shrink怎样回收分区表碎片，针对这个问题，这篇文章详细介绍了相对应的分析和解答，希望可以帮助更多想解决这个问题的小伙伴找到更简单易行的方法。
实验如下：
CREATE TABLE "SCOTT"."T4"
( "A" NUMBER,
"B" NUMBER
)
PARTITION BY RANGE ("A")
(PARTITION "PART1" VALUES LESS THAN (10),
PARTITION "PART2" VALUES LESS THAN (20) ) ;
begin
for v1 in 1..19
loop
insert into scott.t4 values(v1,dbms_random.value(1,100000));
commit;
end loop;
end;
/
INSERT INTO scott.T4 SELECT * FROM sT4;
SQL> CREATE TABLE "SCOTT"."T4"
( "A" NUMBER,
"B" NUMBER
2 3 4 )
5 PARTITION BY RANGE ("A")
6 (PARTITION "PART1" VALUES LESS THAN (10),
7 PARTITION "PART2" VALUES LESS THAN (20) ) ;
Table created.
SQL> begin
2 for v1 in 1..19
3 loop
4 insert into scott.t4 values(v1,dbms_random.value(1,100000));
5 commit;
6 end loop;
7 end;
8 /
PL/SQL procedure successfully completed.
SQL>
SQL> conn scott/tiger;
Connected.
SQL> INSERT INTO T4 SELECT * FROM T4;
19 rows created.
省略.........
SQL> INSERT INTO T4 SELECT * FROM T4;
77824 rows created.
SQL> INSERT INTO T4 SELECT * FROM T4;
155648 rows created.
SQL> INSERT INTO T4 SELECT * FROM T4;
311296 rows created.
SQL> commit;
Commit complete.
SQL> commit;
Commit complete.
SQL> select count(*) from t4;
COUNT(*)
----------
622592
SQL>
--sys用户执行查询表大小及碎片
SQL> col SEGMENT_NAME for a15
SQL> select OWNER,SEGMENT_NAME,PARTITION_NAME,SEGMENT_TYPE,BYTES/1024/1024 size_m from dba_segments where SEGMENT_NAME='T4';
OWNER SEGMENT_NAME PARTITION_NAME SEGMENT_TYPE SIZE_M
------------------------------ --------------- ------------------------------ ------------------ ----------
SCOTT T4 PART1 TABLE PARTITION 16
SCOTT T4 PART2 TABLE PARTITION 16
SQL>
SQL> set lines 200
SQL> select table_name,
2 num_rows,
3 avg_row_len * num_rows / 1024 / 1024 actual_mb,
4 blocks * 8 / 1024 high_water_mb,
5 (blocks * 8 / 1024 - avg_row_len * num_rows / 1024 / 1024) need_recovery_mb
6 from dba_tables
7 where table_name = '&table_name' and owner='&owner';
Enter value for table_name: T4
Enter value for owner: SCOTT
old 7: where table_name = '&table_name' and owner='&owner'
new 7: where table_name = 'T4' and owner='SCOTT'
TABLE_NAME NUM_ROWS ACTUAL_MB HIGH_WATER_MB NEED_RECOVERY_MB
------------------------------ ---------- ---------- ------------- ----------------
T4
SQL>
--分析表：
SQL> analyze table scott.T4 compute statistics;
Table analyzed.
SQL> select OWNER,SEGMENT_NAME,PARTITION_NAME,SEGMENT_TYPE,BYTES/1024/1024 size_m from dba_segments where SEGMENT_NAME='T4';
OWNER SEGMENT_NAME PARTITION_NAME SEGMENT_TYPE SIZE_M
------------------------------ --------------- ------------------------------ ------------------ ----------
SCOTT T4 PART1 TABLE PARTITION 16
SCOTT T4 PART2 TABLE PARTITION 16
SQL> select table_name,
2 num_rows,
3 avg_row_len * num_rows / 1024 / 1024 actual_mb,
4 blocks * 8 / 1024 high_water_mb,
5 (blocks * 8 / 1024 - avg_row_len * num_rows / 1024 / 1024) need_recovery_mb
6 from dba_tables where table_name='T4';
TABLE_NAME NUM_ROWS ACTUAL_MB HIGH_WATER_MB NEED_RECOVERY_MB
------------------------------ ---------- ---------- ------------- ----------------
T4 622592 16.625 31.46875 14.84375
SQL>
--删除表数据：
SQL> delete scott.t4 where rownum < 600000;
599999 rows deleted.
