Spark应用_PageView_UserView_HotChannel

一、PV

对某一个页面的访问量，在页面中进行刷新一次就是一次pv

PV {p1, (u1,u2,u3,u1,u2,u4…)} 对同一个页面的浏览量进行统计，用户可以重复

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
public class PV_ANA {
public static void main(String[] args) {
SparkConf conf = new SparkConf()
.setAppName("PV_ANA")
.setMaster("local")
.set("spark.testing.memory", "2147480000");
JavaSparkContext sc = new JavaSparkContext(conf);
JavaRDD<String> logRDD = sc.textFile("f:/userLog");
String str = "View";
final Broadcast<String> broadcast = sc.broadcast(str);
pvAnalyze(logRDD, broadcast);
}
 
private static void pvAnalyze(JavaRDD<String> logRDD,
final Broadcast<String> broadcast) {
JavaRDD<String> filteredLogRDD = logRDD.filter
(new Function<String, Boolean>() {
private static final long serialVersionUID = 1L;
 
@Override
public Boolean call(String s) throws Exception {
String actionParam = broadcast.value();
String action = s.split("\t")[5];
return actionParam.equals(action);
}
});
JavaPairRDD<String, String> pariLogRDD = filteredLogRDD.mapToPair
(new PairFunction<String, String, String>() {
private static final long serialVersionUID = 1L;
 
@Override
public Tuple2<String, String> call(String s)
throws Exception {
String pageId = s.split("\t")[3];
return new Tuple2<String, String>(pageId, null);
}
});
pariLogRDD.groupByKey().foreach(new VoidFunction
<Tuple2<String, Iterable<String>>>() {
private static final long serialVersionUID = 1L;
 
@Override
public void call(Tuple2<String, Iterable<String>> tuple)
throws Exception {
String pageId = tuple._1;
Iterator<String> iterator = tuple._2.iterator();
long count = 0L;
while (iterator.hasNext()) {
iterator.next();
count++;
}
System.out.println("PAGEID:" + pageId + "\t PV_COUNT:" + count);
}
});
}
 
}

二、UV

UV {p1, (u1,u2,u3,u4,u5…)} 对一个页面有多少用户访问，用户不可以重复

【方式一】

【流程图】

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
public class UV_ANA {
public static void main(String[] args) {
SparkConf conf = new SparkConf()
.setAppName("UV_ANA")
.setMaster("local")
.set("spark.testing.memory", "2147480000");
JavaSparkContext sc = new JavaSparkContext(conf);
JavaRDD<String> logRDD = sc.textFile("f:/userLog");
String str = "View";
final Broadcast<String> broadcast = sc.broadcast(str);
uvAnalyze(logRDD, broadcast);
}
 
private static void uvAnalyze(JavaRDD<String> logRDD,
final Broadcast<String> broadcast) {
JavaRDD<String> filteredLogRDD = logRDD.filter
(new Function<String, Boolean>() {
private static final long serialVersionUID = 1L;
 
@Override
public Boolean call(String s) throws Exception {
String actionParam = broadcast.value();
String action = s.split("\t")[5];
return actionParam.equals(action);
}
});
JavaPairRDD<String, String> pairLogRDD = filteredLogRDD.mapToPair
(new PairFunction<String, String, String>() {
private static final long serialVersionUID = 1L;
 
@Override
public Tuple2<String, String> call(String s) throws Exception {
String pageId = s.split("\t")[3];
String userId = s.split("\t")[2];
return new Tuple2<String, String>(pageId, userId);
}
});
pairLogRDD.groupByKey().foreach(new VoidFunction
<Tuple2<String, Iterable<String>>>() {
private static final long serialVersionUID = 1L;
 
@Override
public void call(Tuple2<String, Iterable<String>> tuple)
throws Exception {
String pageId = tuple._1;
Iterator<String> iterator = tuple._2.iterator();
Set<String> userSets = new HashSet<>();
while (iterator.hasNext()) {
String userId = iterator.next();
userSets.add(userId);
}
System.out.println("PAGEID:" + pageId + "\t " +
"UV_COUNT:" + userSets.size());
}
});
}
}

