	cout << "5.XXXXXX" << endl;

	system("pause");
	return 0;
}

数组名

#include<iostream>
using namespace std;

int main() {

	int arr[] = { 1,2,3,4,5,6,7,8,9,10 };

	cout << "数组在内存中的首地址(十六进制): " << arr << endl;
	cout << "数组在内存中的首地址(十进制): " << (int)arr << endl;
	cout << "数组中的第一个元素在内存中的地址(十进制): " << (int)&arr[0] << endl;
	cout << "数组占用的内存空间: : " << sizeof(arr) << endl;
	cout << "数组中每个元素占用的内存空间 : " << sizeof(arr[0]) << endl;
	cout << "数组元素的个数 : " << sizeof(arr)/sizeof(arr[0]) << endl;

	system("pause");
	return 0;
}

最大值

#include<iostream>
using namespace std;

int main() {
	
	int arr[] = { 300,350,200,400,250 }, max = 0;

	for (int i = 0; i < sizeof(arr) / sizeof(arr[0]); i++) {
		max = arr[i] > max ? arr[i] : max;
	}
	cout << "数组中最大值: " << max << endl;

	system("pause");
	return 0;
}

数组逆序

#include<iostream>
using namespace std;

int main() {
	
	int arr[] = { 1,3,2,5,4,6,2}, temp = 0;

	int num = sizeof(arr) / sizeof(arr[0]) - 1;//数组最后一个元素索引

	for (int i = 0; i <= num/2; i++) {
		temp = arr[i];
		arr[i] = arr[num - i];
		arr[num - i] = temp;
	}

	for (int i = 0; i <= num; i++) {
		cout << arr[i] << endl;
	}

	system("pause");
	return 0;
}

二维数组的数组名

#include<iostream>
using namespace std;

int main() {

	int arr[][3] = {
		{1,2,3},
		{4,5,6}
	};

	cout << "数组大小 " << sizeof(arr) << endl;
	cout << "二维数组第一行大小 " << sizeof(arr[0]) << endl;
	cout << "二维数组元素大小 " << sizeof(arr[0][0]) << endl;

	cout << "二维数组行数 " << sizeof(arr)/ sizeof(arr[0]) << endl;
	cout << "二维数组列数 " << sizeof(arr[0]) / sizeof(arr[0][0]) << endl;

	cout << "二维数组首地址 " << (int)arr << endl;
	cout << "二维数组第一行首地址 " << (int)arr[0] << endl;
	cout << "二维数组第二行首地址 " << (int)arr[1] << endl;

	cout << "二维数组第一个元素首地址 " << (int)&arr[0][0] << endl;
	cout << "二维数组第二个元素首地址 " << (int)&arr[0][1] << endl;
	system("pause");
	return 0;
}

成绩求和

#include<iostream>
#include<string>
using namespace std;

int main() {

	int arr[][3] = {
		{100,100,100},
		{90,50,100},
		{60,70,80}
	};
	
	string name[3] = { "张三","李四","王五" };

	int raw = sizeof(arr) / sizeof(arr[0]);
	int col = sizeof(arr[0]) / sizeof(arr[0][0]);

	for (int i = 0; i < raw; i++) {
		
		int sum = 0;
		for (int j = 0; j < col; j++) {
		
			sum += arr[i][j];
		}
		cout << name[i]  << "的总成绩: " << sum << endl;
	}

	system("pause");
	return 0;
}

两数相加函数

#include<iostream>
using namespace std;

int add(int num1, int num2) {

	return num1 + num2;
}

int main() {

	cout << add(1, 3) << endl;

	system("pause");
	return 0;
}

值传递,形参发生任何改变,都不会影响实参

#include<iostream>
using namespace std;

void swap(int num1, int num2) {

	cout << "交换前" << endl;
	cout << "num1= " << num1 << endl;
	cout << "num2= " << num2 << endl;

	int temp = num1;
	num1 = num2;
	num2 = temp;

	cout << "交换后" << endl;
	cout << "num1= " << num1 << endl;
	cout << "num2= " << num2 << endl;
}

int main() {

	int a = 10, b = 20;

	cout << "a= " << a << endl;
	cout << "b " << b << endl;

	swap(a, b);

	cout << "a= " << a << endl;
	cout << "b " << b << endl;

	system("pause");
	return 0;
}

函数的定义

#include<iostream>
using namespace std;

//在main()之前定义
int max(int num1, int num2) {
	return num1 > num2 ? num1 : num2;
}

int main() {

	cout << max(10, 20) << endl;

	system("pause");
	return 0;
}

函数的声明和定义

#include<iostream>
using namespace std;

//在main()之前声明
int max(int num1, int num2);

int main() {

	cout << max(10, 20) << endl;

	system("pause");
	return 0;
}

//在main()之后定义
int max(int num1, int num2) {
	return num1 > num2 ? num1 : num2;
}

函数的分文件编写

创建.h的头文件
创建.cpp的源文件
在头文件中写函数声明
在源文件中写函数的定义

//swap.h
#pragma once
#include<iostream>
using namespace std;

void swap(int a, int b);

