Skip to content

Commit bbb4e9c

Browse files
committed
update kudu-side文档
1 parent ccd5ede commit bbb4e9c

File tree

2 files changed

+98
-57
lines changed

2 files changed

+98
-57
lines changed

docs/plugin/kuduSide.md

Lines changed: 96 additions & 55 deletions
Original file line numberDiff line numberDiff line change
@@ -1,66 +1,48 @@
1-
2-
## 1.格式:
3-
All:
4-
```
5-
create table sideTable(
6-
id int,
7-
tablename1 VARCHAR,
8-
PRIMARY KEY(id),
9-
PERIOD FOR SYSTEM_TIME
10-
)WITH(
11-
type='kudu',
12-
kuduMasters ='ip1,ip2,ip3',
13-
tableName ='impala::default.testSide',
14-
cache ='ALL',
15-
primaryKey='id,xx',
16-
lowerBoundPrimaryKey='10,xx',
17-
upperBoundPrimaryKey='15,xx',
18-
workerCount='1',
19-
defaultOperationTimeoutMs='600000',
20-
defaultSocketReadTimeoutMs='6000000',
21-
batchSizeBytes='100000000',
22-
limitNum='1000',
23-
isFaultTolerant='false',
24-
partitionedJoin='false'
25-
);
26-
```
27-
LRU:
1+
## 1.格式
2+
通过建表语句中的` PERIOD FOR SYSTEM_TIME`将表标识为维表,其中`PRIMARY KEY(keyInfo)`中的keyInfo,表示用来和源表进行关联的字段,
3+
维表JOIN的条件必须与`keyInfo`字段一致。
284
```
29-
create table sideTable(
30-
id int,
31-
tablename1 VARCHAR,
32-
PRIMARY KEY(id),
5+
CREATE TABLE tableName(
6+
colName cloType,
7+
...
8+
PRIMARY KEY(colName1,colName2) ,
339
PERIOD FOR SYSTEM_TIME
34-
)WITH(
35-
type='kudu',
10+
)WITH(
11+
type ='kudu',
3612
kuduMasters ='ip1,ip2,ip3',
3713
tableName ='impala::default.testSide',
38-
cache ='LRU',
14+
primaryKey='id,xx',
15+
lowerBoundPrimaryKey='10,xx',
16+
upperBoundPrimaryKey='15,xx',
3917
workerCount='1',
4018
defaultOperationTimeoutMs='600000',
4119
defaultSocketReadTimeoutMs='6000000',
4220
batchSizeBytes='100000000',
4321
limitNum='1000',
4422
isFaultTolerant='false',
23+
cache ='LRU',
24+
cacheSize ='10000',
25+
cacheTTLMs ='60000',
26+
parallelism ='1',
4527
partitionedJoin='false'
46-
);
47-
```
48-
28+
);
29+
```
4930
## 2.支持版本
50-
kudu 1.10.0+cdh6.2.0
31+
kudu 1.10.0+cdh6.2.0
5132

5233
## 3.表结构定义
5334

5435
|参数名称|含义|
5536
|----|---|
5637
| tableName | 注册到flink的表名称(可选填;不填默认和hbase对应的表名称相同)|
5738
| colName | 列名称|
58-
| colType | 列类型 [colType支持的类型](docs/colType.md)|
39+
| colType | 列类型 [colType支持的类型](colType.md)|
5940
| PERIOD FOR SYSTEM_TIME | 关键字表明该定义的表为维表信息|
6041
| PRIMARY KEY(keyInfo) | 维表主键定义;多个列之间用逗号隔开|
6142

62-
## 3.参数
43+
## 4.参数
6344

45+
参数详细说明请看[参数详细说明](./sideParams.md)
6446

6547
|参数名称|含义|是否必填|默认值|
6648
|----|---|---|-----|
@@ -78,17 +60,25 @@ kudu 1.10.0+cdh6.2.0
7860
| isFaultTolerant |查询是否容错 查询失败是否扫描第二个副本 默认false 容错 |||
7961
| cache | 维表缓存策略(NONE/LRU/ALL)||NONE|
8062
| partitionedJoin | 是否在維表join之前先根据 設定的key 做一次keyby操作(可以減少维表的数据缓存量)||false|
