Database management system and method

• US 20070022100A1
• Filed: 07/21/2006
• Published: 01/25/2007
• Est. Priority Date: 07/22/2005
• Status: Active Grant

• Current Assignee: Kazuo Goda, Masaru Kitsuregawa
• Original Assignee: Masaru Kitsuregawa
• Inventors: Kitsuregawa, Masaru, Goda, Kazuo
• Granted Patent: US 7,827,167 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24524 Access plan code generation and invalidation; Reuse of access plans
• CPC Class Codes: G06F 16/24542 Plan optimisation
• CPC Class Codes: G06F 16/24549 Run-time optimisation
• CPC Class Codes: G06F 16/24566 Recursive queries

Abstract

A database management system has a query interface for receiving a query and a query executor for executing the received query. The query executor dynamically generates a task, and executes a plurality of tasks in parallel. In execution of each task, each time data is required to be read from a database, the query executor generates a task for acquiring the data, and in execution of the generated task, issues a data read request for reading the data from the database, to thereby shorten a time taken to execute each task.

Claims

1. A database management system comprising:

a query interface for receiving a query;

a query plan generator for generating a query plan from said received query; and

a query executor for executing said received query in accordance with said generated query plan,

wherein said query executor;

executes a database processing for a database in accordance with said query plan;

in said database processing, generates a new task each time each data is accessed, in accordance with a plurality set of data contained in said database, a task from which said new task originates processing data following said data accessed;

issues a data read request for reading data from said database in the generated new task, acquires said data from said database, and executes said database processing for said acquired data;

executes said database processing for each of a plurality of generated new tasks, and thereafter eliminates each of said new tasks; and

executes in parallel at least two tasks of said plurality of generated new tasks.

2. The database management system according to claim 1, wherein said query executor suspends to generate a new task if the number of presently existing tasks reaches a predetermined number, and generates a new task if the number of presently existing tasks becomes smaller than said predetermined number.

3. The database management system according to claim 1, further comprising a read sequence controller for receiving a plurality of data read requests issued from said query executor, and issuing said plurality of received data read requests in a sequence based on block addresses corresponding to said plurality of received data read requests.

4. The database management system according to claim 3, further comprising an I/O optimizer having a plurality of queues having different priority degrees, wherein in executing said generated new task, said query executor distributes said plurality of data read requests to said queues corresponding to priority degrees corresponding to contents representative of emergency of I/O requests contained in said received query, and said read sequence controller issues data read requests stored in higher priority degree queues, taking priority over other data read request.

5. The database management system according to claim 4, wherein upon reception of a predetermined command, said read sequence controller moves at least one read request stored in a queue having a certain priority degree to a queue having a different priority degree.

6. The database management system according to claim 4, wherein said query executor decides the queue having what priority degree is used for storing said data read request in accordance with contents of said query plan or performance requirements for execution of said received query.

7. The database management system according to claim 4, further comprising a task manager for calculating the number of left steps in accordance with said query plan and raising the priority degree of a task having a smaller calculated number of left steps, wherein said query executor stores said data read request issued by executing said task having said raised priority degree, in the queue corresponding to said raised priority degree.

8. The database management system according to claim 3, wherein in executing said generated new task, said query executor suspends to acquire said data in said database, and after said date is acquired, resumes execution of said task in a same sequence as a data acquisition suspend start sequence.

9. A database management method for a database management system connected to a database via a network, comprising:

a step of receiving a query;

a step of generating a query plan from said received query;

a step of executing a database processing in accordance with said query plan;

a step of, in said database processing, generating a new task each time each data is accessed, in accordance with a plurality set of data contained in said database, a task from which said new task originates processing data following said data accessed;

a step of issuing a data read request for reading data from said database in the generated new task, acquiring said data from said database, and executing said database processing for said acquired data;

a step of executing said database processing for each of a plurality of generated new tasks, and thereafter eliminating each of said new tasks; and

a step of executing in parallel at least two tasks of said plurality of generated new tasks.

10. The database management method according to claim 9, wherein a step of suspending to generate a new task if the number of presently existing tasks reaches a predetermined number, and generating a new task if the number of presently existing tasks becomes smaller than said predetermined number.

11. The database management method to claim 10, further comprising a step of receiving a plurality of data read requests issued from said query executor, and issuing said plurality of received data read requests in a sequence based on block addresses corresponding to said plurality of received data read requests.

12. The database management method according to claim 11, further comprising a step of, in executing said generated new task, distributing said plurality of data read requests to said queues corresponding to priority degree corresponding to contents representative of emergency of I/O requests contained in said received query, and a step of issuing data read requests stored in higher priority degree queues, taking priority over other data read request.

13. The database management method according to claim 12, further comprising a step of, upon reception of a predetermined command, moving at least one read request stored in a queue having a certain priority degree to a queue having a different priority degree.

14. The database management method according to claim 13, further comprising a step of deciding the queue having what priority degree is used for storing said data read request in accordance with contents of said query plan or performance requirements for execution of said received query.

15. The database management method according to claim 14, further comprising a step of calculating the number of lest steps in accordance with said query plan and raising the priority degree of a task having a smaller calculated number of left steps, and a step of storing said data read request issued by executing said task having said raised priority degree, in the queue corresponding to said raised priority degree.

16. The database management method according to claim 15, further comprising a step of, in executing said generated new task, suspending to acquire said data in said database, and a step of, after said date is acquired, resuming execution of said task in a same sequence as a data acquisition suspend start sequence.

17. A computer system comprising:

a query interface for receiving a query;

a query plan generator for generating a query plan from said received query; and

a query executor for executing said received query in accordance with said generated query plan,

wherein said query executor;

executes a database processing for a database in accordance with said query plan;

in said database processing, generates a new task each time each data is accessed, in accordance with a plurality set of data contained in said database, a task from which said new task originates processing data following said data accessed;

issues a data read request for reading data from said database in the generated new task, acquires said data from said database, and executes said database processing for said acquired data;

executes said database processing for each of a plurality of generated new tasks, and thereafter eliminates each of said new tasks;

suspends to generate a new task if the number of presently existing tasks reaches a predetermined number, and generates a new task if the number of presently existing tasks becomes smaller than said predetermined number; and

executes in parallel at least two tasks of said plurality of generated new tasks.

18. The computer system according to claim 17, wherein in executing said generated new task, said query executor suspends to acquire said data in said database, and after said date is acquired, resumes execution of said task in a same sequence as a data acquisition suspend start sequence.

19. The computer system according to claim 18, further comprising an I/O optimizer having a plurality of queues having different priority degrees, wherein in executing said generated new task, said query executor distributes said plurality of data read requests to said queues corresponding to priority degrees corresponding to contents representative of emergency of I/O requests contained in said received query, and said read sequence controller issues data read requests stored in higher priority degree queues, taking priority over other data read request.

20. The computer system according to claim 19, wherein upon reception of a predetermined command, said read sequence controller moves at least one read request stored in a queue having a certain priority degree to a queue having a different priority degree.