一、疑問，Oracle資料泵匯入的時候建立索引是否會使用並行？

某客戶需要使用資料泵進行遷移，客戶諮詢匯入的時間能不能加快一點。

那麼如何加快匯入的速度呢？多加一些並行，那麼建立索引內部的索引並行度是否會調整呢？

帶著這些疑問看看Oracle資料泵並行引數與匯入後建立索引的並行度是否有關係！

二、實驗測試

2.1測試資料準備

Oracel11.2.0.4

--分割槽表建立

create user yz identified by yz;

grant dba to yz;

conn yz/yz

create table a1(id number,

    deal_date date, area_code number, contents varchar2(4000))

 partition by range(deal_date)

 (

 partition p1 values less than(to_date('2019-02-01','yyyy-mm-dd')),

 partition p2 values less than(to_date('2019-03-01','yyyy-mm-dd')),

 partition p3 values less than(to_date('2019-04-01','yyyy-mm-dd')),

 partition p4 values less than(to_date('2019-05-01','yyyy-mm-dd')),

 partition p5 values less than(to_date('2019-06-01','yyyy-mm-dd')),

 partition p6 values less than(to_date('2019-07-01','yyyy-mm-dd')),

 partition p7 values less than(to_date('2019-08-01','yyyy-mm-dd')),

 partition p8 values less than(to_date('2019-09-01','yyyy-mm-dd')),

 partition p9 values less than(to_date('2019-10-01','yyyy-mm-dd')),

 partition p10 values less than(to_date('2019-11-01','yyyy-mm-dd')),

 partition p11 values less than(to_date('2019-12-01','yyyy-mm-dd')),

 partition p12 values less than(to_date('2020-01-01','yyyy-mm-dd')),

 partition p13 values less than(to_date('2020-02-01','yyyy-mm-dd')),

 partition p14 values less than(to_date('2020-03-01','yyyy-mm-dd')),

 partition p15 values less than(to_date('2020-04-01','yyyy-mm-dd')),

 partition p16 values less than(to_date('2020-05-01','yyyy-mm-dd')),

 partition p17 values less than(to_date('2020-06-01','yyyy-mm-dd')),

 partition p18 values less than(to_date('2020-07-01','yyyy-mm-dd')),

 partition p19 values less than(to_date('2020-08-01','yyyy-mm-dd')),

 partition p20 values less than(to_date('2020-09-01','yyyy-mm-dd')),

 partition p31 values less than(to_date('2020-10-01','yyyy-mm-dd')),

 partition p32 values less than(to_date('2020-11-01','yyyy-mm-dd')),

 partition p33 values less than(to_date('2020-12-01','yyyy-mm-dd')),

 partition p34 values less than(to_date('2021-01-01','yyyy-mm-dd')),

 partition p35 values less than(to_date('2021-02-01','yyyy-mm-dd')),

 partition p36 values less than(to_date('2021-03-01','yyyy-mm-dd')),

 partition p37 values less than(to_date('2021-04-01','yyyy-mm-dd')),

 partition p38 values less than(to_date('2021-05-01','yyyy-mm-dd')),

 partition p39 values less than(to_date('2021-06-01','yyyy-mm-dd')),

 partition p40 values less than(to_date('2021-07-01','yyyy-mm-dd'))

 );

insert into a1 (id,deal_date,area_code,contents)

 select rownum,

          to_date(to_char(sysdate-900,'J')+ trunc(dbms_random.value(0,200)),'J'),

ceil(dbms_random.value(590,599)),

 rpad('*',400,'*')

 from dual

 connect by rownum <= 100000;

commit;

create table a2(id number,

    deal_date date, area_code number, contents varchar2(4000))

 partition by range(deal_date)