//swap.cpp
#include "swap.h"

void swap(int a, int b) {

	int temp = a;
	a = b;
	b = temp;

	cout << "a= " << a << endl;
	cout << "b= " << b << endl;
}

//source.cpp
#include<iostream>
using namespace std;
#include "swap.h"


int main() {

	swap(10, 20);

	system("pause");
	return 0;
}

指针的定义和使用

#include<iostream>
using namespace std;

int main() {

	int a = 10;
	//1.定义指针
	int* p = &a;
	cout << "a的地址: " << &a << endl;
	cout << "指针p为: " << p << endl;

	//2.使用指针
	*p = 100;
	cout << "a= " << a << endl;
	cout << "*p= " << *p << endl;

	system("pause");
	return 0;
}

指针所占内存空间

#include<iostream>
using namespace std;

int main() {

	int a = 10;
	int* p = &a;

	cout << "a= " << *p << endl;
	cout << "p " << sizeof(p) << endl;//指针所占内存空间
	cout << "*p " << sizeof(*p) << endl;//数据所占内存空间
	cout << "char* " << sizeof(char*) << endl;
	cout << "float* " << sizeof(float*) << endl;
	cout << "double* " << sizeof(double*) << endl;

	system("pause");
	return 0;
}

空指针

#include<iostream>
using namespace std;

int main() {

	int* p = NULL;//空指针

	system("pause");
	return 0;
}

野指针

#include<iostream>
using namespace std;

int main() {
	//野指针
	//int* p = (int*)0x1100;//错误

	system("pause");
	return 0;
}

const 修饰指针

#include<iostream>
using namespace std;

int main() {
	//const修饰的值不能改

	int a = 10, b = 20;
	//1.常量指针
	const int* p1 = &a;
	p1 = &b;//正确

	//2.指针常量
	int* const p2 = &b;
	*p2 = a;

	//3.const 修饰指针和常量
	const int* const p3 = &a;

	system("pause");
	return 0;
}

指针和数组

#include<iostream>
using namespace std;

int main() {
	
	int arr[] = { 1,2,3,4,5,6,7,8,9,10 };

	int* p = arr;

	cout << "第一个元素 " << arr[0] << endl;
	cout << "指针访问第一个元素 " << *p << endl;

	for (int i = 0; i < sizeof(arr) / sizeof(arr[0]); i++, p++) {
	
		cout << *p << endl;
	}

	system("pause");
	return 0;
}

值传递和地址传递

#include<iostream>
using namespace std;

//值传递
void swap01(int a, int b) {

	int temp = a;
	a = b;
	b = temp;
}

//地址传递
void swap02(int* p1, int* p2) {

	int temp = *p1;
	*p1 = *p2;
	*p2 = temp;
}

int main() {
	int a = 10, b = 20;
	cout << "a= " << a << endl;
	cout << "b= " << b << endl;

	cout << "值传递" << endl;
	swap01(a, b);
	cout << "a= " << a << endl;
	cout << "b= " << b << endl;

	cout << "地址传递" << endl;
	swap02(&a, &b);
	cout << "a= " << a << endl;
	cout << "b= " << b << endl;

	system("pause");
	return 0;
}

指针配合数组和函数

#include<iostream>
using namespace std;

void bubleSort(int* arr, int len) {
	
	for (int i = 0; i < len; i++) {
	
		for (int j = 0; j < len - i - 1; j++) {
		
			if (arr[j] > arr[j + 1]) {
			
				int temp = arr[j];
				arr[j] = arr[j + 1];
				arr[j + 1] = temp;
			}
		}
	}
}

int main() {

	int arr[] = {4,3,6,9,1,2,10,8,7,5};
	int len = sizeof(arr) / sizeof(arr[0]);

	bubleSort(arr, len);

	for (int i = 0; i < len; i++) {
	
		cout << arr[i] << "\t";
		
	}
	cout <<endl;

	system("pause");
	return 0;
}

结构体定义和使用

#include<iostream>
#include<string>
using namespace std;

struct Student {
	string name;
	int age;
	int score;
};

int main() {
	
	Student s1;
	s1.name = "张三";
	s1.age = 18;
	s1.score = 60;

	Student s2 = { "李四",19,70 };

	cout << "姓名: " << s2.name 
		 << " 年龄: " << s2.age 
		 << " 成绩: " << s2.score << endl;

	system("pause");
	return 0;
}

结构体数组

#include<iostream>
#include<string>
using namespace std;

struct Student {
	string name;
	int age;
	int score;
};

int main() {
	
	Student stuArray[3] = {
		{"张三",18,60},
		{"李四",19,70},
		{"王五",20,80}
	};

	for (int i = 0; i < 3; i++) {
		cout << "姓名: " << stuArray[i].name
			 << " 年龄: " << stuArray[i].age
			 << " 成绩: " << stuArray[i].score << endl;
	}

	system("pause");
	return 0;
}

结构体指针

#include<iostream>
#include<string>
using namespace std;

struct Student {
	string name;
	int age;
	int score;
};

int main() {
	
	Student s1 = {"张三",18,60};
	cout << "姓名: " << s1.name
		<< " 年龄: " << s1.age
		<< " 分数: " << s1.score << endl;
	