【方式二】

【流程图】

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
public class UV_ANAoptz {
public static void main(String[] args) {
SparkConf conf = new SparkConf()
.setAppName("UV_ANAoptz")
.setMaster("local")
.set("spark.testing.memory", "2147480000");
JavaSparkContext sc = new JavaSparkContext(conf);
JavaRDD<String> logRDD = sc.textFile("f:/userLog");
String str = "View";
final Broadcast<String> broadcast = sc.broadcast(str);
uvAnalyzeOptz(logRDD, broadcast);
}
 
private static void uvAnalyzeOptz(JavaRDD<String> logRDD,
final Broadcast<String> broadcast) {
JavaRDD<String> filteredLogRDD = logRDD.filter
(new Function<String, Boolean>() {
private static final long serialVersionUID = 1L;
 
@Override
public Boolean call(String s) throws Exception {
String actionParam = broadcast.value();
String action = s.split("\t")[5];
return actionParam.equals(action);
}
});
JavaPairRDD<String, String> pairRDD = filteredLogRDD.mapToPair
(new PairFunction<String, String, String>() {
private static final long serialVersionUID = 1L;
 
@Override
public Tuple2<String, String> call(String s)
throws Exception {
String pageId = s.split("\t")[3];
String userId = s.split("\t")[2];
return new Tuple2<String, String>(pageId + "_" +
userId, null);
}
});
JavaPairRDD<String, Iterable<String>> groupUp2LogRDD = pairRDD.groupByKey();
Map<String, Object> countByKey = groupUp2LogRDD.mapToPair
(new PairFunction<Tuple2<String, Iterable<String>>,
String, String>() {
private static final long serialVersionUID = 1L;
 
@Override
public Tuple2<String, String> call(Tuple2<String,
Iterable<String>> tuple)
throws Exception {
String pu = tuple._1;
String[] spilted = pu.split("_");
String pageId = spilted[0];
return new Tuple2<String, String>(pageId, null);
}
}).countByKey();
Set<String> keySet = countByKey.keySet();
for (String key : keySet) {
System.out.println("PAGEID:" + key + "\tUV_COUNT：" +
countByKey.get(key));
}
}
}

三、热门版块下用户访问的数量

统计出热门版块中最活跃的top3用户。

【方式一】

【流程图】

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
public class HotChannel {
public static void main(String[] args) {
SparkConf conf = new SparkConf()
.setAppName("HotChannel")
.setMaster("local")
.set("spark.testing.memory", "2147480000");
JavaSparkContext sc = new JavaSparkContext(conf);
JavaRDD<String> logRDD = sc.textFile("f:/userLog");
String str = "View";
final Broadcast<String> broadcast = sc.broadcast(str);
hotChannel(sc, logRDD, broadcast);
}
 
private static void hotChannel(JavaSparkContext sc, JavaRDD<String> logRDD,
final Broadcast<String> broadcast) {
JavaRDD<String> filteredLogRDD = logRDD.filter
(new Function<String, Boolean>() {
private static final long serialVersionUID = 1L;
 
@Override
public Boolean call(String v1) throws Exception {
String actionParam = broadcast.value();
String action = v1.split("\t")[5];
return actionParam.equals(action);
}
});
JavaPairRDD<String, String> channel2nullRDD = filteredLogRDD.mapToPair
(new PairFunction<String, String, String>() {
private static final long serialVersionUID = 1L;
 
@Override
public Tuple2<String, String> call(String s) throws Exception {
String channel = s.split("\t")[4];
return new Tuple2<String, String>(channel, null);
}
});
Map<String, Object> channelPVMap = channel2nullRDD.countByKey();
Set<String> keySet = channelPVMap.keySet();
List<SortObj> channels = new ArrayList<>();
for (String channel : keySet) {
channels.add(new SortObj(channel, Integer.valueOf
(channelPVMap.get(channel) + "")));
}
Collections.sort(channels, new Comparator<SortObj>() {
 