 (

 partition p1 values less than(to_date('2019-02-01','yyyy-mm-dd')),

 partition p2 values less than(to_date('2019-03-01','yyyy-mm-dd')),

 partition p3 values less than(to_date('2019-04-01','yyyy-mm-dd')),

 partition p4 values less than(to_date('2019-05-01','yyyy-mm-dd')),

 partition p5 values less than(to_date('2019-06-01','yyyy-mm-dd')),

 partition p6 values less than(to_date('2019-07-01','yyyy-mm-dd')),

 partition p7 values less than(to_date('2019-08-01','yyyy-mm-dd')),

 partition p8 values less than(to_date('2019-09-01','yyyy-mm-dd')),

 partition p9 values less than(to_date('2019-10-01','yyyy-mm-dd')),

 partition p10 values less than(to_date('2019-11-01','yyyy-mm-dd')),

 partition p11 values less than(to_date('2019-12-01','yyyy-mm-dd')),

 partition p12 values less than(to_date('2020-01-01','yyyy-mm-dd')),

 partition p13 values less than(to_date('2020-02-01','yyyy-mm-dd')),

 partition p14 values less than(to_date('2020-03-01','yyyy-mm-dd')),

 partition p15 values less than(to_date('2020-04-01','yyyy-mm-dd')),

 partition p16 values less than(to_date('2020-05-01','yyyy-mm-dd')),

 partition p17 values less than(to_date('2020-06-01','yyyy-mm-dd')),

 partition p18 values less than(to_date('2020-07-01','yyyy-mm-dd')),

 partition p19 values less than(to_date('2020-08-01','yyyy-mm-dd')),

 partition p20 values less than(to_date('2020-09-01','yyyy-mm-dd')),

 partition p31 values less than(to_date('2020-10-01','yyyy-mm-dd')),

 partition p32 values less than(to_date('2020-11-01','yyyy-mm-dd')),

 partition p33 values less than(to_date('2020-12-01','yyyy-mm-dd')),

 partition p34 values less than(to_date('2021-01-01','yyyy-mm-dd')),

 partition p35 values less than(to_date('2021-02-01','yyyy-mm-dd')),

 partition p36 values less than(to_date('2021-03-01','yyyy-mm-dd')),

 partition p37 values less than(to_date('2021-04-01','yyyy-mm-dd')),

 partition p38 values less than(to_date('2021-05-01','yyyy-mm-dd')),

 partition p39 values less than(to_date('2021-06-01','yyyy-mm-dd')),

 partition p40 values less than(to_date('2021-07-01','yyyy-mm-dd'))

 );

insert into a2 (id,deal_date,area_code,contents)

 select rownum,

          to_date(to_char(sysdate-900,'J')+ trunc(dbms_random.value(0,200)),'J'),

ceil(dbms_random.value(590,599)),

 rpad('*',400,'*')

 from dual

 connect by rownum <= 200000;

commit;

alter table a1 add constraint pk_id primary key (id);

alter table a2 add constraint pk_id_time primary key(id,deal_date);

SQL> create index cc_id on a1(id);

create index cc_id on a1(id)

                         *

ERROR at line 1:

ORA-01408: such column list already indexed

SQL> select index_name,status from user_indexes where table_name in('A1','A2');

INDEX_NAME                     STATUS

------------------------------ --------

PK_ID_TIME                     VALID

PK_ID                          VALID    

Alter table a1 drop constraint pk_id;

Alter table a2 drop constraint pk_id_time;

create index cc_id on a1(id) LOCAL;

alter table a1 add constraint pk_id primary key (id) USING INDEX cc_id ;

ORA-14196: Specified index cannot be used to enforce the constraint.

DROP INDEX CC_ID;

create index cc_id on a1(id) ;

alter table a1 add constraint pk_id primary key (id) USING INDEX cc_id ;

create index cc_id_DATE on a2(id,DEAL_DATE) LOCAL;

alter table a2 add constraint pk_id_DATE primary key (id,DEAL_DATE) USING INDEX cc_id_DATE ;

https://www.cnblogs.com/lvcha001/p/10218318.html

索引可以認為分3種，非分割槽索引，全域性XX索引，可以是全域性範圍分割槽、全域性雜湊分割槽，這種情況會根據規則將資料打散，而不是根據實際表的資料進行打散！
  本地索引，完全根據分割槽表的子分割槽，一個子分割槽一個索引！但是我們觀察重建索引其實是一樣的操作，本次測試為了後續重建索引，建立不同的分割槽型別，非分割槽索引使用主鍵！

create index ind_hash on a1(id,0) global partition by hash (id) partitions 8 online;

SQL>  select index_name,status from user_indexes where table_name in('A1','A2');

INDEX_NAME                     STATUS

------------------------------ --------

CC_ID                          VALID

CC_ID_DATE                     N/A

IND_HASH                       N/A

select index_name,PARTITION_NAME,HIGH_VALUE,STATUS,TABLESPACE_NAME from dba_ind_partitions
 where index_owner='YZ' and index_name IN('CC_ID_DATE','IND_HASH');

2.2 匯入SQL檔案測試

nohup time expdp \'/ as sysdba\' directory=dump dumpfile=yang%u.dmp logfile=yang.log tables=yz.a1,yz.a2 FLASHBACK_SCN=1017463 parallel=2 &

情況一、匯出並行度2，匯入並行度2，觀察SQL指令碼

nohup time impdp \'/ as sysdba\' directory=dump dumpfile=yang%u.dmp logfile=yang.log tables=yz.a1,yz.a2 parallel=2 sqlfile=table01.sql &