	Student* p = &s1;
	cout << "姓名: " << p->name
		<< " 年龄: " << p->age
		<< " 分数: " << p->score << endl;

	system("pause");
	return 0;
}

结构体嵌套结构体

#include<iostream>
#include<string>
using namespace std;

struct Student {
	string name;
	int age;
	int score;
};

struct Teacher {
	int id;
	string name;
	int age;
	Student stuArray[2];
};

int main() {
	Teacher t;
	t.name = "老王";
	t.id = 001;
	t.age = 50;

	t.stuArray[0].name = "张三";
	t.stuArray[0].age = 18;
	t.stuArray[0].score = 60;

	t.stuArray[1].name = "李四";
	t.stuArray[1].age = 19;
	t.stuArray[1].score = 70;

	system("pause");
	return 0;
}

结构体做函数参数

#include<iostream>
#include<string>
using namespace std;

struct Student {
	string name;
	int age;
	int score;
};

void print1(Student stu) {
	cout << "值传递打印 " << " 姓名: " << stu.name
		<< " 年龄: " << stu.age
		<< " 分数: " << stu.score << endl;
};

void print2(Student* p) {
	p->name = "李四";
	cout << "地址传递打印 " << " 姓名: " << p->name
		<< " 年龄: " << p->age
		<< " 分数: " << p->score << endl;
};

int main() {
	
	Student s = { "张三",18,60 };
	cout << "main函数中打印 " << " 姓名: " << s.name
		<< " 年龄: " << s.age
		<< " 分数: " << s.score << endl;

	print1(s);
	print2(&s);

	system("pause");
	return 0;
}

结构体中const的使用场景

#include<iostream>
#include<string>
using namespace std;

struct Student {
	string name;
	int age;
	int score;
};

void print1(const Student* p) {
	cout << "地址传递打印 " << " 姓名: " << p->name
		<< " 年龄: " << p->age
		<< " 分数: " << p->score << endl;
};

int main() {
	
	Student s = { "张三",18,60 };
	cout << "main函数中打印 " << " 姓名: " << s.name
		<< " 年龄: " << s.age
		<< " 分数: " << s.score << endl;

	print1(&s);

	system("pause");
	return 0;
}

结构体案例1

#include<iostream>
#include<string>
#include<ctime>
using namespace std;

struct Student {
	string sName;
	int score;
};

struct Teacher
{
	string tName;
	Student sArray[5];
};

void allocatespace(Teacher tArray[], int len) {
	string nameSeed = "ABCDE";
	for (int i = 0; i < len; i++) {
		tArray[i].tName = "teacher_";
		tArray[i].tName += nameSeed[i];
		for (int j = 0; j < 5; j++) {
			tArray[i].sArray[j].sName = "student_";
			tArray[i].sArray[j].sName += nameSeed[j];

			int random = 60 + rand() % 41;
			tArray[i].sArray[j].score = random;
		}
	}
};

void printinfo( Teacher tArray[],int len) {
	for (int i = 0; i < len; i++) {
		cout << "老师姓名: " << tArray[i].tName << endl;
		for (int j = 0; j < 5; j++) {
			cout << "\t学生姓名: " << tArray[i].sArray[j].sName
				<< " 得分: " << tArray[i].sArray[j].score << endl;
		}
	}
};

int main() {
	
	srand((unsigned int)time(NULL));

	Teacher tArray[3];
	int len = sizeof(tArray) / sizeof(tArray[0]);
	allocatespace(tArray, len);

	printinfo(tArray, len);
	system("pause");
	return 0;
}

结构体案例2

#include<iostream>
#include<string>
using namespace std;

struct Hero {

	string name;
	int age;
	string sex;
};

void bubbleSorrt(Hero HeroArray[], int len) {
	for (int i = 0; i < len - 1; i++) {
		for (int j = 0; j < len - 1 - i; j++) {
			if (HeroArray[j].age > HeroArray[j + 1].age) {
				Hero temp = HeroArray[j];
				HeroArray[j] = HeroArray[j + 1];
				HeroArray[j + 1] = temp;
			}
		}
	}
};

void printinfo(Hero HeroArray[], int len) {
	for (int i = 0; i < len; i++) {
		cout << HeroArray[i].name << "\t"
			<< HeroArray[i].age << "\t"
			<< HeroArray[i].sex << endl;
	}
};

