procedure performs a major compaction on a table. The compaction is
performed on the table and on all of its index and constraint tables.
A major compaction actually reads every block of data from the every store file in a Region, and rewrites only the live data to a single store file. This permanently deletes the rows that were previously marked as deleted. HBase runs major compactions on a scheduled interval, which is specified in the
hbase.hregion.majorcompaction property; the default value for this property in Splice Machine is 7 days.
Splice Machine recommends running a major compaction on a table after you’ve imported a large dataset into the table or deleted a large number of rows from the table in your database.
For more information about compactions, see Using Compaction and Vacuuming.
SYSCS_UTIL.SYSCS_PERFORM_MAJOR_COMPACTION_ON_TABLE( schemaName, tableName)
A string that specifies the Splice Machine schema name to which the table belongs.
A string that specifies name of the Splice Machine table on which to perform the compaction.
Major compaction is synchronous, which means that when you invoke this procedure from the command line, your command line prompt won’t be available again until the compaction completes, which can take a little time.
This procedure does not return a result.
splice> CALL SYSCS_UTIL.SYSCS_PERFORM_MAJOR_COMPACTION_ON_TABLE('SPLICE','Pitching'); Statement executed.