still have to talk to people about standards for this sort of thing and all that. But as a first checkin – funkyness++!<\/p>\n
\r\nmysql> select * from INFORMATION_SCHEMA.DATAFILES; select * \r\nfrom INFORMATION_SCHEMA.TABLESPACES;\r\nEmpty set (0.03 sec)\r\n\r\nEmpty set (0.00 sec)\r\n\r\nmysql> CREATE TABLESPACE ts1 ADD DATAFILE 'datafile.dat' USE \r\nLOGFILE GROUP lg1 INITIAL_SIZE = 12M ENGINE NDB;\r\nQuery OK, 0 rows affected (2.35 sec)\r\n\r\nmysql> select * from INFORMATION_SCHEMA.DATAFILES; select * \r\nfrom INFORMATION_SCHEMA.TABLESPACES;\r\n+--------------+--------+--------------+----------+------+------------+\r\n| NAME | ENGINE | PATH | SIZE | FREE | TABLESPACE |\r\n+--------------+--------+--------------+----------+------+------------+\r\n| datafile.dat | NDB | datafile.dat | 12582912 | 11 | |\r\n+--------------+--------+--------------+----------+------+------------+\r\n1 row in set (0.00 sec)\r\n\r\n+------+--------+---------+-------------+-----------------------+\r\n| NAME | ENGINE | VERSION | EXTENT_SIZE | DEFAULT_LOGFILE_GROUP |\r\n+------+--------+---------+-------------+-----------------------+\r\n| ts1 | NDB | 1 | 1048576 | 0 |\r\n+------+--------+---------+-------------+-----------------------+\r\n1 row in set (0.00 sec)\r\n\r\nmysql> CREATE TABLE t1 \r\n(pk1 int not null primary key auto_increment,\r\n b int not null, \r\nc int not null) \r\ntablespace ts1 storage disk engine ndb; \r\nQuery OK, 0 rows affected (0.62 sec)\r\n\r\nmysql> insert into t1 (b,c) values (1,2),(2,3),(3,4),(1,2),(2,3),(3,4),(1,2),(2,3),(3,4),\r\n(1,2),(2,3),(3,4),(1,2),(2,3),(3,4),(1,2),(2,3),(3,4),(1,2),(2,3),(3,4),(1,2),\r\n(2,3),(3,4),(1,2),(2,3),(3,4),(1,2),(2,3),(3,4),(1,2),(2,3),(3,4),(1,2),(2,3),(3,4);\r\nQuery OK, 36 rows affected (0.11 sec)\r\nRecords: 36 Duplicates: 0 Warnings: 0\r\n\r\nmysql> select * from INFORMATION_SCHEMA.DATAFILES; select * \r\nfrom INFORMATION_SCHEMA.TABLESPACES; \r\n+--------------+--------+--------------+----------+------+------------+ \r\n| NAME | ENGINE | PATH | SIZE | FREE | TABLESPACE |\r\n+--------------+--------+--------------+----------+------+------------+\r\n| datafile.dat | NDB | datafile.dat | 12582912 | 9 | |\r\n+--------------+--------+--------------+----------+------+------------+\r\n1 row in set (0.02 sec)\r\n\r\n+------+--------+---------+-------------+-----------------------+\r\n| NAME | ENGINE | VERSION | EXTENT_SIZE | DEFAULT_LOGFILE_GROUP |\r\n+------+--------+---------+-------------+-----------------------+\r\n| ts1 | NDB | 1 | 1048576 | 0 |\r\n+------+--------+---------+-------------+-----------------------+\r\n1 row in set (0.00 sec)\r\n\r\nmysql> CREATE TABLESPACE ts2 ADD DATAFILE 'datafile2.dat' \r\nUSE LOGFILE GROUP lg1 INITIAL_SIZE = 12M ENGINE NDB;\r\nQuery OK, 0 rows affected (2.18 sec)\r\n\r\nmysql> select * from INFORMATION_SCHEMA.DATAFILES; select * \r\nfrom INFORMATION_SCHEMA.TABLESPACES;\r\n+---------------+--------+---------------+----------+------+------------+\r\n| NAME | ENGINE | PATH | SIZE | FREE | TABLESPACE |\r\n+---------------+--------+---------------+----------+------+------------+\r\n| datafile2.dat | NDB | datafile2.dat | 12582912 | 11 | |\r\n| datafile.dat | NDB | datafile.dat | 12582912 | 9 | |\r\n+---------------+--------+---------------+----------+------+------------+\r\n2 rows in set (0.02 sec)\r\n\r\n+------+--------+---------+-------------+-----------------------+\r\n| NAME | ENGINE | VERSION | EXTENT_SIZE | DEFAULT_LOGFILE_GROUP |\r\n+------+--------+---------+-------------+-----------------------+\r\n| ts1 | NDB | 1 | 1048576 | 0 |\r\n| ts2 | NDB | 1 | 1048576 | 0 |\r\n+------+--------+---------+-------------+-----------------------+\r\n2 rows in set (0.00 sec)\r\n\r\nmysql> ALTER TABLESPACE ts1 ADD DATAFILE 'datafile3.dat' \r\nINITIAL_SIZE=12M ENGINE NDB;\r\nQuery OK, 0 rows affected (1.85 sec)\r\n\r\nmysql> select * from INFORMATION_SCHEMA.DATAFILES; select *\r\n from INFORMATION_SCHEMA.TABLESPACES;\r\n+---------------+--------+---------------+----------+------+------------+\r\n| NAME | ENGINE | PATH | SIZE | FREE | TABLESPACE |\r\n+---------------+--------+---------------+----------+------+------------+\r\n| datafile2.dat | NDB | datafile2.dat | 12582912 | 11 | |\r\n| datafile3.dat | NDB | datafile3.dat | 12582912 | 11 | |\r\n| datafile.dat | NDB | datafile.dat | 12582912 | 9 | |\r\n+---------------+--------+---------------+----------+------+------------+\r\n3 rows in set (0.02 sec)\r\n\r\n+------+--------+---------+-------------+-----------------------+\r\n| NAME | ENGINE | VERSION | EXTENT_SIZE | DEFAULT_LOGFILE_GROUP |\r\n+------+--------+---------+-------------+-----------------------+\r\n| ts1 | NDB | 1 | 1048576 | 0 |\r\n| ts2 | NDB | 1 | 1048576 | 0 |\r\n+------+--------+---------+-------------+-----------------------+\r\n2 rows in set (0.00 sec)\r\n<\/pre>\nthe ‘free’ column is really the number of free extents. Not exactly ideal… maybe… but since that’s the unit of allocation in the data files, it sort of makes sense. The other option is to list number of extents * extent size. Maybe that’s clearer for people… there is the option of denormalising the tables and have extent size in the DATAFILES table too. There is something in my brain that makes that a hard leap though.<\/p>\n
Although…. if you’re going to be querying the tables directly and not just using a pretty gui on top of it all, you should probably know what you’re doing anyway.<\/p>\n
Although, both a great benefit (and curse) of commoditising the database market is the fact that you get all sorts as users. This is interesting in cluster as it is naturally a bit more complex than a simple client-server RDBMS.<\/p>\n
we also need a NODE column as well. which will probably cause confusion for non-cluster users and the like :)<\/p>\n
(for the unintiated, this is work being done in a branch off the 5.1 tree for NDB disk data. we’ll push it to the main 5.1 tree at some point). don’t go thinking this is production ready any time soon (in other words insert a standard disclaimer).<\/p>\n","protected":false},"excerpt":{"rendered":"
still have to talk to people about standards for this sort of thing and all that. But as a first checkin – funkyness++! mysql> select * from INFORMATION_SCHEMA.DATAFILES; select * from INFORMATION_SCHEMA.TABLESPACES; Empty set (0.03 sec) Empty set (0.00 sec) … Continue reading