int main() {
	
	Hero HeroArray[5] = {
		{"刘备",23,"男"},
		{"关羽",22,"男"},
		{"张飞",20,"男"},
		{"赵云",21,"男"},
		{"貂蝉",19,"男"}
	};

	int len = sizeof(HeroArray) / sizeof(HeroArray[0]);
	cout << "\t排序前" << endl;
	printinfo(HeroArray, len);
	
	bubbleSorrt(HeroArray, len);

	cout << "\t排序后" << endl;
	printinfo(HeroArray, len);

	system("pause");
	return 0;
}

MPI并行——梯形积分

#include<iostream>
#include<math.h>
#include "mpi.h"
using namespace std;

double f(double x) {
	return x*x*x;
}

double trap(double a, double b, int n) {

	double h = (b - a) / n;
	double sum = (f(a) + f(b)) / 2;
	for (int i = 1; i < n; i++)
	{
		sum += f(a + i * h);
	}
	return sum * h;
}

int main(void) {
	int n = 9000;
	double a = -2, b = 2;

	int rank, comm_sz;
	MPI_Status status;
	MPI_Init(NULL, NULL);
	MPI_Comm_rank(MPI_COMM_WORLD, &rank);
	MPI_Comm_size(MPI_COMM_WORLD, &comm_sz);

	int local_n = n / comm_sz;
	double local_width = (b - a) / comm_sz;
	double local_a = a + rank * local_width;
	double local_b = local_a + local_width;

	double local_sum = trap(local_a, local_b, local_n);

	double cacha_sum;

	MPI_Reduce(&local_sum, &cacha_sum, 1, MPI_DOUBLE, MPI_SUM, 0, MPI_COMM_WORLD);
	if (rank == 0) {
		cout << "With n = " << n << " trapeziods, our estimate" << endl;
		cout << "of the integral from " << a << " to " << b << " = " << cacha_sum << endl;
	}
	MPI_Finalize();
	return 0;
}

With n = 9000 trapeziods, our estimate
of the integral from -2 to 2 = -8.88178e-16

MPI并行——局部二分自适应区间加密积分

$\int_0^1\frac{4}{1+x^2}\mathrm{d}x=\pi$

#include<iostream>
#include "mpi.h"
using namespace std;

//被积函数
double f(double x) {
	return 4.0 / (1 + x * x);
}

//局部二分自适应区间加密积分函数（梯形）
double trap(double a, double b, double e) {

	double Xc = 0.5 * (a + b), h = b - a, V0 = 0.5 * h * (f(a) + f(b));

	if (Xc == a || Xc == b) return V0;

	double fc = f(Xc), V = (V0 + h * fc) * 0.5, sub_e = abs(V - V0);
	if (sub_e >= 3 * h * e)
	{
		return trap(a, Xc, e) + trap(Xc, b, e);
	}
	else return V;
}

int main(void) {
	double A[4] = { 0,0,0,0 };//下限，上限，误差，积分结果
	double PI = 3.141592653589793;//pi精确值

	int rank, comm_sz;
	MPI_Init(NULL, NULL);
	MPI_Comm_rank(MPI_COMM_WORLD, &rank);
	MPI_Comm_size(MPI_COMM_WORLD, &comm_sz);

	if (rank == 0)
	{
		cout << "积分下限：" << "\t";
		cin >> A[0];
		cout << "积分上限：" << "\t";
		cin >> A[1];
		if (A[0] == A[1]) return 0;
		cout << "误差：" << "\t";
		cin >> A[2];
	}
	MPI_Bcast(A, 3, MPI_DOUBLE, 0, MPI_COMM_WORLD);

	//各区域的积分上下限、允许的误差
	double local_a = A[0] + (rank + 0.) * (A[1] - A[0]) / comm_sz;
	double local_b = A[0] + (rank + 1.) * (A[1] - A[0]) / comm_sz;
	double local_e = A[2] / comm_sz;

	//各区域积分值
	double cache_sum = trap(local_a, local_b, local_e);

	//归约到进程0
	MPI_Reduce(&cache_sum, &A[3], 1, MPI_DOUBLE, MPI_SUM, 0, MPI_COMM_WORLD);

	if (rank == 0) {
		cout << "\n从 " << A[0] << " 到 " << A[1] << " ，误差小于 " << A[2] << " 的积分值为：" << endl;
		printf("%.15lf，与真值差%.1e\n", A[3], abs(A[3] - PI));

	}
	MPI_Finalize();
	return 0;
}

MPI并行——9点平滑（阻塞式通讯）

#include<iostream>
#include "mpi.h"
using namespace std;

int main(void) {
	const int rowsize = 250, colsize = 1000;

	//定义二维数组
	float** a = new float* [rowsize + 2], ** b = new float* [rowsize + 2];

	for (int i = 0; i < rowsize + 2; i++)
	{
		a[i] = new float[colsize + 2];
		b[i] = new float[colsize + 2];
	}

