极客大学架构师训练营系统架构一致性哈希 Consistent Hash 第五次作业

 关注
发布于: 2020 年 07 月 08 日
用你熟悉的编程语言实现一致性 hash 算法。
编写测试用例测试这个算法，测试 100 万 KV 数据，10 个服务器节点的情况下，计算这些 KV 数据在服务器上分布数量的标准差，以评估算法的存储负载不均衡性。
答：
1、首先选用Java的TreeMap，因为它基于红黑树实现，查找效率比较高，而且已经原生提供了ceilingEntry和firstEntry方法；
其次，虚拟结点的Hash计算必须是稳定不变的，否则服务重启可能导致虚拟结点跳变；
还有，过程中用Mock类，代替物理服务器完成缓存的实际读写操作；
如果直接采用String.hashCode，发现计算得到的hash值比较集中，不适合使用，于是去网上另外找了一个hash算法替代。
﻿
2、项目的2个关键流程简画如下：
﻿
﻿
3、类设计图如下：
﻿
﻿
CacheImplement
package hash;
﻿
public class CacheImplement implements Cache {
  private ConsistentHashRing ring = new ConsistentHashRing();
  private String[] servers;
﻿
  public CacheImplement(String[] servers, int virtualFactorPerNode) {
    ring.addServers(servers, virtualFactorPerNode);
  }
﻿
  @Override
  public String get(String key) {
    Cache serverKey = ring.getServer(key);
    return serverKey.get(key);
  }
﻿
  @Override
  public void set(String key, String val) {
    Cache serverKey = ring.getServer(key);
    serverKey.set(key,val);
  }
﻿
  @Override
  public String toString() {
    return ring.toString();
  }
﻿
  static String[] ipServers = new String[] {
      "192.168.0.1",
      "192.168.0.2",
      "192.168.0.3",
      "192.168.0.4",
      "192.168.0.5",
      "192.168.0.6",
      "192.168.0.7",
      "192.168.0.8",
      "192.168.0.9",
      "192.168.0.10"
  };
﻿
  static int maxCacheNumber = 1_000_000;
﻿
  public static void main(String[] args) {
    testCache(10);
    testCache(50);
    testCache(100);
    testCache(150);
    testCache(200);
    testCache(500);
    testCache(1_000);
    testCache(1_500);
    testCache(2_000);
  }
﻿
  static void testCache(int virtualFactorPerNode) {
    Cache cache = new CacheImplement(ipServers, virtualFactorPerNode);
    fillCache(cache, maxCacheNumber);
    System.out.println(cache);
  }
﻿
  static void fillCache(Cache cache, int cacheNumber) {
    for (int i = 0; i < cacheNumber; i++) {
      String key = "key" + i;
      String val = "val" + key;
      cache.set(key, val);
    }
  }
}
﻿
哈希环类 ConsistentHashRing
package hash;
import java.util.*;
public class ConsistentHashRing {
  TreeMap<Integer, Cache> map = new TreeMap<Integer, Cache>();
  List<Cache> servers = new ArrayList<>();
  final String virtualKeyString = "&&Virtual";
  public void addServers(String[] servers, int virtualFactorPerNode) {
    if (servers == null) {
      return;
    }
    for (String server: servers) {
      Cache cacheServer = new CacheServerMock(server);
      map.put(getHashCode(server), cacheServer);
      // add virtual nodes
      for (int i = 1; i <= virtualFactorPerNode; i++) {
        map.put(getVirtualHashCode(server, i), cacheServer);
      }
      this.servers.add(cacheServer);
    }
  }
  private int getVirtualHashCode(String key, int i) {
    key += virtualKeyString + i;
    return getHashCode(key);
  }
  public static int getHashCode(String key) {
    // int ret = key.hashCode(); // result is too close, deprecated
    final int p = 16777619;
    int ret = (int) 2166136261L;
    for (int i = 0; i < key.length(); i++) {
      ret = (ret ^ key.charAt(i)) * p;
    }
    ret += ret << 13;
    ret ^= ret >> 7;
    ret += ret << 3;
    ret ^= ret >> 17;
    ret += ret << 5;