@Override
public int compare(SortObj o1, SortObj o2) {
return o2.getValue() - o1.getValue();
}
});
List<String> hotChannelList = new ArrayList<>();
for (int i = 0; i < 3; i++) {
hotChannelList.add(channels.get(i).getKey());
}
for (String channel : hotChannelList) {
System.out.println("channel:" + channel);
}
final Broadcast<List<String>> hotChannelListBroadcast =
sc.broadcast(hotChannelList);
JavaRDD<String> filterRDD = logRDD.filter(new Function<String, Boolean>() {
@Override
public Boolean call(String s) throws Exception {
List<String> hostChannels = hotChannelListBroadcast.value();
String channel = s.split("\t")[4];
String userId = s.split("\t")[2];
return hostChannels.contains(channel) && !"null".equals(userId);
}
});
JavaPairRDD<String, String> channel2UserRDD = filterRDD.mapToPair
(new PairFunction<String, String, String>() {
@Override
public Tuple2<String, String> call(String s)
throws Exception {
String[] splited = s.split("\t");
String channel = splited[4];
String userId = splited[2];
return new Tuple2<String, String>(channel, userId);
}
});
channel2UserRDD.groupByKey().foreach(new VoidFunction
<Tuple2<String, Iterable<String>>>() {
@Override
public void call(Tuple2<String, Iterable<String>> tuple)
throws Exception {
String channel = tuple._1;
Iterator<String> iterator = tuple._2.iterator();
Map<String, Integer> userNumMap = new HashMap<>();
while (iterator.hasNext()) {
String userId = iterator.next();
Integer count = userNumMap.get(userId);
if (count == null) {
count = 1;
} else {
count++;
}
userNumMap.put(userId, count);
}
List<SortObj> lists = new ArrayList<>();
Set<String> keys = userNumMap.keySet();
for (String key : keys) {
lists.add(new SortObj(key, userNumMap.get(key)));
}
Collections.sort(lists, new Comparator<SortObj>() {
 
@Override
public int compare(SortObj O1, SortObj O2) {
return O2.getValue() - O1.getValue();
}
});
System.out.println("HOT_CHANNEL:" + channel);
for (int i = 0; i < 3; i++) {
SortObj sortObj = lists.get(i);
System.out.println(sortObj.getKey() + "=="
+ sortObj.getValue());
}
}
});
}
}

【方式二】

【流程图】

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
public class HotChannelOpz {
public static void main(String[] args) {
SparkConf conf = new SparkConf()
.setAppName("hotChannelOpz")
.setMaster("local")
.set("spark.testing.memory", "2147480000");
JavaSparkContext sc = new JavaSparkContext(conf);
JavaRDD<String> logRDD = sc.textFile("f:/userLog");
String str = "View";
final Broadcast<String> broadcast = sc.broadcast(str);
hotChannelOpz(sc, logRDD, broadcast);
}
 
private static void hotChannelOpz(JavaSparkContext sc, JavaRDD<String> logRDD,
final Broadcast<String> broadcast) {
JavaRDD<String> filteredLogRDD = logRDD.filter
(new Function<String, Boolean>() {
private static final long serialVersionUID = 1L;
 
@Override
public Boolean call(String v1) throws Exception {
String actionParam = broadcast.value();
String action = v1.split("\t")[5];
return actionParam.equals(action);
}
});
 
JavaPairRDD<String, String> channel2nullRDD = filteredLogRDD.mapToPair
(new PairFunction<String, String, String>() {
private static final long serialVersionUID = 1L;
 
@Override
public Tuple2<String, String> call(String val)
throws Exception {
String channel = val.split("\t")[4];
 
return new Tuple2<String, String>(channel, null);
}
});
Map<String, Object> channelPVMap = channel2nullRDD.countByKey();
Set<String> keySet = channelPVMap.keySet();
List<SortObj> channels = new ArrayList<>();
for (String channel : keySet) {
channels.add(new SortObj(channel, Integer.valueOf
(channelPVMap.get(channel) + "")));
}
Collections.sort(channels, new Comparator<SortObj>() {
 