	//数组a整体赋值为1
	for (int i = 0; i < rowsize + 2; i++)
	{
		for (int j = 0; j < colsize + 2; j++)
		{
			a[i][j] = 1;
		}
	}

	int rank, comm_sz;
	MPI_Status status;
	MPI_Init(NULL, NULL);
	MPI_Comm_rank(MPI_COMM_WORLD, &rank);
	MPI_Comm_size(MPI_COMM_WORLD, &comm_sz);

	//应使用4个进程并行
	if (comm_sz != 4)
	{
		printf("*** This program uses exactly 4 processes！Not %d ***", comm_sz);
		MPI_Abort(MPI_COMM_WORLD, 1);
	}

	//边界赋值
	for (int i = 0; i < rowsize + 2; i++) {
		a[i][0] = 10;//左边界
		a[i][colsize + 1] = 10;//右边界
	}

	if (rank == 0)
	{
		for (int i = 0; i < colsize + 2; i++)
		{
			a[0][i] = 10;//上边界
		}
	}

	if (rank == 3)
	{
		for (int i = 0; i < colsize + 2; i++)
		{
			a[rowsize + 1][i] = 10;//下边界
		}
	}

	//b=a
	for (int i = 0; i < rowsize + 2; i++)
	{
		for (int j = 0; j < colsize + 2; j++)
		{
			b[i][j] = a[i][j];
		}
	}

	int before_rank = rank > 0 ? rank - 1 : MPI_PROC_NULL;
	int after_rank = rank < 3 ? rank + 1 : MPI_PROC_NULL;

	int itag1 = 321;
	int itag2 = 123;

	for (int times = 0; times < 100; times++)
	{
		MPI_Send(a[1], colsize + 2, MPI_FLOAT, before_rank, itag1, MPI_COMM_WORLD);
		MPI_Recv(a[rowsize + 1], colsize + 2, MPI_FLOAT, after_rank, itag1, MPI_COMM_WORLD, &status);

		MPI_Send(a[rowsize], colsize + 2, MPI_FLOAT, after_rank, itag2, MPI_COMM_WORLD);
		MPI_Recv(a[0], colsize + 2, MPI_FLOAT, before_rank, itag2, MPI_COMM_WORLD, &status);

		for (int i = 1; i < rowsize + 1; i++)
		{
			for (int j = 1; j < colsize + 1; j++)
			{	//9点平滑
				b[i][j] = (a[i][j] + a[i + 1][j] + a[i - 1][j] + a[i][j + 1] + a[i][j - 1] + a[i - 1][j - 1] + a[i - 1][j + 1] + a[i + 1][j + 1] + a[i + 1][j - 1]) / 9;
			}
		}

		for (int i = 1; i < rowsize + 1; i++)
		{
			for (int j = 1; j < colsize + 1; j++)
			{
				a[i][j] = b[i][j];
			}
		}
	}

	if (rank == 0)
	{
		cout << "Process " << rank << "\n " << endl;
		for (int i = 0; i < 10; i++)
		{
			for (int j = 0; j < 10; j++)
			{
				cout << a[i][j] << "\t";
			}
			cout << endl;
		}
	}

	//释放内存
	delete[]a;
	delete[]b;

	MPI_Finalize();
	return 0;
}

MPI并行——9点平滑（非阻塞式通讯）

#include<iostream>
#include "mpi.h"
using namespace std;

int main(void) {
	const int rowsize = 250, colsize = 1000;

	//定义二维数组
	float** a = new float* [rowsize + 2], ** b = new float* [rowsize + 2];

	for (int i = 0; i < rowsize + 2; i++)
	{
		a[i] = new float[colsize + 2];
		b[i] = new float[colsize + 2];
	}

	//数组a整体赋值为1
	for (int i = 0; i < rowsize + 2; i++)
	{
		for (int j = 0; j < colsize + 2; j++)
		{
			a[i][j] = 1;
		}
	}

	int rank, comm_sz;
	MPI_Status status[4];
	MPI_Request req[4];
	MPI_Init(NULL, NULL);
	MPI_Comm_rank(MPI_COMM_WORLD, &rank);
	MPI_Comm_size(MPI_COMM_WORLD, &comm_sz);

	//应使用4个进程并行
	if (comm_sz != 4)
	{
		printf("*** This program uses exactly 4 processes！Not %d ***", comm_sz);
		MPI_Abort(MPI_COMM_WORLD, 1);
	}

	//边界赋值
	for (int i = 0; i < rowsize + 2; i++) {
		a[i][0] = 10;//左边界
		a[i][colsize + 1] = 10;//右边界
	}

	if (rank == 0)
	{
		for (int i = 0; i < colsize + 2; i++)
		{
			a[0][i] = 10;//上边界
		}
	}

	if (rank == 3)
	{
		for (int i = 0; i < colsize + 2; i++)
		{
			a[rowsize + 1][i] = 10;//下边界
		}
	}

