Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

sql-statements: improve SHOW TABLE REGIONS examples #3389

Merged
merged 8 commits into from Jul 23, 2020
78 changes: 61 additions & 17 deletions sql-statements/sql-statement-show-table-regions.md
Original file line number Diff line number Diff line change
Expand Up @@ -57,43 +57,83 @@ Executing `SHOW TABLE REGIONS` returns the following columns:

## Examples

Create an example table with enough data that it fills a few regions:
This conversation was marked as resolved.
Show resolved Hide resolved

{{< copyable "sql" >}}

```sql
test> create table t (id int key,name varchar(50), index (name));
Query OK, 0 rows affected
CREATE TABLE t1 (
id INT NOT NULL PRIMARY KEY auto_increment,
b INT NOT NULL,
pad1 VARBINARY(1024),
pad2 VARBINARY(1024),
pad3 VARBINARY(1024)
);
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM dual;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
INSERT INTO t1 SELECT NULL, FLOOR(RAND()*1000), RANDOM_BYTES(1024), RANDOM_BYTES(1024), RANDOM_BYTES(1024) FROM t1 a JOIN t1 b JOIN t1 c LIMIT 10000;
SELECT SLEEP(5);
SHOW TABLE t1 REGIONS;
```

After a table is created, the table data is stored in a newly split Region by default. In this initial phase, all row data and index data of the table are written into this Region.
The output should show that the table is split into regions. The `REGION_ID`, `START_KEY` and `END_KEY` may not match exactly:
This conversation was marked as resolved.
Show resolved Hide resolved

```sql
test> show table t regions;
+-----------+-----------+---------+-----------+-----------------+-----------+------------+---------------+------------+----------------------+------------------+
| REGION_ID | START_KEY | END_KEY | LEADER_ID | LEADER_STORE_ID | PEERS | SCATTERING | WRITTEN_BYTES | READ_BYTES | APPROXIMATE_SIZE(MB) | APPROXIMATE_KEYS |
+-----------+-----------+---------+-----------+-----------------+-----------+------------+---------------+------------+----------------------+------------------+
| 3 | t_43_ | | 73 | 9 | 5, 73, 93 | 0 | 35 | 0 | 1 | 0 |
+-----------+-----------+---------+-----------+-----------------+-----------+------------+---------------+------------+----------------------+------------------+
1 row in set
..
This conversation was marked as resolved.
Show resolved Hide resolved
mysql> SHOW TABLE t1 REGIONS;
+-----------+--------------+--------------+-----------+-----------------+-------+------------+---------------+------------+----------------------+------------------+
| REGION_ID | START_KEY | END_KEY | LEADER_ID | LEADER_STORE_ID | PEERS | SCATTERING | WRITTEN_BYTES | READ_BYTES | APPROXIMATE_SIZE(MB) | APPROXIMATE_KEYS |
+-----------+--------------+--------------+-----------+-----------------+-------+------------+---------------+------------+----------------------+------------------+
| 94 | t_75_ | t_75_r_31717 | 95 | 1 | 95 | 0 | 0 | 0 | 112 | 207465 |
| 96 | t_75_r_31717 | t_75_r_63434 | 97 | 1 | 97 | 0 | 0 | 0 | 97 | 0 |
| 2 | t_75_r_63434 | | 3 | 1 | 3 | 0 | 269323514 | 66346110 | 245 | 162020 |
+-----------+--------------+--------------+-----------+-----------------+-------+------------+---------------+------------+----------------------+------------------+
3 rows in set (0.00 sec)
```

In the above result, `t_43_` is the value of `START_KEY` row. In this value, `t` is the table prefix and `43` is the table ID. The value of `END_KEY` row is empty (""), which means that it is an infinite value.
In the output above, a `START_KEY` of `t_75_r_31717` and `END_KEY` of `t_75_r_63434` shows that data with a PRIMARY KEY between `31717` and `63434` will be stored in this region. The prefix `t_75_` indicates that this is the region for a table (`t`) which has an internal table ID of `75`. An empty key value for `START_KEY` or `END_KEY` indicates negative infinity or positive infinity respectively.
This conversation was marked as resolved.
Show resolved Hide resolved

Use the `SPLIT TABLE REGION` statement to split row data into five Regions.
TiDB will automatically rebalance regions as needed. For manual rebalancing, use the `SPLIT TABLE REGION` statement:
This conversation was marked as resolved.
Show resolved Hide resolved

```sql
test> split table t between (0) and (100000) regions 5;
mysql> SPLIT TABLE t1 BETWEEN (31717) AND (63434) REGIONS 2;
+--------------------+----------------------+
| TOTAL_SPLIT_REGION | SCATTER_FINISH_RATIO |
+--------------------+----------------------+
| 5 | 1.0 |
| 1 | 1 |
+--------------------+----------------------+
1 row in set (42.34 sec)

mysql> SHOW TABLE t1 REGIONS;
+-----------+--------------+--------------+-----------+-----------------+-------+------------+---------------+------------+----------------------+------------------+
| REGION_ID | START_KEY | END_KEY | LEADER_ID | LEADER_STORE_ID | PEERS | SCATTERING | WRITTEN_BYTES | READ_BYTES | APPROXIMATE_SIZE(MB) | APPROXIMATE_KEYS |
+-----------+--------------+--------------+-----------+-----------------+-------+------------+---------------+------------+----------------------+------------------+
| 94 | t_75_ | t_75_r_31717 | 95 | 1 | 95 | 0 | 0 | 0 | 112 | 207465 |
| 98 | t_75_r_31717 | t_75_r_47575 | 99 | 1 | 99 | 0 | 1325 | 0 | 53 | 12052 |
| 96 | t_75_r_47575 | t_75_r_63434 | 97 | 1 | 97 | 0 | 1526 | 0 | 48 | 0 |
| 2 | t_75_r_63434 | | 3 | 1 | 3 | 0 | 0 | 55752049 | 60 | 0 |
+-----------+--------------+--------------+-----------+-----------------+-------+------------+---------------+------------+----------------------+------------------+
4 rows in set (0.00 sec)
```

In the above example:
The above output shows that region 96 was split, with a new region 98 being created. The remaining regions in the table were unaffected by the split operation. This is confirmed by the output statistics:
This conversation was marked as resolved.
Show resolved Hide resolved

* `TOTAL_SPLIT_REGION` indicates the number of newly split Regions. In this example, the number is 5.
* `TOTAL_SPLIT_REGION` indicates the number of newly split Regions. In this example, the number is 1.
* `SCATTER_FINISH_RATIO` indicates the rate at which the newly split Regions are successfully scattered. `1.0` means that all Regions are scattered.

For a more detailed example:

```sql
test> show table t regions;
mysql> show table t regions;
+-----------+--------------+--------------+-----------+-----------------+---------------+------------+---------------+------------+----------------------+------------------+
| REGION_ID | START_KEY | END_KEY | LEADER_ID | LEADER_STORE_ID | PEERS | SCATTERING | WRITTEN_BYTES | READ_BYTES | APPROXIMATE_SIZE(MB) | APPROXIMATE_KEYS |
+-----------+--------------+--------------+-----------+-----------------+---------------+------------+---------------+------------+----------------------+------------------+
Expand Down Expand Up @@ -155,6 +195,10 @@ test> show table t regions;
7 rows in set
```

## MySQL compatibility

This statement is a TiDB extension to MySQL syntax.

## See also

* [SPLIT REGION](/sql-statements/sql-statement-split-region.md)
Expand Down