    return ret < 0 ? -ret : ret;
  }
  /**
   * get the server from consistent hash ring
   * @param cacheKey
   * @return
   */
  public Cache getServer(String cacheKey) {
    if (map.size() <= 0) {
      throw  new IllegalArgumentException("You must add server before call this method");
    }
    if (cacheKey == null || cacheKey.isEmpty()) {
      throw new IllegalArgumentException("cacheKey can't be empty");
    }
    int hash = getHashCode(cacheKey);
    Map.Entry<Integer, Cache> server = map.ceilingEntry(hash);
    return server == null ? map.firstEntry().getValue() : server.getValue();
  }
  @Override
  public String toString() {
    int totalNum = 0;
    for (Cache cache : servers) {
      CacheServerMock mock = (CacheServerMock) cache;
      totalNum += mock.getCaches().size();
    }
    int avg = totalNum / servers.size();
    StringBuilder sb = new StringBuilder();
    sb.append("总缓存个数：").append(totalNum)
        .append(" 服务器台数：").append(servers.size())
        .append(" 每服务器虚拟节点数：").append(map.size()/servers.size() - 1)
        .append(" 平均值：").append(avg)
        .append("\r\n");
    calculateStandardDeviation(avg, sb);
    return sb.toString();
  }
  private void calculateStandardDeviation(int avg, StringBuilder sb) {
    // 标准差公式： (每个样本 - 平均值)的平方 / (服务器数 - 1)，结果再取平方根
    long diff = 0;
    for (Cache cache : servers) {
      CacheServerMock mock = (CacheServerMock) cache;
      int size = mock.getCaches().size();
      sb.append("Server-Name:")
          .append(mock.getServerName())
          .append(" Cache-num: ")
          .append(size)
          .append("\r\n");
      long tmp = size - avg;
      diff += (tmp * tmp);
    }
    diff /= (servers.size() - 1);
    sb.append("标准差为： ").append(Math.sqrt(diff)).append("\r\n");
  }
}
﻿
Cache
package hash;
public interface Cache {
  // get cache data by key
  String get(String key);
  // set cache
  void set(String key, String val);
}
﻿
CacheServerMoke
package hash;
import java.util.HashMap;
import java.util.Map;
/**
 * concrete server class
 */
public class CacheServerMock implements Cache {
  private Map<String, String> caches = new HashMap<>();
  private String serverName;
  public CacheServerMock(String serverName) {
    this.serverName = serverName;
  }
  public String getServerName() {
    return serverName;
  }
  public Map<String, String> getCaches() {
    return caches;
  }
  @Override
  public String get(String key) {
    return caches.get(key);
  }
  @Override
  public void set(String key, String val) {
    caches.put(key, val);
  }
}
﻿
输出结果：
/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/bin/java "-javaagent:/Applications/IntelliJ IDEA.app/Contents/lib/idea_rt.jar=64424:/Applications/IntelliJ IDEA.app/Contents/bin" -Dfile.encoding=UTF-8 -classpath /Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/charsets.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/deploy.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/ext/cldrdata.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/ext/dnsns.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/ext/jaccess.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/ext/jfxrt.