	//b=a
	for (int i = 0; i < rowsize + 2; i++)
	{
		for (int j = 0; j < colsize + 2; j++)
		{
			b[i][j] = a[i][j];
		}
	}

	int before_rank = rank > 0 ? rank - 1 : MPI_PROC_NULL;
	int after_rank = rank < 3 ? rank + 1 : MPI_PROC_NULL;

	int itag1 = 321;
	int itag2 = 123;

	for (int times = 0; times < 100; times++)
	{
		MPI_Isend(a[1], colsize + 2, MPI_FLOAT, before_rank, itag1, MPI_COMM_WORLD, &req[0]);
		MPI_Irecv(a[rowsize + 1], colsize + 2, MPI_FLOAT, after_rank, itag1, MPI_COMM_WORLD, &req[1]);

		MPI_Isend(a[rowsize], colsize + 2, MPI_FLOAT, after_rank, itag2, MPI_COMM_WORLD, &req[2]);
		MPI_Irecv(a[0], colsize + 2, MPI_FLOAT, before_rank, itag2, MPI_COMM_WORLD, &req[3]);

		//不需要通讯的部分 9点平滑
		for (int i = 2; i < rowsize; i++)
		{
			for (int j = 1; j < colsize + 1; j++)
			{	
				b[i][j] = (a[i][j] + a[i + 1][j] + a[i - 1][j] + a[i][j + 1] + a[i][j - 1] + a[i - 1][j - 1] + a[i - 1][j + 1] + a[i + 1][j + 1] + a[i + 1][j - 1]) / 9;
			}
		}
		MPI_Waitall(4, req, status);

		//非阻塞通讯完成后，开始计算 边的平滑
		int iRow[2] = {1, rowsize};//上、下边索引
		for (int ii = 0; ii < 2; ii++)
		{	
			const int i = iRow[ii];
			for (int j = 1; j < colsize + 1; j++)
			{	
				// 9点平滑
				b[i][j] = (a[i][j] + a[i + 1][j] + a[i - 1][j] + a[i][j + 1] + a[i][j - 1] + a[i - 1][j - 1] + a[i - 1][j + 1] + a[i + 1][j + 1] + a[i + 1][j - 1]) / 9;
			}
		}

		for (int i = 1; i < rowsize + 1; i++)
		{
			for (int j = 1; j < colsize + 1; j++)
			{
				a[i][j] = b[i][j];
			}
		}
	}


	if (rank == 0)
	{
		cout << "Process " << rank << "\n " << endl;
		for (int i = 0; i < 10; i++)
		{
			for (int j = 0; j < 10; j++)
			{
				cout << b[i][j] << "\t";
			}
			cout << endl;
		}
	}

	//释放内存
	delete[]a;
	delete[]b;

	MPI_Finalize();
	return 0;
}

MPI并行——Jacobi迭代解二维 Poisson 方程（待完善）

#include<iostream>
#include<fstream>
#include<math.h>
#include<string>
#include "mpi.h"
using namespace std;

double f(double x, double y) {//f(x,y)
	//DEL^2 U = f(x,y)
	return 4;
}

double Uedge(double x, double y) {//外边界条件

	return x * x + y * y;
}

int main(void) {
	double X[2] = { 0,10 }, Y[2] = { 0,10 };
	const int Nx = 128, Ny = 128;
	double error = 1e-4;

	double dx = (X[1] - X[0]) / Nx, dy = (Y[1] - Y[0]) / Ny;//每格长度

	const int nx = Nx / 2, ny = Ny / 2;//2*2的进程分配

	int rank, comm_sz;
	MPI_Status status[8];
	MPI_Request req[8];
	MPI_Init(NULL, NULL);
	MPI_Comm_rank(MPI_COMM_WORLD, &rank);
	MPI_Comm_size(MPI_COMM_WORLD, &comm_sz);

	//应使用4个进程并行
	if (comm_sz != 4)
	{
		printf("*** This program uses exactly 4 processes！Not %d ***", comm_sz);
		MPI_Abort(MPI_COMM_WORLD, 1);
	}

	//各进程分配的位置
	int top_rank, bottom_rank, left_rank, right_rank;
	if (rank > 1) {
		top_rank = rank - 2;
		bottom_rank = MPI_PROC_NULL;
		if (rank < 3)//rank == 2
		{
			left_rank = MPI_PROC_NULL;
			right_rank = rank + 1;
		}
		else {//rank == 3
			left_rank = rank - 1;
			right_rank = MPI_PROC_NULL;
		}
	}
	else {
		top_rank = MPI_PROC_NULL;
		bottom_rank = rank + 2;
		if (rank < 1)//rank == 0
		{
			left_rank = MPI_PROC_NULL;
			right_rank = rank + 1;
		}
		else {//rank == 1
			
			left_rank = rank - 1;
			right_rank = MPI_PROC_NULL;
		}
	}
	/*cout << "Process " << rank
		<< "\t top_rank " << top_rank
		<< "\t bottom_rank " << bottom_rank
		<< "\t left_rank " << left_rank
		<< "\t right_rank " << right_rank << "\n" << endl;*/

	double** U = new double* [nx + 2], ** temp = new double* [nx + 2];//定义U和temp，用temp储存上一次的U
	for (int i = 0; i < nx + 2; i++)
	{
		U[i] = new double[ny + 2];
		temp[i] = new double[ny + 2];
	}

	for (int i = 0; i < nx + 2; i++)
	{
		for (int j = 0; j < ny + 2; j++)
		{	//中心区域赋值
			U[i][j] = Uedge(X[0] + i * dx, Y[0] + j * dy);