@Override
public int compare(SortObj o1, SortObj o2) {
return o2.getValue() - o1.getValue();
}
});
List<String> hotChannelList = new ArrayList<>();
for (int i = 0; i < 3; i++) {
hotChannelList.add(channels.get(i).getKey());
}
final Broadcast<List<String>> hotChannelListBroadcast =
sc.broadcast(hotChannelList);
JavaRDD<String> filtedRDD = logRDD.filter
(new Function<String, Boolean>() {
 
@Override
public Boolean call(String v1) throws Exception {
List<String> hostChannels = hotChannelListBroadcast.value();
String channel = v1.split("\t")[4];
String userId = v1.split("\t")[2];
return hostChannels.contains(channel) &&
!"null".equals(userId);
}
});
JavaPairRDD<String, String> user2ChannelRDD = filtedRDD.mapToPair
(new PairFunction<String, String, String>() {
private static final long serialVersionUID = 1L;
 
@Override
public Tuple2<String, String> call(String val)
throws Exception {
String[] splited = val.split("\t");
String userId = splited[2];
String channel = splited[4];
return new Tuple2<String, String>(userId, channel);
}
});
JavaPairRDD<String, String> userVistChannelsRDD =
user2ChannelRDD.groupByKey().
flatMapToPair(new PairFlatMapFunction
<Tuple2<String, Iterable<String>>, String, String>() {
private static final long serialVersionUID = 1L;
 
@Override
public Iterable<Tuple2<String, String>> call
(Tuple2<String, Iterable<String>> tuple)
throws Exception {
String userId = tuple._1;
Iterator<String> iterator = tuple._2.iterator();
Map<String, Integer> channelMap = new HashMap<>();
while (iterator.hasNext()) {
String channel = iterator.next();
Integer count = channelMap.get(channel);
if (count == null)
count = 1;
else
count++;
channelMap.put(channel, count);
}
List<Tuple2<String, String>> list = new ArrayList<>();
Set<String> keys = channelMap.keySet();
for (String channel : keys) {
Integer channelNum = channelMap.get(channel);
list.add(new Tuple2<String, String>(channel,
userId + "_" + channelNum));
}
return list;
}
});
userVistChannelsRDD.groupByKey().foreach(new VoidFunction
<Tuple2<String, Iterable<String>>>() {
 
@Override
public void call(Tuple2<String, Iterable<String>> tuple)
throws Exception {
String channel = tuple._1;
Iterator<String> iterator = tuple._2.iterator();
List<SortObj> list = new ArrayList<>();
while (iterator.hasNext()) {
String ucs = iterator.next();
String[] splited = ucs.split("_");
String userId = splited[0];
Integer num = Integer.valueOf(splited[1]);
list.add(new SortObj(userId, num));
}
Collections.sort(list, new Comparator<SortObj>() {
@Override
public int compare(SortObj o1, SortObj o2) {
return o2.getValue() - o1.getValue();
}
});
System.out.println("HOT_CHANNLE:" + channel);
for (int i = 0; i < 3; i++) {
SortObj sortObj = list.get(i);
System.out.println(sortObj.getKey() + "==="
+ sortObj.getValue());
}
}
});
}
}

Spark应用_PageView_UserView_HotChannel的更多相关文章

  1. Spark踩坑记——Spark Streaming+Kafka

    [TOC] 前言 在WeTest舆情项目中,需要对每天千万级的游戏评论信息进行词频统计,在生产者一端,我们将数据按照每天的拉取时间存入了Kafka当中,而在消费者一端,我们利用了spark strea ...

  2. Spark RDD 核心总结

    摘要: 1.RDD的五大属性 1.1 partitions(分区) 1.2 partitioner(分区方法) 1.3 dependencies(依赖关系) 1.4 compute(获取分区迭代列表) ...

  3. spark处理大规模语料库统计词汇

    最近迷上了spark,写一个专门处理语料库生成词库的项目拿来练练手, github地址:https://github.com/LiuRoy/spark_splitter.代码实现参考wordmaker ...