Processing object type TABLE_EXPORT/TABLE/PROCACT_INSTANCE

Processing object type TABLE_EXPORT/TABLE/TABLE

Processing object type TABLE_EXPORT/TABLE/INDEX/INDEX

Processing object type TABLE_EXPORT/TABLE/INDEX/FUNCTIONAL_INDEX/INDEX

Processing object type TABLE_EXPORT/TABLE/CONSTRAINT/CONSTRAINT

Processing object type TABLE_EXPORT/TABLE/INDEX/STATISTICS/INDEX_STATISTICS

Processing object type TABLE_EXPORT/TABLE/INDEX/STATISTICS/FUNCTIONAL_INDEX/INDEX_STATISTICS

Job "SYS"."SYS_SQL_FILE_TABLE_01" successfully completed at Wed Aug 11 07:00:04 2021 elapsed 0 00:00:03

-- new object type path: TABLE_EXPORT/TABLE/TABLE

-- CONNECT SYS

CREATE TABLE "YZ"."A2"

   (    "ID" NUMBER,

······

-- new object type path: TABLE_EXPORT/TABLE/INDEX/INDEX

-- CONNECT YZ

CREATE INDEX "YZ"."CC_ID_DATE" ON "YZ"."A2" ("ID", "DEAL_DATE")

  PCTFREE 10 INITRANS 2 MAXTRANS 255

  STORAGE(

  BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT) LOCAL

 (PARTITION "P1"

  PCTFREE 10 INITRANS 2 MAXTRANS 255 LOGGING

  STORAGE(

  BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT)

  TABLESPACE "USERS" ,

 PARTITION "P2"

······

 PARTITION "P40"

  PCTFREE 10 INITRANS 2 MAXTRANS 255 LOGGING

  STORAGE(

  BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT)

  TABLESPACE "USERS" ) PARALLEL 1 ;

  ALTER INDEX "YZ"."CC_ID_DATE" NOPARALLEL;

CREATE INDEX "YZ"."CC_ID" ON "YZ"."A1" ("ID")

  PCTFREE 10 INITRANS 2 MAXTRANS 255

  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" PARALLEL 1 ;

  ALTER INDEX "YZ"."CC_ID" NOPARALLEL;

-- new object type path: TABLE_EXPORT/TABLE/INDEX/FUNCTIONAL_INDEX/INDEX

CREATE INDEX "YZ"."IND_HASH" ON "YZ"."A1" ("ID", 0)

  PCTFREE 10 INITRANS 2 MAXTRANS 255

  STORAGE(

  BUFFER_POOL DEFAULT FLASH_CACHE DEFAULT CELL_FLASH_CACHE DEFAULT)

  TABLESPACE "USERS"  GLOBAL PARTITION BY HASH ("ID")

 (PARTITION "SYS_P41"

  TABLESPACE "USERS" ,

 PARTITION "SYS_P42"

  TABLESPACE "USERS" ,

 PARTITION "SYS_P43"

  TABLESPACE "USERS" ,

 PARTITION "SYS_P44"

  TABLESPACE "USERS" ,

 PARTITION "SYS_P45"

  TABLESPACE "USERS" ,

 PARTITION "SYS_P46"

  TABLESPACE "USERS" ,

 PARTITION "SYS_P47"

  TABLESPACE "USERS" ,

 PARTITION "SYS_P48"

  TABLESPACE "USERS" ) PARALLEL 1 ;

  ALTER INDEX "YZ"."IND_HASH" NOPARALLEL;

-- new object type path: TABLE_EXPORT/TABLE/CONSTRAINT/CONSTRAINT

-- CONNECT SYS

ALTER TABLE "YZ"."A1" ADD CONSTRAINT "PK_ID" PRIMARY KEY ("ID")

  USING INDEX "YZ"."CC_ID"  ENABLE;

ALTER TABLE "YZ"."A1" ADD SUPPLEMENTAL LOG GROUP "GGS_87350" ("ID") ALWAYS;

ALTER TABLE "YZ"."A1" ADD SUPPLEMENTAL LOG DATA (PRIMARY KEY) COLUMNS;

ALTER TABLE "YZ"."A1" ADD SUPPLEMENTAL LOG DATA (UNIQUE INDEX) COLUMNS;

ALTER TABLE "YZ"."A1" ADD SUPPLEMENTAL LOG DATA (FOREIGN KEY) COLUMNS;

ALTER TABLE "YZ"."A2" ADD CONSTRAINT "PK_ID_DATE" PRIMARY KEY ("ID", "DEAL_DATE")

  USING INDEX "YZ"."CC_ID_DATE"  ENABLE;