SQL>
--再查看表发小及碎片情况：
SQL> select OWNER,SEGMENT_NAME,PARTITION_NAME,SEGMENT_TYPE,BYTES/1024/1024 size_m from dba_segments where SEGMENT_NAME='T4';
OWNER SEGMENT_NAME PARTITION_NAME SEGMENT_TYPE SIZE_M
------------------------------ --------------- ------------------------------ ------------------ ----------
SCOTT T4 PART1 TABLE PARTITION 16
SCOTT T4 PART2 TABLE PARTITION 16
SQL>
SQL>
SQL> select table_name,
2 num_rows,
3 avg_row_len * num_rows / 1024 / 1024 actual_mb,
4 blocks * 8 / 1024 high_water_mb,
5 (blocks * 8 / 1024 - avg_row_len * num_rows / 1024 / 1024) need_recovery_mb
6 from dba_tables where table_name='T4';
TABLE_NAME NUM_ROWS ACTUAL_MB HIGH_WATER_MB NEED_RECOVERY_MB
------------------------------ ---------- ---------- ------------- ----------------
T4 622592 16.625 31.46875 14.84375
SQL>
--再次分析一下表
SQL> analyze table scott.T4 compute statistics;
Table analyzed.
SQL>
SQL> select OWNER,SEGMENT_NAME,PARTITION_NAME,SEGMENT_TYPE,BYTES/1024/1024 size_m from dba_segments where SEGMENT_NAME='T4';
OWNER SEGMENT_NAME PARTITION_NAME SEGMENT_TYPE SIZE_M
------------------------------ --------------- ------------------------------ ------------------ ----------
SCOTT T4 PART1 TABLE PARTITION 16
SCOTT T4 PART2 TABLE PARTITION 16
SQL> select table_name,
2 num_rows,
3 avg_row_len * num_rows / 1024 / 1024 actual_mb,
4 blocks * 8 / 1024 high_water_mb,
5 (blocks * 8 / 1024 - avg_row_len * num_rows / 1024 / 1024) need_recovery_mb
6 from dba_tables where table_name='T4';
TABLE_NAME NUM_ROWS ACTUAL_MB HIGH_WATER_MB NEED_RECOVERY_MB
------------------------------ ---------- ---------- ------------- ----------------
T4 22593 .603298187 31.46875 30.8654518
SQL>
发现使劲的表大小已经变为0.603298187MB了.
--使用dbms_stat包分析分区表，如下：
SQL> EXEC DBMS_STATS.GATHER_TABLE_STATS(OWNNAME=> 'SCOTT',TABNAME=> 'T4',METHOD_OPT=> 'for all indexed columns size auto',CASCADE=> TRUE,ESTIMATE_PERCENT=> 100);
PL/SQL procedure successfully completed.
SQL> select OWNER,SEGMENT_NAME,PARTITION_NAME,SEGMENT_TYPE,BYTES/1024/1024 size_m from dba_segments where SEGMENT_NAME='T4';
OWNER SEGMENT_NAME PARTITION_NAME SEGMENT_TYPE SIZE_M
------------------------------ --------------- ------------------------------ ------------------ ----------
SCOTT T4 PART1 TABLE PARTITION 16
SCOTT T4 PART2 TABLE PARTITION 16
SQL> select table_name,
2 num_rows,
3 avg_row_len * num_rows / 1024 / 1024 actual_mb,
4 blocks * 8 / 1024 high_water_mb,
5 (blocks * 8 / 1024 - avg_row_len * num_rows / 1024 / 1024) need_recovery_mb
6 from dba_tables where table_name='T4';
TABLE_NAME NUM_ROWS ACTUAL_MB HIGH_WATER_MB NEED_RECOVERY_MB
------------------------------ ---------- ---------- ------------- ----------------
T4 22593 .538659096 31.46875 30.9300909
结果同上。
--查询每个分区的碎片情况；
SQL> select t3.table_name, t3.partition_name,t3.num_rows * t3.AVG_ROW_LEN / 1024 / 1024 actual_mb,T3.blocks * 8 / 1024 high_water_mb,(blocks * 8 / 1024 - avg_row_len * num_rows / 1024 / 1024 ) need_recovery_mb,t3.empty_blocks,t3.last_analyzed from dba_tab_partitions t3 where t3.table_name ='T4';
TABLE_NAME PARTITION_NAME ACTUAL_MB HIGH_WATER_MB NEED_RECOVERY_MB EMPTY_BLOCKS LAST_ANALYZED
------------------------------ ------------------------------ ---------- ------------- ---------------- ------------ -------------------
T4 PART1 0 15.734375 15.734375 34 2017-10-18 06:06:49
T4 PART2 .538659096 15.734375 15.1957159 34 2017-10-18 06:06:49
SQL>
--进行shrink表t4（整个分区shrink，当然也可以针对子分区shrink,例如alter table PTABLE MODIFY PARTITION PTABLE_P2 shrink space）
SQL> alter table scott.t4 enable row movement;
Table altered.
