本文演示以Spark作為分析引擎,Cassandra作為數(shù)據(jù)存儲(chǔ),而使用Spring Boot來(lái)開發(fā)驅(qū)動(dòng)程序的示例。
1.前置條件
- 安裝Spark(本文使用Spark-1.5.1,如安裝目錄為/opt/spark)
- 安裝Cassandra(3.0+)
創(chuàng)建keyspace
|
1
|
CREATE KEYSPACE hfcb WITH REPLICATION = { 'class' : 'SimpleStrategy', 'replication_factor' : 3 }; |
創(chuàng)建table
|
1
2
3
4
5
|
CREATE TABLE person ( id text PRIMARY KEY, first_name text, last_name text); |
插入測(cè)試數(shù)據(jù)
|
1
2
3
4
5
|
insert into person (id,first_name,last_name) values('1','wang','yunfei');insert into person (id,first_name,last_name) values('2','peng','chao');insert into person (id,first_name,last_name) values('3','li','jian');insert into person (id,first_name,last_name) values('4','zhang','jie');insert into person (id,first_name,last_name) values('5','liang','wei'); |
2.spark-cassandra-connector安裝
讓Spark-1.5.1能夠使用Cassandra作為數(shù)據(jù)存儲(chǔ)，需要加上下面jar包的依賴(示例將包放置于 /opt/spark/managed-lib/ 目錄，可任意):
|
1
2
3
4
5
6
7
8
|
cassandra-clientutil-3.0.2.jarcassandra-driver-core-3.1.4.jarguava-16.0.1.jarcassandra-thrift-3.0.2.jar joda-convert-1.2.jarjoda-time-2.9.9.jarlibthrift-0.9.1.jarspark-cassandra-connector_2.10-1.5.1.jar |
在 /opt/spark/conf 目錄下，新建 spark-env.sh 文件，輸入下面內(nèi)容
|
1
|
SPARK_CLASSPATH=/opt/spark/managed-lib/* |
3.Spring Boot應(yīng)用開發(fā)
添加 spark-cassandra-connector 和 spark 依賴
|
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
|
<dependency> <groupId>com.datastax.spark</groupId> <artifactId>spark-cassandra-connector_2.10</artifactId> <version>1.5.1</version> </dependency> <dependency> <groupId>org.apache.spark</groupId> <artifactId>spark-core_2.10</artifactId> <version>1.5.1</version> </dependency> <dependency> <groupId>org.apache.spark</groupId> <artifactId>spark-sql_2.10</artifactId> <version>1.5.1</version> </dependency> |
在 application.yml 中配置 spark 與 cassandra 路徑
|
1
2
3
|
spark.master: spark://master:7077cassandra.host: 192.168.1.140cassandra.keyspace: hfcb |
此處特別說(shuō)明 spark://master:7077 是域名形式而不是ip地址，可修改本地 hosts 文件將 master 與 ip 地址映射。
配置 SparkContext 和 CassandraSQLContext
|
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
|
@Configurationpublic class SparkCassandraConfig { @Value("${spark.master}") String sparkMasterUrl; @Value("${cassandra.host}") String cassandraHost; @Value("${cassandra.keyspace}") String cassandraKeyspace; @Bean public JavaSparkContext javaSparkContext(){ SparkConf conf = new SparkConf(true) .set("spark.cassandra.connection.host", cassandraHost)// .set("spark.cassandra.auth.username", "cassandra")// .set("spark.cassandra.auth.password", "cassandra") .set("spark.submit.deployMode", "client"); JavaSparkContext context = new JavaSparkContext(sparkMasterUrl, "SparkDemo", conf); return context; } @Bean public CassandraSQLContext sqlContext(){ CassandraSQLContext cassandraSQLContext = new CassandraSQLContext(javaSparkContext().sc()); cassandraSQLContext.setKeyspace(cassandraKeyspace); return cassandraSQLContext; } } |
簡(jiǎn)單調(diào)用
|
1
2
3
4
5
6
7
8
9
|
@Repositorypublic class PersonRepository { @Autowired CassandraSQLContext cassandraSQLContext; public Long countPerson(){ DataFrame people = cassandraSQLContext.sql("select * from person order by id"); return people.count(); }} |
啟動(dòng)即可如常規(guī)Spring Boot程序一樣執(zhí)行。
源碼地址: https://github.com/wiselyman/spring-spark-cassandra.git
總結(jié)
以上所述是小編給大家介紹的Spring Boot與Spark、Cassandra系統(tǒng)集成開發(fā)示例，希望對(duì)大家有所幫助，如果大家有任何疑問(wèn)請(qǐng)給我留言，小編會(huì)及時(shí)回復(fù)大家的。在此也非常感謝大家對(duì)服務(wù)器之家網(wǎng)站的支持！
原文鏈接：http://www.wisely.top/2018/02/01/spring_boot-spark-cassandra-integration/