ALTER TABLE "YZ"."A2" ADD SUPPLEMENTAL LOG GROUP "GGS_87381" ("ID", "DEAL_DATE") ALWAYS;

ALTER TABLE "YZ"."A2" ADD SUPPLEMENTAL LOG DATA (PRIMARY KEY) COLUMNS;

ALTER TABLE "YZ"."A2" ADD SUPPLEMENTAL LOG DATA (UNIQUE INDEX) COLUMNS;

ALTER TABLE "YZ"."A2" ADD SUPPLEMENTAL LOG DATA (FOREIGN KEY) COLUMNS;

-- new object type path: TABLE_EXPORT/TABLE/INDEX/STATISTICS/INDEX_STATISTICS

······

並行度1！

情況二、匯出並行度2，匯入並行度4，觀察SQL指令碼

$ nohup time impdp \'/ as sysdba\' directory=dump dumpfile=yang%u.dmp logfile=yang.log tables=yz.a1,yz.a2 parallel=4 sqlfile=table02.sql &

$ cat dump/table02.sql |grep PARALLEL

TABLESPACE "USERS" ) PARALLEL 1 ;

ALTER INDEX "YZ"."CC_ID_DATE" NOPARALLEL;

TABLESPACE "USERS" PARALLEL 1 ;

ALTER INDEX "YZ"."CC_ID" NOPARALLEL;

TABLESPACE "USERS" ) PARALLEL 1 ;

ALTER INDEX "YZ"."IND_HASH" NOPARALLEL;

通過測試，我們可以得知資料泵匯入建立索引並行度就是1！！！這種情況除非資料庫引數設定物件是AUTO選擇並行，如果是Manual的情況則無法使用並行加快速度！

2.3 如何對資料泵匯入使用並行建立索引加快速度！

參考

https://blog.51cto.com/wyzwl/2333565

？為什麼指令碼要排除約束？    感興趣的小夥伴可以測試一下。
目標端建立使用者、授權後，匯入表資料！

*****************************************************資料匯入**************************************

cat >imp_data.par <<EOF
userid='/ as sysdba'
directory=dump
dumpfile=yang%u.dmp
logfile=imp_data.log
cluster=no
parallel=2
exclude= index,constraint
EOF
--排除索引和約束，執行匯入
nohup impdp parfile=imp_data.par > imp_data.out &

*****************************************************索引及約束匯入**************************************

--通過sqlfile引數生成建立索引語句
cat >imp_ind_con.par <<EOF
userid='/ as sysdba'
directory=dump
dumpfile=yang%u.dmp
sqlfile=imp_ind_con.sql
logfile=imp_ind_con.log
cluster=no
parallel=2
tables=yz.a1,yz.a2
include=index,constraint
EOF

--執行生成建立索引語句（實際並不會匯入）
nohup impdp parfile= imp_ind_con.par > imp_ind_con.out &
--修改建立索引的並行度，並行度建議不超過CPU核數的1.5倍
--LINUX環境使用
sed -i 's/PARALLEL 1/PARALLEL 16/g' imp_ind_con.sql
--因AIX環境sed沒有-i引數，可以使用如下兩種方法：
perl -pi -e 's/ PARALLEL 1/PARALLEL 16/g' imp_ind_con.sql
或者
vi imp_ind_con.sql << EOF
:%s/ PARALLEL 1/PARALLEL 16/g
:wq
EOF

*****************************************************替換效果***************************************

[oracle@t2 dump]$ cat imp_ind_con.sql|grep PARALLEL
TABLESPACE "USERS" ) PARALLEL 1 ;
ALTER INDEX "YZ"."CC_ID_DATE" NOPARALLEL;
TABLESPACE "USERS" PARALLEL 1 ;
ALTER INDEX "YZ"."CC_ID" NOPARALLEL;
TABLESPACE "USERS" ) PARALLEL 1 ;
ALTER INDEX "YZ"."IND_HASH" NOPARALLEL;
[oracle@t2 dump]$ sed -i 's/PARALLEL 1/PARALLEL 16/g' imp_ind_con.sql
[oracle@t2 dump]$ cat imp_ind_con.sql|grep PARALLEL
TABLESPACE "USERS" ) PARALLEL 16 ;
ALTER INDEX "YZ"."CC_ID_DATE" NOPARALLEL;
TABLESPACE "USERS" PARALLEL 16 ;
ALTER INDEX "YZ"."CC_ID" NOPARALLEL;
TABLESPACE "USERS" ) PARALLEL 16 ;
ALTER INDEX "YZ"."IND_HASH" NOPARALLEL;