81-
82-
8363
--------------
84-
> 缓存策略
85-
* NONE: 不做内存缓存
86-
* LRU:
87-
* cacheSize: 缓存的条目数量
88-
* cacheTTLMs:缓存的过期时间(ms)
8964

90-
## 4.样例
91-
All:
65+
## 5.样例
66+
### LRU维表示例
67+
```
68+
create table sideTable(
69+
id int,
70+
tablename1 VARCHAR,
71+
PRIMARY KEY(id),
72+
PERIOD FOR SYSTEM_TIME
73+
)WITH(
74+
type='kudu',
75+
kuduMasters ='ip1,ip2,ip3',
76+
tableName ='impala::default.testSide',
77+
cache ='LRU',
78+
partitionedJoin='false'
79+
);
80+
```
81+
### ALL维表示例
9282
```
9383
create table sideTable(
9484
id int,
@@ -106,19 +96,70 @@ create table sideTable(
10696
partitionedJoin='false'
10797
);
10898
```
109-
LRU:
99+
100+
## 6.kudu异步关联完整样例
101+
110102
```
111-
create table sideTable(
103+
CREATE TABLE MyTable(
104+
id bigint,
105+
name varchar,
106+
address varchar
107+
)WITH(
108+
type = 'kafka10',
109+
bootstrapServers = '172.16.101.224:9092',
110+
zookeeperQuorm = '172.16.100.188:2181/kafka',
111+
offsetReset = 'latest',
112+
topic = 'tiezhu_test_in2',
113+
timezone = 'Asia/Shanghai',
114+
topicIsPattern = 'false',
115+
parallelism = '1'
116+
);
117+
118+
CREATE TABLE sideTable(
112119
id int,
113-
tablename1 VARCHAR,
120+
message varchar,
114121
PRIMARY KEY(id),
115122
PERIOD FOR SYSTEM_TIME
116-
)WITH(
123+
)WITH(
117124
type='kudu',
118125
kuduMasters ='ip1,ip2,ip3',
119126
tableName ='impala::default.testSide',
120127
cache ='LRU',
121128
partitionedJoin='false'
122-
);
123-
```
129+
);
130+
131+
CREATE TABLE MyResult(
132+
id bigint,
133+
name varchar,
134+
address varchar,
135+
message varchar
136+
)WITH(
137+
type ='console',
138+
address ='192.168.80.106:9042,192.168.80.107:9042',
139+
userName='cassandra',
140+
password='cassandra',
141+
database ='tiezhu',
142+
tableName ='stu_out',
143+
parallelism ='1'
144+
);
145+
146+
insert
147+
into
148+
MyResult
149+
select
150+
t1.id AS id,
151+
t1.name AS name,
152+
t1.address AS address,
153+
t2.message AS message
154+
from(
155+
select
156+
id,
157+
name,
158+
address
159+
from
160+
MyTable
161+
) t1
162+
join sideTable t2
163+
on t1.id = t2.id;
164+
```
124165

pom.xml

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -25,10 +25,10 @@
2525
<!-- <module>sqlserver</module>-->
2626
<!-- <module>oracle</module>-->
2727
<!-- <module>cassandra</module>-->
28-
<!-- <module>kudu</module>-->
28+
<module>kudu</module>
2929
<!-- <module>postgresql</module>-->
3030
<!-- <module>serversocket</module>-->
31-
<!-- <module>console</module>-->
31+
<module>console</module>
3232
<!-- <module>clickhouse</module>-->
3333
<!-- <module>impala</module>-->
3434
<!-- <module>db2</module>-->

0 commit comments

Comments
 (0)