标签:
上一篇介绍了如何使用nginx+iis部署一个简单的分布式系统,文章结尾留下了几个问题,其中一个是"如何解决多站点下Session共享"。这篇文章将会介绍如何使用Redis,下一篇在此基础上实现Session。
这里特别说明一下,其实没有必要使用Redis来解决Session共享。Asp.net提供了StateServer模式来共享Session,这里重复造轮子的目的1:熟悉Redis的基本知识和使用 2.学习和巩固Session的实现原理。
3.学习Redis应用场景
阅读目录
redis是一个key-value存储系统。和Memcached类似,它支持存储的value类型相对更多,包括string(字符串)、list(链表)、set(集合)、zset(sorted set --有序集合)和hash(哈希类型)。这些数据类型都支持push/pop、add/remove及取交集并集和差集及更丰富的操作,而且这些操作都是原子性的。在此基础上,redis支持各种不同方式的排序。与memcached一样,为了保证效率,数据都是缓存在内存中。区别的是redis会周期性的把更新的数据写入磁盘或者把修改操作写入追加的记录文件,并且在此基础上实现了master-slave(主从)同步。
最新版本的redis版本为3.0.3,支持集群功能。我这下载的是window版本的,实际场景都是安装在linux系统下的。下载地址:redis-2.8.19.rar 。更多下载地址:
下载完成之后解压运行redis-server.exe就启动了redis了,启动后会在进程里面看到reids。
1.读写分离配置
redis的读写分离需要修改配置文件,把解压的文件复制了一份。两份文件是一样的,分别命名为MasterRedis-2.8.19(主redis服务),SlaveRedis-2.8.19(从redis服务)。redis默认绑定的是6379端口,
我们保持主服务配置不变,修改从服务配置。
从服务启动以后,主服务会发送一条同步的sync命令,同步从服务器的缓存数据。
服务搭建好以后可以使用.net版本redis操作类库ServiceStack.Redis来操作redis,本文会用到以下三个dll。
初始化RedisClient对象
var client = new RedisClient("120.26.197.185", 6379);
1.String
String是最常用的一种数据类型,普通的key/value存储都可以归为此类,value其实不仅是String,也可以是数字:比如想知道什么时候封锁一个IP地址(访问超过几次)。INCRBY命令让这些变得很容易,通过原子递增保持计数。
#region "字符串类型"
client.Set<string>("name", "laowang");
string userName = client.Get<string>("name");
Console.WriteLine(userName);
//访问次数
client.Set<int>("IpAccessCount", 0);
//次数递增
client.Incr("IpAccessCount");
Console.WriteLine(client.Get<int>("IpAccessCount"));
#endregion
2.Hash
一个hashid可以存储多项信息,每一项信息也有自己的key
client.SetEntryInHash("userInfoId", "name", "zhangsan");
client.SetEntryInHash("userInfoId", "name1", "zhangsan1");
client.SetEntryInHash("userInfoId", "name2", "zhangsan2");
client.SetEntryInHash("userInfoId", "name3", "zhangsan3");
client.GetHashKeys("userInfoId").ForEach(e => Console.WriteLine(e));
client.GetHashValues("userInfoId").ForEach(e => Console.WriteLine(e));
3.List
应用场景:
#region "List类型"
client.AddItemToList("userInfoId1", "123");
client.AddItemToList("userInfoId1", "1234");
Console.WriteLine("List数据项条数:" + client.GetListCount("userInfoId1"));
Console.WriteLine("List数据项第一条数据:" + client.GetItemFromList("userInfoId1", 0));
Console.WriteLine("List所有数据");
client.GetAllItemsFromList("userInfoId1").ForEach(e => Console.WriteLine(e));
#endregion
#region "List类型做为队列和栈使用"
Console.WriteLine(client.GetListCount("userInfoId1"));
//队列先进先出
//Console.WriteLine(client.DequeueItemFromList("userInfoId1"));
//Console.WriteLine(client.DequeueItemFromList("userInfoId1"));
//栈后进先出
Console.WriteLine("出栈"+client.PopItemFromList("userInfoId1"));
Console.WriteLine("出栈"+client.PopItemFromList("userInfoId1"));
#endregion
4.Set
应用场景:
client.AddItemToSet("A", "B");
client.AddItemToSet("A", "C");
client.AddItemToSet("A", "D");
client.AddItemToSet("A", "E");
client.AddItemToSet("A", "F");
client.AddItemToSet("B", "C");
client.AddItemToSet("B", "F");
//求差集
Console.WriteLine("A,B集合差集");
client.GetDifferencesFromSet("A", "B").ToList<string>().ForEach(e => Console.Write(e + ","));
//求集合交集
Console.WriteLine("\nA,B集合交集");
client.GetIntersectFromSets(new string[] { "A", "B" }).ToList<string>().ForEach(e => Console.Write(e + ","));
//求集合并集
Console.WriteLine("\nA,B集合并集");
client.GetUnionFromSets(new string[] { "A", "B" }).ToList<string>().ForEach(e => Console.Write(e + ","));
5.Sort Set(排序)
应用场景:
#region "有序Set操作"
client.AddItemToSortedSet("SA", "B", 2);
client.AddItemToSortedSet("SA", "C", 1);
client.AddItemToSortedSet("SA", "D", 5);
client.AddItemToSortedSet("SA", "E", 3);
client.AddItemToSortedSet("SA", "F", 4);
//有序集合降序排列
Console.WriteLine("\n有序集合降序排列");
client.GetAllItemsFromSortedSetDesc("SA").ForEach(e => Console.Write(e + ","));
Console.WriteLine("\n有序集合升序序排列");
client.GetAllItemsFromSortedSet("SA").ForEach(e => Console.Write(e + ","));
client.AddItemToSortedSet("SB", "C", 2);
client.AddItemToSortedSet("SB", "F", 1);
client.AddItemToSortedSet("SB", "D", 3);
Console.WriteLine("\n获得某个值在有序集合中的排名,按分数的升序排列");