SQL>
SQL> alter table scott.t4 shrink space cascade;
Table altered.
SQL> alter table scott.t4 disable row movement;
Table altered.
--查看表大小及碎片情况：
SQL> select OWNER,SEGMENT_NAME,PARTITION_NAME,SEGMENT_TYPE,BYTES/1024/1024 size_m from dba_segments where SEGMENT_NAME='T4';
OWNER SEGMENT_NAME PARTITION_NAME SEGMENT_TYPE SIZE_M
------------------------------ --------------- ------------------------------ ------------------ ----------
SCOTT T4 PART1 TABLE PARTITION .1875
SCOTT T4 PART2 TABLE PARTITION .8125
SQL>
SQL> select table_name,
2 num_rows,
3 avg_row_len * num_rows / 1024 / 1024 actual_mb,
4 blocks * 8 / 1024 high_water_mb,
5 (blocks * 8 / 1024 - avg_row_len * num_rows / 1024 / 1024) need_recovery_mb
6 from dba_tables where table_name='T4';
TABLE_NAME NUM_ROWS ACTUAL_MB HIGH_WATER_MB NEED_RECOVERY_MB
------------------------------ ---------- ---------- ------------- ----------------
T4 22593 .538659096 31.46875 30.9300909
SQL>
--再次收集统计信息：
SQL> EXEC DBMS_STATS.GATHER_TABLE_STATS(OWNNAME=> 'SCOTT',TABNAME=> 'T4',METHOD_OPT=> 'for all indexed columns size auto',CASCADE=> TRUE,ESTIMATE_PERCENT=> 100);
PL/SQL procedure successfully completed.
SQL>
SQL> select OWNER,SEGMENT_NAME,PARTITION_NAME,SEGMENT_TYPE,BYTES/1024/1024 size_m from dba_segments where SEGMENT_NAME='T4';
OWNER SEGMENT_NAME PARTITION_NAME SEGMENT_TYPE SIZE_M
------------------------------ --------------- ------------------------------ ------------------ ----------
SCOTT T4 PART1 TABLE PARTITION .1875
SCOTT T4 PART2 TABLE PARTITION .8125
SQL> select table_name,
2 num_rows,
3 avg_row_len * num_rows / 1024 / 1024 actual_mb,
4 blocks * 8 / 1024 high_water_mb,
5 (blocks * 8 / 1024 - avg_row_len * num_rows / 1024 / 1024) need_recovery_mb
6 from dba_tables where table_name='T4';
TABLE_NAME NUM_ROWS ACTUAL_MB HIGH_WATER_MB NEED_RECOVERY_MB
------------------------------ ---------- ---------- ------------- ----------------
T4 22593 .538659096 .6640625 .125403404
--同时查看各子分区情况：
SQL> select t3.table_name, t3.partition_name,t3.num_rows * t3.AVG_ROW_LEN / 1024 / 1024 actual_mb,T3.blocks * 8 / 1024 high_water_mb,(blocks * 8 / 1024 - avg_row_len * num_rows / 1024 / 1024 ) need_recovery_mb,t3.empty_blocks,t3.last_analyzed from dba_tab_partitions t3 where t3.table_name ='T4';
TABLE_NAME PARTITION_NAME ACTUAL_MB HIGH_WATER_MB NEED_RECOVERY_MB EMPTY_BLOCKS LAST_ANALYZED
------------------------------ ------------------------------ ---------- ------------- ---------------- ------------ -------------------
T4 PART1 0 .0078125 .0078125 34 2017-10-18 06:15:37
T4 PART2 .538659096 .65625 .117590904 34 2017-10-18 06:15:37
关于shrink怎样回收分区表碎片问题的解答就分享到这里了，希望以上内容可以对大家有一定的帮助，如果你还有很多疑惑没有解开，可以关注亿速云行业资讯频道了解更多相关知识。