  4. Hive on Spark安装配置详解（都是坑啊）

    个人主页:http://www.linbingdong.com 简书地址:http://www.jianshu.com/p/a7f75b868568 简介 本文主要记录如何安装配置Hive on Sp ...

  5. Spark踩坑记——数据库（Hbase+Mysql）

    [TOC] 前言 在使用Spark Streaming的过程中对于计算产生结果的进行持久化时,我们往往需要操作数据库,去统计或者改变一些值.最近一个实时消费者处理任务,在使用spark streami ...

  6. Spark踩坑记——初试

    [TOC] Spark简介 整体认识 Apache Spark是一个围绕速度.易用性和复杂分析构建的大数据处理框架.最初在2009年由加州大学伯克利分校的AMPLab开发,并于2010年成为Apach ...

  7. Spark读写Hbase的二种方式对比

    作者:Syn良子 出处:http://www.cnblogs.com/cssdongl 转载请注明出处 一.传统方式 这种方式就是常用的TableInputFormat和TableOutputForm ...

  8. （资源整理）带你入门Spark

    一.Spark简介: 以下是百度百科对Spark的介绍: Spark 是一种与 Hadoop 相似的开源集群计算环境,但是两者之间还存在一些不同之处,这些有用的不同之处使 Spark 在某些工作负载方 ...

  9. Spark的StandAlone模式原理和安装、Spark-on-YARN的理解

    Spark是一个内存迭代式运算框架,通过RDD来描述数据从哪里来,数据用那个算子计算,计算完的数据保存到哪里,RDD之间的依赖关系.他只是一个运算框架,和storm一样只做运算,不做存储. Spark ...

随机推荐

  1. 初识QT

    前言:这是写给纯小白看的文章,大神可以自行绕道. QT的优势(摘自360百科): Qt支持下列操作系统: Microsoft Windows 95/98, Microsoft Windows NT, ...

  2. thinkinginjava学习笔记10_容器

    Java中并没有像Perl.Python.Ruby那样对容器有直接的支持,但是可以依靠容器类来完成相同的工作: 泛型 使用一个ArrayList对象可以保存一系列的对象,如: ArrayList ap ...

  3. Android应用程序启动时发生AndroidRuntime : ClassNotFoundException for Activity class的解决方法

    在android应用程序启动时抛出下面异常导致启动失败:07-09 17:12:35.709: ERROR/AndroidRuntime(3866): Uncaught handler: thread ...

  4. openldap 搭建

    环境构建 1)软件安装: yum -y install openldap-servers openldap-clients openldap openldap-devel migrationtools ...

  5. TurnipBit开发板DIY呼吸的吃豆人教程实例

    转载请以链接形式注明文章来源(MicroPythonQQ技术交流群:157816561,公众号:MicroPython玩家汇) 0x00前言 吃豆人是耳熟能详的可爱形象,如今我们的TurnipBit也 ...

  6. onoffswitch-checkbox

    @foreach (EmailSubscription es in Model)   { if(true){ <div class="onoffswitch">     ...

  7. Django入门(一)

    官方网站: 点击 Django 项目是一个python定制框架,它源自一个在线新闻 Web 站点,于 2005 年以开源的形式被释放出来.Django 框架的核心组件有: 用于创建模型的对象关系映射 ...

  8. windows PAE扩展和AWE编程

    在32位windows上只能看到最大3GB的内存空间,而且每个应用程序只能访问4GB的的内存,这个限制是windows独有的,为了使程序能够访问大于4GB的内存空间,需要使用AWE编程接口,同时需要开 ...

  9. Jerry的碎碎念：SAPUI5, Angular, React和Vue

    去年我去一个国内客户现场时,曾经和他们IT部门的一位架构师聊到关于在SAP平台上进行UI应用的二次开发时,UI框架是选用UI5还是Vue这个话题. 我们代表SAP, 向客户推荐使用UI5是基于以下六点 ...

  10. 【转载】从头编写 asp.net core 2.0 web api 基础框架 (4) EF配置

    Github源码地址:https://github.com/solenovex/Building-asp.net-core-2-web-api-starter-template-from-scratc ...