Increasing buffer locality during multiple table access operations

• US 8,010,766 B2
• Filed: 10/12/2006
• Issued: 08/30/2011
• Est. Priority Date: 10/12/2006
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bhattacharjee, Bishwaranjan, Padmanabhan, Sriram, Lang, Christian A., Malkemus, Timothy R., Wong, Kwai
• Primary Examiner(s): Elmore; Reba I
• Application Number:
  US11/548,875
• Publication Number:
  US 20080144128A1
• Time in Patent Office: 1,783 Days
• Field of Search: 711/159, 711/154, 711/166, 711/168
• US Class Current: 711/168
• CPC Class Codes: G06F 16/24552 Database cache management
• CPC Class Codes: G06F 16/9017 using directory or table look-up use of a directory or look-up table in file systems G06F16/13
• CPC Class Codes: G06F 16/90344 by using string matching techniques

Abstract

Disclosed are a method, information processing system, and computer readable medium for managing table scan processes. The method includes monitoring a plurality of storage medium table scan processes. Each storage medium table scan process in the plurality of storage medium table scan processes is placed into a plurality of scan groups based on storage medium pages to be scanned by each of the storage medium table scan processes. Each storage medium table scan process in a scan group can share data within a storage medium page.

Claims

1. A method in a table scan manager for managing table scan processes, the method comprising:

monitoring a plurality of storage medium table scan processes that are reading data out of blocks of memory in a fast memory to identify a set of data that satisfies a query associated with the scanning processes;

placing each storage medium table scan process, in the plurality of storage medium table scan processes, into a set of scans organized as a plurality of scan groups based on blocks of memory in the fast memory to be scanned by each of the storage medium table scan processes, wherein each storage medium table scan process in a scan group shares data within a block of memory with other storage medium table scan process in the scan group;

retrieving, for a front most storage medium table scan process in a scan group, a set of data from a slow memory;

storing the set of data that has been retrieved into blocks of memory in the fast memory associated with the scan group; and

maintaining the set of data in the blocks of memory for all storage medium table scan processes that are subsequent to the front most storage medium table scan process in the scan group.

2. The method of claim 1, wherein the placing each storage medium table scan process in the plurality of storage medium table scan processes further comprises:

sorting each storage medium table scan process within a scan group based on a current scanning location of each storage medium table scan process.

3. The method of claim 1, further comprising at least one of:

associating statistical information with at least one scan group in the plurality of scan groups; and

associating statistical information with at least one storage medium table scan process in the at least one scan group.

4. The method of claim 3, wherein the statistical information associated with the at least one scan group includes at least one of:

an average scan speed of the at least one scan group; and

an age of the scan group.

5. The method of claim 3, wherein the statistical information associated with the at least one storage medium table scan process includes at least one of:

an average scan speed of the at least one storage medium table scan process; and

an age of the at least one storage medium table scan process.

6. The method of claim 1, further comprising:

detecting a new storage medium table scan request; and

determining a starting location within a storage medium table associated with the new storage medium table scan request based on information associated with at least the plurality of scan groups.

7. The method of claim 6, wherein the determining the starting location further comprises:

determining if at least one scan group exists;

if at least one scan group fails to exist, creating a new scan group;

placing a storage medium table scan process associated with the new storage medium table scan request into the new scan group;

selecting the starting location to be one of;

a starting storage medium page of the storage medium table; and

a storage medium page within the storage medium table associated with a previous storage medium table scan process; and

notifying the storage medium table scan process of the starting location which has been selected.

8. The method of claim 7, wherein if at least one scan group does exist:

selecting a scan group from the at least one scan group based on at least statistical information associated with the at least one scan group;

selecting a location within the selected scan group to place the storage medium table scan process based on at least statistical information associated with each storage medium table scan process within the selected scan group;

placing the storage medium table scan process within the selected scan group at the selected location, wherein selected location indicates a starting location of the storage medium table scan process; and

notifying the storage medium table scan process of the selected starting location.

9. The method of claim 1, further comprising:

updating a scan location associated with at least one of the storage medium table scan processes.

10. The method of claim 9, wherein the updating further comprises:

receiving scan location update information from the storage medium table scan process, wherein the scan location update information is associated with a new scan location of the storage medium table scan process;

removing the storage medium table scan process from its current scan group;

determining if at least on scan group is available;

if at least one scan group fails to be available, creating a new scan group;

placing the storage medium table scan process into the new scan group;

if at least one scan group is available, determining if the new scan location is located within the at least one scan group;

placing, in response to the new scan location being located within the at least one scan group, the storage medium table scan process into the at least one scan group.