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/ext/localedata.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/ext/nashorn.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/ext/sunec.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/ext/sunjce_provider.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/ext/sunpkcs11.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/ext/zipfs.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/javaws.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/jce.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/jfr.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/jfxswt.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/jsse.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/management-agent.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/plugin.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/resources.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/jre/lib/rt.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/lib/ant-javafx.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/lib/dt.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/lib/javafx-mx.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/lib/jconsole.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/lib/packager.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/lib/sa-jdi.jar:/Library/Java/JavaVirtualMachines/jdk1.8.0_172.jdk/Contents/Home/lib/tools.jar:/Users/zgpeace/Workspace/Java/awesome-java-leetcode/code/LeetCode/out/production/LeetCode hash.CacheImplement
总缓存个数：1000000 服务器台数：10 每服务器虚拟节点数：10 平均值：100000
Server-Name:192.168.0.1 Cache-num: 93989
Server-Name:192.168.0.2 Cache-num: 114372
Server-Name:192.168.0.3 Cache-num: 137352
Server-Name:192.168.0.4 Cache-num: 137654
Server-Name:192.168.0.5 Cache-num: 64539
Server-Name:192.168.0.6 Cache-num: 126010
Server-Name:192.168.0.7 Cache-num: 80109
Server-Name:192.168.0.8 Cache-num: 96205
Server-Name:192.168.0.9 Cache-num: 79377
Server-Name:192.168.0.10 Cache-num: 70393
标准差为： 27287.81011733994
总缓存个数：1000000 服务器台数：10 每服务器虚拟节点数：50 平均值：100000
Server-Name:192.168.0.1 Cache-num: 112108
Server-Name:192.168.0.2 Cache-num: 116560
Server-Name:192.168.0.3 Cache-num: 85448
Server-Name:192.168.0.4 Cache-num: 105785
Server-Name:192.168.0.5 Cache-num: 87262
Server-Name:192.168.0.6 Cache-num: 90845
Server-Name:192.168.0.7 Cache-num: 89956
Server-Name:192.168.0.8 Cache-num: 133565
Server-Name:192.168.0.9 Cache-num: 89618
Server-Name:192.168.0.10 Cache-num: 88853
标准差为： 16233.25651248079
总缓存个数：1000000 服务器台数：10 每服务器虚拟节点数：100 平均值：100000
Server-Name:192.168.0.1 Cache-num: 91865
Server-Name:192.168.0.2 Cache-num: 117698
Server-Name:192.168.0.3 Cache-num: 87517
Server-Name:192.168.0.4 Cache-num: 92990
Server-Name:192.168.0.5 Cache-num: 90695
Server-Name:192.168.0.6 Cache-num: 88082
Server-Name:192.168.0.7 Cache-num: 108513
Server-Name:192.168.0.8 Cache-num: 114522
Server-Name:192.168.0.9 Cache-num: 109090
Server-Name:192.168.0.10 Cache-num: 99028
标准差为： 11449.97379909666
总缓存个数：1000000 服务器台数：10 每服务器虚拟节点数：150 平均值：100000
Server-Name:192.168.0.1 Cache-num: 98153
Server-Name:192.168.0.2 Cache-num: 116658
Server-Name:192.168.0.3 Cache-num: 96638