***************************************************************************************************

$more 觀察SQL指令碼

-- new object type path: TABLE_EXPORT/TABLE/INDEX/INDEX
-- CONNECT YZ
CREATE INDEX "YZ"."CC_ID_DATE" ON "YZ"."A2" ("ID", "DEAL_DATE")
PCTFREE 10 INITRANS 2 MAXTRANS 255
······
-- new object type path: TABLE_EXPORT/TABLE/CONSTRAINT/CONSTRAINT
-- CONNECT SYS
ALTER TABLE "YZ"."A1" ADD CONSTRAINT "PK_ID" PRIMARY KEY ("ID")
USING INDEX "YZ"."CC_ID" ENABLE;

先建立索引在整約束！

--等匯入完資料之後，執行建索引的SQL：
$vi imp_ind_con.sh
sqlplus / as sysdba <<EOF
set timing on
set echo on
set verify on
spool imp_ind_con.log
@imp_ind_con.sql
spool off
exit
EOF
--執行建索引的SQL
nohup sh imp_ind_con.sh> imp_ind_con.out &

疑問一、匯出表的dump有建立使用者的語句嗎？如何匯出建立使用者的SQL語句

--從只匯出表的dump，匯入create user 提示Not found user

$ nohup time impdp \'/ as sysdba\' directory=dump dumpfile=yang%u.dmp logfile=yang.log include=user parallel=1 sqlfile=user01.sql &

ORA-39168: Object path USER was not found.

--從匯出整個schema dump再次測試！

$nohup time expdp \'/ as sysdba\' directory=dump dumpfile=yanga%u.dmp logfile=yang.log SCHEMAS=yz parallel=2

$ scp /home/oracle/script/dump/yanga*.dmp t2:/home/oracle/script/dump/.

$ nohup time impdp \'/ as sysdba\' directory=dump dumpfile=yanga%u.dmp logfile=yang.log include=user parallel=1 sqlfile=user02.sql &

$ cat user02.sql

-- CONNECT SYSTEM

 CREATE USER "YZ" IDENTIFIED BY VALUES 'S:C9A5297B9802EBB85A3BE800929ECE1BFCCB00146E58E0FBB055A937869F;86EF13A1088170F5'

      DEFAULT TABLESPACE "USERS"

      TEMPORARY TABLESPACE "TEMP";

疑問二、資料泵匯出使用者後，匯入這個使用者，這個使用者之前在DB不存在，Oracle會自動建立這個使用者嗎？

Starting "SYS"."SYS_EXPORT_SCHEMA_01": "/******** AS SYSDBA" directory=dump dumpfile=yanga%u.dmp logfile=yang.log SCHEMAS=yz parallel=2

nohup time impdp \'/ as sysdba\' directory=dump dumpfile=yanga%u.dmp logfile=yanga.log parallel=4 &

SQL> select username,account_status from dba_users where username='YZ';

no rows selected

SQL>
SQL> r
1* select username,account_status from dba_users where username='YZ'

USERNAME ACCOUNT_STATUS
------------------------------ --------------------------------
YZ OPEN

是可以自動建立使用者的！

Processing object type SCHEMA_EXPORT/USER
Processing object type SCHEMA_EXPORT/SYSTEM_GRANT
Processing object type SCHEMA_EXPORT/ROLE_GRANT
Processing object type SCHEMA_EXPORT/DEFAULT_ROLE
Processing object type SCHEMA_EXPORT/PRE_SCHEMA/PROCACT_SCHEMA
Processing object type SCHEMA_EXPORT/TABLE/PROCACT_INSTANCE
Processing object type SCHEMA_EXPORT/TABLE/TABLE
Processing object type SCHEMA_EXPORT/TABLE/TABLE_DATA
. . imported "YZ"."A2":"P3" 12.53 MB 30997 rows

. . imported "YZ"."A2":"P40" 0 KB 0 rows
Processing object type SCHEMA_EXPORT/TABLE/INDEX/INDEX
Processing object type SCHEMA_EXPORT/TABLE/INDEX/FUNCTIONAL_INDEX/INDEX
Processing object type SCHEMA_EXPORT/TABLE/CONSTRAINT/CONSTRAINT
Processing object type SCHEMA_EXPORT/TABLE/INDEX/STATISTICS/INDEX_STATISTICS
Processing object type SCHEMA_EXPORT/TABLE/INDEX/STATISTICS/FUNCTIONAL_INDEX/INDEX_STATISTICS
Job "SYS"."SYS_IMPORT_FULL_01" successfully completed at Wed Aug 11 12:47:16 2021 elapsed 0 00:00:17

Oracle資料泵匯入的時候建立索引是否會使用並行？