Predictive memory management
A query for data stored in a database that includes a set of segments is received at a computer system. The set of segments are divided into a plurality of columns and at least one column of the plurality of columns includes one or more fields. The system analyzes the query to determine fields required to be retrieved from the database. The system determines whether a required field of the query is located in a main memory of the computer system. The system creates an input/output request for a column containing the required field for a plurality of segments of the set of segments prior to executing the query.
1. A method of retrieving information from a computer system, comprising:
receiving a query for data stored in a database that includes a set of segments, wherein the query comprises one or more fields, wherein the set of segments are divided into a plurality of columns, wherein at least one column of the plurality of columns includes the one or more fields;
analyzing the query to determine which of the one or more fields are required to be retrieved from the database;
determining that a required field of the query is not located in a main memory of the computer system;
in response to determining that the required field of the query is not located in the main memory of the computer system, creating an input/output request for a column containing the required field for a plurality of segments of the set of segments prior to executing the query; and
reducing an amount of time needed to execute the query at least in part by:
loading data of the column containing the required field from a long term storage that includes the database into the main memory; and
executing the query when at least a portion of data corresponding to the one or more fields is stored in the main memory.
2. The method of claim 1 , wherein the at least the portion of data corresponding to the one or more fields corresponds to the data of the column.
3. The method of claim 1 , wherein correlated fields are included together in a column.
4. The method of claim 3 , wherein the correlated fields are frequently part of a same query.
5. The method of claim 1 , wherein creating an input/output request for a column containing the required field for a plurality of segments of the set of segments prior to executing the query further includes determining whether a segment of the set of segments stores values of the required field that are within a range of values specified in the query.
6. The method of claim 1 , wherein creating an input/output request for a column containing the required field for a plurality of segments of the set of segments prior to executing the query further includes creating a list of memory segments and columns corresponding to the required field.
7. The method of claim 1 , wherein the loaded data overwrites data that corresponds to least frequently queried fields previously stored in the main memory.
8. The method of claim 1 , wherein the set of segments are sequentially stored in the long term storage.
9. The method of claim 1 , wherein the set of segments are stored in the long term storage in an order based at least in part on the columns.
10. The method of claim 1 , wherein the query is executed first on segments already found in main memory.
11. A system for retrieving information from a computer system, comprising:
main memory;
long term storage; and
a query analyzer implemented on a processor configured to:
receive a query for data stored in a database that includes a set of segments, wherein the query comprises one or more fields, wherein the set of segments are divided into a plurality of columns, wherein at least one column of the plurality of columns includes one or more fields;
analyze the query to determine which of the one or more fields are required to be retrieved from the database;
determine that a required field of the query is not located in the main memory;
in response to determining that the required field of the query is not located in the main memory of the computer system, create an input/output request for a column containing the required field for a plurality of segments of the set of segments prior to executing the query; and
reduce an amount of time needed to execute the query at least in part by:
load data of the column containing the required field from the long term storage into the main memory; and
execute the query when at least a portion of the data corresponding to the one or more fields is stored in the main memory.
12. The system of claim 11 , wherein the at least the portion of data corresponding to the one or more fields corresponds to the data of the column.
13. The system of claim 11 , wherein correlated fields are included together in a column.
14. The system of claim 13 , wherein the correlated fields are frequently part of a same query.
15. The system of claim 11 , wherein to create an input/output request for a column containing the required field for a plurality of segments of the set of segments prior to executing the query further includes to determine whether a segment of the set of segments stores values of the required field that are within a range of values specified in the query.
16. The system of claim 11 , wherein to create an input/output request for a column containing the required field for a plurality of segments of the set of segments prior to executing the query further includes to create a list of memory segments and columns corresponding to the required field.
17. The system of claim 11 , wherein the loaded data overwrites data that corresponds to least frequently queried fields previously stored in the main memory.
18. The system of claim 11 , wherein the set of segments are sequentially stored in the long term storage.
19. The system of claim 11 , wherein the set of segments are stored in the long term storage in an order based on the columns.
20. The system of claim 11 , wherein the query is executed first on segments already found in main memory.