标签:
用Map&Reduce计算几个班级中,每个班级10岁和20岁之间学生的数量:
db.students.insert({classid:1, age:14, name:‘Tom‘})
将classid随机1和2、age在8-25岁之间随机,name在3-7个字符之间随机。
往mrtask库中students写入1000万条数据:
package org.test;
import java.util.ArrayList;
import java.util.List;
import java.util.Random;
import com.mongodb.BasicDBObject;
import com.mongodb.DB;
import com.mongodb.DBCollection;
import com.mongodb.DBCursor;
import com.mongodb.DBObject;
import com.mongodb.MongoClient;
import com.mongodb.ServerAddress;
public class TestMongoDBReplSet {
public static void main(String[] args) {
try {
List<ServerAddress> addresses = new ArrayList<ServerAddress>();
ServerAddress address1 = new ServerAddress("172.16.16.89", 27017);
addresses.add(address1);
MongoClient client = new MongoClient(addresses);
DB db = client.getDB("mrtask");
DBCollection coll = db.getCollection("students");
// 数据写入
BasicDBObject object = new BasicDBObject();
for (int i = 1; i <= 10000000; i++) {
object.append("classid", 1 + (int) (Math.random() * 2));
object.append("age", 8 + (int) (Math.random() * 17));
object.append("name", getName());
coll.insert(object);
object.clear();
}
} catch (Exception e) {
e.printStackTrace();
}
}
public static String getName() {
ArrayList list = new ArrayList();
for (char c = ‘a‘; c <= ‘z‘; c++) {
list.add(c);
}
String str = "";
int end = 3 + (int) (Math.random() * 4);
for (int i = 0; i < end; i++) {
int num = (int) (Math.random() * 26);
str = str + list.get(num);
}
return str;
}
}
经查看,mrtask库中students表中有1000万条的数据:
[root@localhost bin]# ./mongo
MongoDB shell version: 2.6.11
connecting to: test
> show dbs
admin (empty)
local 0.078GB
mrtask 3.952GB
test 0.453GB
> use mrtask
switched to db mrtask
> db.students.find().count()
10000000
> mapfun = function(){emit(this.classid,1)}
> reducefun=function (key, values) { var count = 0; values.forEach(function (v) {count += v;}); return count; }
> ff = function (key, value) { return {classid:key, count:value}; }
> classid_res = db.runCommand({
mapreduce:"students",
map:mapfun,
reduce:reducefun,
out:"students_classid_res",
finalize:ff,
query:{age:{$gt:10,$lt:20}}
});
> db.students_classid_res.find()
{ "_id" : 1, "value" : { "classid" : 1, "count" : 2643128 } }
{ "_id" : 2, "value" : { "classid" : 2, "count" : 2650870 } }
标签:
原文地址:http://www.cnblogs.com/ljai/p/5017864.html