Storage array compression based on the structure of the data being compressed
In one aspect, a method includes storing a database on a storage array, determining a plurality of most used values in the database, forming a table using the plurality of most used values and compressing data in the database according to the table. In another aspect, an apparatus includes electronic hardware circuitry configured to store a database on a storage array, determine a plurality of most used values in the database, form a table using the plurality of most used values and compress data in the database according to the table. In a further aspect, an article includes a computer-readable medium that stores computer-executable instructions. The instructions cause a machine to store a database on a storage array, determine a plurality of most used values in the database, form a table using the plurality of most used values and compress data in the database according to the table.
1. A method comprising:
determining a plurality of most used values in a database stored on a storage array;
forming a table containing the plurality of most used values;
retrieving statistical information about the database;
identifying a structure of data from the statistical information;
identifying a manner in which the data is stored;
selecting a compression technique from a plurality of compression techniques as a function of the structure of the data and the manner in which the data is stored; and
compressing the data in the database according to the table and selected compression technique;
wherein upon selecting a block and table level compression as the compression technique, the compressing further comprises:
generating a global mapping table from the table by mapping a key to each of a corresponding one of the most used values; and
generating a block mapping table according to one or more keys from the global mapping table, the block mapping table substituting the most used values from the global mapping table with corresponding keys;
wherein compression logic for performing the plurality of compression techniques is stored in the storage array, and the compression technique is selected prior to writing to the storage array.
2. The method of claim 1 , wherein upon selecting a block level compression as the compression technique, the compressing the data in the database according to the table and the selected compression technique comprises, for each header of each block in the table, adding pointers to point to other data associated with the data.
3. The method of claim 2 , wherein the block contains database records spanning multiple database tables of the database.
4. The method of claim 1 , wherein upon selecting the block and table level compression as the compression technique, the compressing the data in the database according to the table further comprises, for each block in the block mapping table, adding pointers to point to other data associated with the data.
5. The method of claim 1 , wherein the statistical information about the database is retrieved from a histogram.
6. The method of claim 1 , further comprising:
selecting data to compress based on a type of storage device in the storage array.
7. The method of claim 1 , wherein the plurality of most used values comprises a user-tunable parameter.
8. The method of claim 1 , wherein the global mapping table is configured for access across a plurality of nodes in the storage array.
9. The method of claim 1 , wherein the compression technique selected is a table-level compression technique and the forming a table using the plurality of most used values comprises forming the global mapping table, wherein mapping corresponds to all blocks within the database.
10. An apparatus, comprising:
electronic hardware circuitry configured to:
determine a plurality of most used values in a database stored on a storage array;
form a table containing the plurality of most used values;
retrieve statistical information about the database;
identify a structure of data from the statistical information;
identify a manner in which the data is stored;
select a compression technique from a plurality of compression techniques as a function of the structure of the data and the manner in which the data is stored; and
compress the data in the database according to the table and selected compression technique;
wherein upon selecting a block and table level compression as the compression technique, the compressing further comprises:
generating a global mapping table from the table by mapping a key to each of a corresponding one of the most used values; and
generating a block mapping table according to one or more keys from the global mapping table, the block mapping table substituting the most used values from the global mapping table with corresponding keys;
wherein compression logic for performing the plurality of compression techniques is stored in the storage array, and the compression technique is selected prior to writing to the storage array.
11. The apparatus of claim 10 , wherein the circuitry comprises at least one of a processor, a memory, a programmable logic device or a logic gate.
12. The apparatus of claim 11 , wherein the processor determines a system state, and the compressing the data in the database according to the table and selected compression technique includes compressing the data upon determining the system state indicates a value representing a non-busy state.
13. The apparatus of claim 10 , wherein upon selecting a block level compression as the compression technique, the circuitry configured to compress the data in the database according to the table comprises circuitry configured to, for each header of each block in the table, add pointers to point to other data associated with the data.
14. The apparatus of claim 10 , wherein upon selecting the block and table level compression as the compression technique, the circuitry configured to compress the data in the database according to the table further comprises, for each block in the block mapping table, add pointers to point to other data associated with the data.
15. The apparatus of claim 10 , wherein the storage array is a flash disk array.
16. An article comprising:
a non-transitory computer-readable medium that stores computer-executable instructions, the instructions causing a machine to:
determine a plurality of most used values in a database stored on a storage array;
form a table containing the plurality of most used values;
retrieve statistical information about the database;
identify a structure of data from the statistical information;
identify a manner in which the data is stored;
select a compression technique from a plurality of compression techniques as a function of the structure of the data and the manner in which the data is stored; and
compress the data in the database according to the table and selected compression technique;
wherein upon selecting a block and table level compression as the compression technique, the compressing further comprises:
generating a global mapping table from the table by mapping a key to each of a corresponding one of the most used values; and
generating a block mapping table according to one or more keys from the global mapping table, the block mapping table substituting the most used values from the global mapping table with corresponding keys;
wherein compression logic for performing the plurality of compression techniques is stored in the storage array, and the compression technique is selected prior to writing to the storage array.
17. The article of claim 16 , wherein upon selecting a block level compression as the compression technique, the instructions causing the machine to compress the data in the database according to the table comprises instructions causing the machine to, for each header of each block in the table, add pointers to point to other data associated with the data.
18. The article of claim 16 , wherein upon selecting the block and table level compression as the compression technique, the instructions causing the machine to compress the data in the database according to the table further comprises instructions causing the machine to, for each block in the block mapping table, add pointers to point to other data associated with the data.