Server-Name:192.168.0.4 Cache-num: 100941
Server-Name:192.168.0.5 Cache-num: 98168
Server-Name:192.168.0.6 Cache-num: 92365
Server-Name:192.168.0.7 Cache-num: 95321
Server-Name:192.168.0.8 Cache-num: 94687
Server-Name:192.168.0.9 Cache-num: 117472
Server-Name:192.168.0.10 Cache-num: 89597
标准差为： 9535.620640524663
总缓存个数：1000000 服务器台数：10 每服务器虚拟节点数：200 平均值：100000
Server-Name:192.168.0.1 Cache-num: 99807
Server-Name:192.168.0.2 Cache-num: 102766
Server-Name:192.168.0.3 Cache-num: 101971
Server-Name:192.168.0.4 Cache-num: 103664
Server-Name:192.168.0.5 Cache-num: 95750
Server-Name:192.168.0.6 Cache-num: 94986
Server-Name:192.168.0.7 Cache-num: 100927
Server-Name:192.168.0.8 Cache-num: 90744
Server-Name:192.168.0.9 Cache-num: 120784
Server-Name:192.168.0.10 Cache-num: 88601
标准差为： 8923.409662231135
总缓存个数：1000000 服务器台数：10 每服务器虚拟节点数：500 平均值：100000
Server-Name:192.168.0.1 Cache-num: 107026
Server-Name:192.168.0.2 Cache-num: 100697
Server-Name:192.168.0.3 Cache-num: 98181
Server-Name:192.168.0.4 Cache-num: 101271
Server-Name:192.168.0.5 Cache-num: 96521
Server-Name:192.168.0.6 Cache-num: 97214
Server-Name:192.168.0.7 Cache-num: 99982
Server-Name:192.168.0.8 Cache-num: 93274
Server-Name:192.168.0.9 Cache-num: 106760
Server-Name:192.168.0.10 Cache-num: 99074
标准差为： 4300.3152210041535
总缓存个数：1000000 服务器台数：10 每服务器虚拟节点数：1000 平均值：100000
Server-Name:192.168.0.1 Cache-num: 99394
Server-Name:192.168.0.2 Cache-num: 98163
Server-Name:192.168.0.3 Cache-num: 96362
Server-Name:192.168.0.4 Cache-num: 98742
Server-Name:192.168.0.5 Cache-num: 102872
Server-Name:192.168.0.6 Cache-num: 103469
Server-Name:192.168.0.7 Cache-num: 97127
Server-Name:192.168.0.8 Cache-num: 97118
Server-Name:192.168.0.9 Cache-num: 106913
Server-Name:192.168.0.10 Cache-num: 99840
标准差为： 3386.525210300375
总缓存个数：1000000 服务器台数：10 每服务器虚拟节点数：1500 平均值：100000
Server-Name:192.168.0.1 Cache-num: 101235
Server-Name:192.168.0.2 Cache-num: 96578
Server-Name:192.168.0.3 Cache-num: 101237
Server-Name:192.168.0.4 Cache-num: 94736
Server-Name:192.168.0.5 Cache-num: 105713
Server-Name:192.168.0.6 Cache-num: 99410
Server-Name:192.168.0.7 Cache-num: 100591
Server-Name:192.168.0.8 Cache-num: 100123
Server-Name:192.168.0.9 Cache-num: 101931
Server-Name:192.168.0.10 Cache-num: 98446
标准差为： 3017.90440537801
总缓存个数：1000000 服务器台数：10 每服务器虚拟节点数：2000 平均值：100000
Server-Name:192.168.0.1 Cache-num: 100787
Server-Name:192.168.0.2 Cache-num: 96876
Server-Name:192.168.0.3 Cache-num: 102202
Server-Name:192.168.0.4 Cache-num: 96784
Server-Name:192.168.0.5 Cache-num: 108713
Server-Name:192.168.0.6 Cache-num: 98230
Server-Name:192.168.0.7 Cache-num: 99731
Server-Name:192.168.0.8 Cache-num: 98804
Server-Name:192.168.0.9 Cache-num: 99857
Server-Name:192.168.0.10 Cache-num: 98016
标准差为： 3497.283946150212
Process finished with exit code 0
﻿
参考https://xie.infoq.cn/article/8651e7b5ef05177005b474dfc
﻿
发布于: 2020 年 07 月 08 日阅读数: 43
原文链接:【http://xie.infoq.cn/article/fb3c2581ebfeb29d81fa5d83d】。
John(易筋)

 关注
问渠那得清如许？为有源头活水来 2018.07.17 加入
工作10+年，架构师，曾经阿里巴巴资深无线开发，汇丰银行架构师/专家。开发过日活过亿的淘宝Taobao App，擅长架构、算法、数据结构、设计模式、iOS、Java Spring Boot。易筋为阿里巴巴花名。
发布
暂无评论
极客大学架构师训练营 系统架构 一致性哈希 Consistent Hash 第五次作业

参考

John(易筋)

评论

极客大学架构师训练营系统架构一致性哈希 Consistent Hash 第五次作业
