这篇文章主要介绍“Oracle怎么迁移数据到Greenplum”,在日常操作中,相信很多人在Oracle怎么迁移数据到Greenplum问题上存在疑惑,小编查阅了各式资料,整理出简单好用的操作方法,希望对大家解答”Oracle怎么迁移数据到Greenplum”的疑惑有所帮助!接下来,请跟着小编一起来学习吧!
创新互联建站长期为上1000家客户提供的网站建设服务,团队从业经验10年,关注不同地域、不同群体,并针对不同对象提供差异化的产品和服务;打造开放共赢平台,与合作伙伴共同营造健康的互联网生态环境。为牙克石企业提供专业的网站设计、成都网站制作,牙克石网站改版等技术服务。拥有十多年丰富建站经验和众多成功案例,为您定制开发。
Oracle端表结构
SQL> select dbms_metadata.get_ddl('TABLE','TAB_ORA','ZWC') from dual;
CREATE TABLE "ZWC"."TAB_ORA"
( "ID" NUMBER,
"OWNER" VARCHAR2(30),
"NAME" VARCHAR2(128),
CONSTRAINT "PK_ID" PRIMARY KEY ("ID")
USING INDEX PCTFREE 10 INITRANS 2 MAXTRANS 255 COMPUTE STATISTICS
STORAGE(INITIAL 65536 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645
PCTINCREASE 0 FREELISTS 1 FREELIST GROUPS 1
BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT)
TABLESPACE "USERS" ENABLE
) SEGMENT CREATION IMMEDIATE
PCTFREE 10 PCTUSED 40 INITRANS 1 MAXTRANS 255
NOCOMPRESS LOGGING
STORAGE(INITIAL 65536 NEXT 1048576 MINEXTENTS 1 MAXEXTENTS 2147483645
PCTINCREASE 0 FREELISTS 1 FREELIST GROUPS 1
BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT)
TABLESPACE "USERS"
使用sqluldr工具导出Oracle表数据
[oracle@ggos ~]$ ./sqluldr zwc/zwc@oraprod query="select * from tab_ora" field=0x7c records=0x0d0x0a charset=gbk file=tab_ora.csv head=off
0 rows exported at 2013-12-04 22:08:14, size 0 MB.
87975 rows exported at 2013-12-04 22:08:14, size 3 MB.
output file tab_ora.csv closed at 87975 rows, size 3 MB.
[oracle@ggos ~]$ more tab_ora.csv
20|SYS|ICOL$
46|SYS|I_USER1
28|SYS|CON$
15|SYS|UNDO$
29|SYS|C_COBJ#
3|SYS|I_OBJ#
25|SYS|PROXY_ROLE_DATA$
41|SYS|I_IND1
54|SYS|I_CDEF2
40|SYS|I_OBJ5
26|SYS|I_PROXY_ROLE_DATA$_1
17|SYS|FILE$
13|SYS|UET$
9|SYS|I_FILE#_BLOCK#
43|SYS|I_FILE1
51|SYS|I_CON1
38|SYS|I_OBJ3
7|SYS|I_TS#
56|SYS|I_CDEF4
19|SYS|IND$
14|SYS|SEG$
6|SYS|C_TS#
44|SYS|I_FILE2
21|SYS|COL$
.......................................
GP端创建表
[gpadmin@mdw gpseg-1]$ psql -h 192.168.1.23 -d zwcdb -U zhongwc -W
Password for user zhongwc:
psql (8.2.15)
Type "help" for help.
zwcdb=# create table tab_gp(
zwcdb(# id integer primary key,
zwcdb(# owner varchar(200),
zwcdb(# name varchar(200)
zwcdb(# );
NOTICE: CREATE TABLE / PRIMARY KEY will create implicit index "tab_gp_pkey" for table "tab_gp"
CREATE TABLE
zwcdb=# \d+ tab_gp
Table "public.tab_gp"
Column | Type | Modifiers | Storage | Description
--------+------------------------+-----------+----------+-------------
id | integer | not null | plain |
owner | character varying(200) | | extended |
name | character varying(200) | | extended |
Indexes:
"tab_gp_pkey" PRIMARY KEY, btree (id)
Has OIDs: no
Distributed by: (id)
导入数据
zwcdb=# copy tab_gp from '/tmp/tab_ora.csv' delimiter '|';
COPY 87975
zwcdb=# select count(*) from tab_gp;
count
-------
87975
(1 row)
zwcdb=# select * from tab_gp;
id | owner | name
-------+--------------------+--------------------------------
15 | SYS | UNDO$
29 | SYS | C_COBJ#
3 | SYS | I_OBJ#
25 | SYS | PROXY_ROLE_DATA$
41 | SYS | I_IND1
17 | SYS | FILE$
13 | SYS | UET$
9 | SYS | I_FILE#_BLOCK#
43 | SYS | I_FILE1
51 | SYS | I_CON1
7 | SYS | I_TS#
19 | SYS | IND$
21 | SYS | COL$
45 | SYS | I_TS1
35 | SYS | I_UNDO2
5 | SYS | CLU$
23 | SYS | PROXY_DATA$
47 | SYS | I_USER2
49 | SYS | I_COL2
37 | SYS | I_OBJ2
39 | SYS | I_OBJ4
57 | SYS | I_CCOL1
59 | SYS | BOOTSTRAP$
33 | SYS | I_TAB1
31 | SYS | CDEF$
53 | SYS | I_CDEF1
55 | SYS | I_CDEF3
11 | SYS | I_USER#
27 | SYS | I_PROXY_ROLE_DATA$_2
61 | SYS | OBJAUTH$
63 | SYS | I_OBJAUTH2
65 | SYS | I_UGROUP1
67 | SYS | TSQ$
69 | SYS | VIEW$
71 | SYS | SUPEROBJ$
73 | SYS | I_SUPEROBJ2
75 | SYS | I_VIEW1
77 | SYS | I_SYN1
79 | SYS | I_SEQ1
81 | SYS | I_LOB1
83 | SYS | COLTYPE$
85 | SYS | I_COLTYPE2
zwcdb=# select gp_segment_id,count(*) from tab_gp group by gp_segment_id;
gp_segment_id | count
---------------+-------
1 | 43982
0 | 43993
(2 rows)
到此,关于“Oracle怎么迁移数据到Greenplum”的学习就结束了,希望能够解决大家的疑惑。理论与实践的搭配能更好的帮助大家学习,快去试试吧!若想继续学习更多相关知识,请继续关注创新互联网站,小编会继续努力为大家带来更多实用的文章!
网站标题:Oracle怎么迁移数据到Greenplum
网站网址:http://lswzjz.com/article/jccssc.html