			//外边界赋值
			if (rank == 0) {
				U[0][j] = Uedge(X[0], Y[0] + j * dy);//U(0,y:0->Y/2)
				U[i][0] = Uedge(X[0] + i * dx, Y[0]);//U(x:0->X/2,0)
			}

			if (rank == 1) {
				U[0][j] = Uedge(X[0], (Y[1] - Y[0]) / 2 + j * dy);//U(0,y:Y/2->Y)
				U[i][ny + 1] = Uedge(X[0] + i * dx, Y[1] + dy);//U(x:0->X/2,Y)
			}

			if (rank == 2) {
				U[nx + 1][j] = Uedge((X[1] - X[0]) / 2 + i * dx, Y[0] + j * dy);//U(x:X/2->X,y:0->Y/2)
				U[i][0] = Uedge((X[1] - X[0]) / 2 + i * dx, Y[0]);//U(x:X/2->X,Y)
			}

			if (rank == 3) {
				U[nx + 1][j] = Uedge(X[1] + dx, (Y[1] - Y[0]) / 2 + j * dy);//U(X,y:Y/2->Y)
				U[i][ny + 1] = Uedge((X[1] - X[0]) / 2 + i * dx, Y[1] + dy);//U(x:X/2->X,Y)
			}

			//给临时数组temp赋值
			temp[i][j] = U[i][j];
		}
	}

	double H = -(dx * dx * dy * dy) / 2 / (dx * dx + dy * dy), Hx = -H / dx / dx, Hy = -H / dy * dy;//定义迭代公式系数

	int itag1 = 1234, itag2 = 2341, itag3 = 3412, itag4 = 4123;//消息标签

	double sumError = 0;//储存误差的变量
	bool TorF;//是否终止循环的变量

	int times = 100000;//防止死循环，设置一个迭代上限
	do
	{
		MPI_Isend(U[nx], ny + 2, MPI_DOUBLE, bottom_rank, itag1, MPI_COMM_WORLD, &req[0]);//向下发送
		MPI_Irecv(U[0], ny + 2, MPI_DOUBLE, top_rank, itag1, MPI_COMM_WORLD, &req[1]);

		MPI_Isend(U[1], ny + 2, MPI_DOUBLE, top_rank, itag2, MPI_COMM_WORLD, &req[2]);//向上发送
		MPI_Irecv(U[nx + 1], ny + 2, MPI_DOUBLE, bottom_rank, itag2, MPI_COMM_WORLD, &req[3]);

		double colArrSendRecv[2][nx + 2] = { 0 };//用于发送和接收 列的数组

		//构造向右发送的列
		for (int i = 0; i < nx + 2; i++)
		{
			colArrSendRecv[0][i] = U[i][ny];
		}
		MPI_Isend(colArrSendRecv[0], nx + 2, MPI_DOUBLE, right_rank, itag3, MPI_COMM_WORLD, &req[4]);//向右发送
		MPI_Irecv(colArrSendRecv[1], nx + 2, MPI_DOUBLE, left_rank, itag3, MPI_COMM_WORLD, &req[5]);

		//构造向左发送的列
		for (int i = 0; i < nx + 2; i++)
		{
			colArrSendRecv[0][i] = U[i][1];
		}
		MPI_Isend(colArrSendRecv[0], nx + 2, MPI_DOUBLE, left_rank, itag4, MPI_COMM_WORLD, &req[6]);//向左发送
		MPI_Irecv(colArrSendRecv[1], nx + 2, MPI_DOUBLE, right_rank, itag4, MPI_COMM_WORLD, &req[7]);

		//计算不需要通讯的区域
		for (int i = 2; i < nx; i++)
		{
			for (int j = 2; j < ny; j++)
			{
				temp[i][j] = H * f(X[0] + i * dx, Y[0] + j * dy) + Hx * (U[i - 1][j] + U[i + 1][j]) + Hy * (U[i][j - 1] + U[i][j + 1]);
			}
		}

		MPI_Waitall(8, req, status);
		//列边界赋值
		if (right_rank == -1)
		{
			for (int i = 0; i < nx + 2; i++)
			{
				U[i][0] = colArrSendRecv[1][i];
			}
		}
		else//right_rank != -1
		{
			for (int i = 0; i < nx + 2; i++)
			{
				U[i][ny + 1] = colArrSendRecv[1][i];
			}
		}