11. The method of claim 10, wherein the placing the storage medium table scan process into the new scan group further comprises:

determining, in response to the new scan location failing to be located within the at least one scan group, if regrouping at least two scan groups in the at least one scan group can accommodate the new scan location;

regrouping, in response to determining that regrouping can accommodate the new scan location, the at least two scan groups;

placing the storage medium table scan process into a regrouped scan group in which the new scan location of the storage medium table scan process is located;

determining, in response to determining that regrouping fails to accommodate the new scan location, if at least scan group can be increased in size;

increasing, in response to at least one scan being able to be increased in size, the size of the at least one group;

placing the storage medium table scan process in to the at least one group;

creating, in response to at least one scan group failing to be able to be increased in size, a new scan group; and

placing the storage medium table scan process into the new scan group.

12. The method of claim 1, further comprising:

determining that a leading storage medium table scan in scan group has changed scanning locations;

determining a distance between the leading storage medium table scan and a trailing storage medium table scan, wherein the trailing storage medium table scan is a last storage medium table scan in the scan group;

determining if the distance is greater than a first given threshold associated with the scan group;

if the distance is greater than the given threshold, signaling the leading storage medium table scan to halt its scanning operation; and

notifying the leading storage medium table scan to resume its scanning operating once the distance between the leading storage medium table scan and the trailing storage medium table scan is below a second given threshold.

13. An information processing system for managing table scan processes, the information processing system comprising:

a memory;

a processor communicatively coupled to the memory; and

a table scan manager communicatively coupled to the memory and the processor, wherein the table scan manager is for;

monitoring a plurality of storage medium table scan processes that are reading data out of blocks of memory in a fast memory to identify a set of data that satisfies a query associated with the scanning processes;

placing each storage medium table scan process, in the plurality of storage medium table scan processes, into a set of scans organized as a plurality of scan groups based on blocks of memory in the fast memory to be scanned by each of the storage medium table scan processes, wherein each storage medium table scan process in a scan group shares data within a block of memory with other storage medium table scan process in the scan group;

retrieving, for a front most storage medium table scan process in a scan group, a set of data from a slow memory;

storing the set of data that has been retrieved into blocks of memory in the fast memory associated with the scan group; and

maintaining the set of data in the blocks of memory for all storage medium table scan processes that are subsequent to the front most storage medium table scan process in the scan group.

14. The information processing system of claim 13, wherein the placing further comprises:

sorting each storage medium table scan process within a scan group based on a current scanning location of each storage medium table scan process.

15. The information processing system of claim 13, wherein the table scan manager is for at least one of:

associating statistical information with at least one scan group in the plurality of scan groups; and

associating statistical information with at least one storage medium table scan process in the at least one scan group.

16. The information processing system of claim 13, wherein the table scan manager is for:

detecting a new table scan request;

determining a starting location within a storage medium table associated with the new table scan request based on information associated with at least the plurality of scan groups; and

updating a scan location associated with at least one of the storage medium table scan processes.

17. A computer readable storage medium for managing table scan processes, the computer readable medium comprising instructions for:

monitoring a plurality of storage medium table scan processes that are reading data out of blocks of memory in a fast memory to identify a set of data that satisfies a query associated with the scanning processes;

placing each storage medium table scan process, in the plurality of storage medium table scan processes, into a set of scans organized as a plurality of scan groups based on blocks of memory in the fast memory to be scanned by each of the storage medium table scan processes, wherein each storage medium table scan process in a scan group shares data within a block of memory with other storage medium table scan process in the scan group;

retrieving, for a front most storage medium table scan process in a scan group, a set of data from a slow memory;

storing the set of data that has been retrieved into blocks of memory in the fast memory associated with the scan group; and

maintaining the set of data in the blocks of memory for all storage medium table scan processes that are subsequent to the front most storage medium table scan process in the scan group.

18. The computer readable storage medium of claim 17, wherein the instructions for placing further comprises instructions for:

sorting each storage medium table scan process within a scan group based on a current scanning location of each storage medium table scan process.

19. The computer readable storage medium of claim 17, further comprising instructions for at least one of:

associating statistical information with at least one scan group in the plurality of scan groups; and

associating statistical information with at least one storage medium table scan process in the at least one scan group.

20. The computer readable storage medium of claim 17, further comprising instructions for:

detecting a new table scan request;

determining a starting location within a storage medium table associated with the new table scan request based on information associated with at least the plurality of scan groups; and

updating a scan location associated with at least one of the storage medium table scan processes.