Console.WriteLine(client.GetItemIndexInSortedSet("SB", "D"));
Console.WriteLine("\n获得有序集合中某个值得分数");
Console.WriteLine(client.GetItemScoreInSortedSet("SB", "D"));
Console.WriteLine("\n获得有序集合中,某个排名范围的所有值");
client.GetRangeFromSortedSet("SA", 0, 3).ForEach(e => Console.Write(e + ","));
#endregion
最后提供一份别人写好的Redis操作的帮助类,用到了PooledRedisClientManager连接池来获取RedisClient,同时用到了读写分离的概念,可以直接拿来使用
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using ServiceStack.Redis;
namespace Com.Redis
{
/// <summary>
/// 来源:http://blog.wx6.org/2013/349.htm
/// </summary>
public class RedisBase
{
private static string[] ReadWriteHosts = System.Configuration.ConfigurationSettings.AppSettings["readWriteHosts"].Split(new char[] { ‘;‘ });
private static string[] ReadOnlyHosts = System.Configuration.ConfigurationSettings.AppSettings["readOnlyHosts"].Split(new char[] { ‘;‘ });
#region -- 连接信息 --
public static PooledRedisClientManager prcm = CreateManager(ReadWriteHosts, ReadOnlyHosts);
private static PooledRedisClientManager CreateManager(string[] readWriteHosts, string[] readOnlyHosts)
{
// 支持读写分离,均衡负载
return new PooledRedisClientManager(readWriteHosts, readOnlyHosts, new RedisClientManagerConfig
{
MaxWritePoolSize = 5, // “写”链接池链接数
MaxReadPoolSize = 5, // “读”链接池链接数
AutoStart = true,
});
}
#endregion
#region -- Item --
/// <summary>
/// 设置单体
/// </summary>
/// <typeparam name="T"></typeparam>
/// <param name="key"></param>
/// <param name="t"></param>
/// <param name="timeSpan"></param>
/// <returns></returns>
public static bool Item_Set<T>(string key, T t)
{
try
{
using (IRedisClient redis = prcm.GetClient())
{
return redis.Set<T>(key, t, new TimeSpan(1, 0, 0));
}
}
catch (Exception ex)
{
// LogInfo
}
return false;
}
/// <summary>
/// 获取单体
/// </summary>
/// <typeparam name="T"></typeparam>
/// <param name="key"></param>
/// <returns></returns>
public static T Item_Get<T>(string key) where T : class
{
using (IRedisClient redis = prcm.GetClient())
{
return redis.Get<T>(key);
}
}
/// <summary>
/// 移除单体
/// </summary>
/// <param name="key"></param>
public static bool Item_Remove(string key)
{
using (IRedisClient redis = prcm.GetClient())
{
return redis.Remove(key);
}
}
#endregion
#region -- List --
public static void List_Add<T>(string key, T t)
{
using (IRedisClient redis = prcm.GetClient())
{
var redisTypedClient = redis.GetTypedClient<T>();
redisTypedClient.AddItemToList(redisTypedClient.Lists[key], t);
}
}
public static bool List_Remove<T>(string key, T t)
{
using (IRedisClient redis = prcm.GetClient())
{
var redisTypedClient = redis.GetTypedClient<T>();
return redisTypedClient.RemoveItemFromList(redisTypedClient.Lists[key], t) > 0;
}
}
public static void List_RemoveAll<T>(string key)
{
using (IRedisClient redis = prcm.GetClient())
{
var redisTypedClient = redis.GetTypedClient<T>();
redisTypedClient.Lists[key].RemoveAll();
}
}
public static int List_Count(string key)
{
using (IRedisClient redis = prcm.GetReadOnlyClient())
{
return redis.GetListCount(key);
}
}
public static List<T> List_GetRange<T>(string key, int start, int count)
{
using (IRedisClient redis = prcm.GetReadOnlyClient())
{
var c = redis.GetTypedClient<T>();
return c.Lists[key].GetRange(start, start + count - 1);
}
}
public static List<T> List_GetList<T>(string key)
{
using (IRedisClient redis = prcm.GetReadOnlyClient())
{
var c = redis.GetTypedClient<T>();
return c.Lists[key].GetRange(0, c.Lists[key].Count);
}
}
public static List<T> List_GetList<T>(string key, int pageIndex, int pageSize)
{
int start = pageSize * (pageIndex - 1);
return List_GetRange<T>(key, start, pageSize);
}
/// <summary>
/// 设置缓存过期
/// </summary>
/// <param name="key"></param>
/// <param name="datetime"></param>
public static void List_SetExpire(string key, DateTime datetime)
{
using (IRedisClient redis = prcm.GetClient())
{
redis.ExpireEntryAt(key, datetime);
}
}
#endregion
#region -- Set --
public static void Set_Add<T>(