		//非阻塞通讯完成后，开始计算 边界处的Uij
		int iRow[2] = { 1, nx };//上、下边索引
		for (int ii = 0; ii < 2; ii++)
		{
			const int i = iRow[ii];
			for (int j = 1; j < ny + 1; j++)
			{
				//计算上、下边的Uij
				temp[i][j] = H * f(X[0] + i * dx, Y[0] + j * dy) + Hx * (U[i - 1][j] + U[i + 1][j]) + Hy * (U[i][j - 1] + U[i][j + 1]);
			}
		}

		int iCol[2] = { 1, ny };//左、右边索引
		for (int i = 1; i < nx + 1; i++)
		{
			for (int jj = 0; jj < 2; jj++)
			{
				const int j = iCol[jj];
				//计算左、右边的Uij
				temp[i][j] = H * f(X[0] + i * dx, Y[0] + j * dy) + Hx * (U[i - 1][j] + U[i + 1][j]) + Hy * (U[i][j - 1] + U[i][j + 1]);
			}
		}

		//U = temp
		for (int i = 1; i < nx + 1; i++)
		{
			for (int j = 1; j < ny + 1; j++)
			{
				sumError += pow(temp[i][j] - U[i][j], 2) / pow(temp[i][j], 2);//相对误差 平方和
				U[i][j] = temp[i][j];
			}
		}
		//cout << "sqrt(sumError) " << sqrt(sumError) << endl;
		//cout << "error " << error << endl;

		TorF = sqrt(sumError) > error ? true : false;

	} while (TorF && times--);

	if (rank == 0)
	{
		cout << "Process " << rank << "\n" << endl;
		for (int i = 0; i < nx + 2; i++)
		{
			for (int j = 0; j < ny + 2; j++)
			{
				cout << U[i][j] << ",";
			}
			cout << endl;
		}
	}

	/*ofstream fout("rank " + to_string(rank) + ".csv");//csv方便查看
	fout << "Process " << rank << "\n" << endl;
	for (int i = 0; i < nx + 2; i++)
	{
		for (int j = 0; j < ny + 2; j++)
		{
			fout << U[i][j] << ",";
		}
		fout << endl;
	}
	fout.close();*/

	delete[]U, temp;

	MPI_Finalize();
	return 0;
}

MPI并行——显式偏移读写

#include<iostream>
#include<mpi.h>
using namespace std;

int main(void) {

	int rank, comm_sz;
	MPI_Init(NULL, NULL);
	MPI_Comm_rank(MPI_COMM_WORLD, &rank);
	MPI_Comm_size(MPI_COMM_WORLD, &comm_sz);
	MPI_Status status[2];
	MPI_File fh,fh1;

	//写文件
	MPI_File_open(MPI_COMM_WORLD, "data.dat", MPI_MODE_CREATE | MPI_MODE_WRONLY, MPI_INFO_NULL, &fh);
	int writeData[2] = { rank * rank ,rank * rank * rank };
	int offset = rank*sizeof(writeData);
	MPI_File_write_at(fh, offset, writeData, 2, MPI_INT, &status[0]);
	MPI_File_close(&fh);
	
	//读文件
	MPI_File_open(MPI_COMM_WORLD, "data.dat", MPI_MODE_RDONLY, MPI_INFO_NULL, &fh1);
	int readData[2] = { 0 };
	MPI_File_read_at(fh1, offset, readData, 2, MPI_INT, &status[1]);
	cout << "Process " << rank << "\treadData\t" << readData[0] << "\t" << readData[1] << endl;
	MPI_File_close(&fh1);

	//收集（组收集MPI_Allgather）
	int gatherData[8];
	MPI_Gather(readData, 2, MPI_INT, gatherData, 2, MPI_INT, 0,MPI_COMM_WORLD);
	if (rank == 0)
	{
		cout << "\nProcess " << rank << " Gather Data" << endl;
		for (int i = 0; i < 8; i++)
		{
			cout << gatherData[i] << "\t";
		}
		cout << "\n" << endl;
	}

	//散发
	int scatterData[2];
	MPI_Scatter(gatherData, 2, MPI_INT, scatterData, 2, MPI_INT, 0, MPI_COMM_WORLD);
	cout << "Process " << rank << " Recv ScatterData\t" << scatterData[0] << "\t" << scatterData[1] << endl;

	//组归约
	int reduceData[2];
	MPI_Allreduce(scatterData, reduceData, 2, MPI_INT, MPI_SUM, MPI_COMM_WORLD);
	if (rank == 0)//rank == 0,1,2,3
 	{
		cout << "\nProcess " << rank << " All reduce Data" << endl;
		for (int i = 0; i < 2; i++)
		{
			cout << reduceData[i] << "\t";
		}
		cout << "\n" << endl;
	}


	MPI_Finalize();
	return 0;
}