Shared temporary storage management in a shared disk database cluster

• US 9,047,019 B2
• Filed: 11/08/2011
• Issued: 06/02/2015
• Est. Priority Date: 11/08/2011
• Status: Active Grant

• Current Assignee: Sybase Incorporated (SAP SE)
• Original Assignee: Sybase Incorporated (SAP SE)
• Inventors: Florendo, Colin Joseph
• Primary Examiner(s): Choe, Yong
• Assistant Examiner(s): Chappell, Daniel C
• Application Number:
  US13/291,157
• Publication Number:
  US 20130117526A1
• Time in Patent Office: 1,302 Days
• Field of Search: 711/171
• US Class Current: 1/1
• CPC Class Codes: G06F 3/0622 in relation to access
• CPC Class Codes: G06F 3/0637 Permissions
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]

Abstract

System, method, computer program product embodiments and combinations and sub-combinations thereof for temporary storage management in a shared disk database cluster are provided. Included is the reserving of units on-demand and of variable size from shared temporary storage space in the SDC. The utilization of the reserved units of the shared temporary storage space is tracked, and the shared temporary storage space is administered based on the tracking.

Claims

1. A method for shared temporary storage management in a shared disk database cluster (SDC), the method comprising:

reserving units on-demand and of variable size from shared temporary storage space in the SDC, wherein the reserving controls a size of the reserved units based upon configuration of the SDC, remaining space of the shared temporary storage space, and a number of configured nodes, and wherein the shared temporary storage space exists independent from the reserving units;

tracking utilization of the reserved units of the shared temporary storage space, wherein coordinator and secondary nodes of the SDC maintain bitmaps which track the utilization of the reserved units, and each bit in the bitmaps represents a logical disk block and tracks whether the logical disk block is free or in use, and wherein the secondary nodes periodically send content of the bitmaps associated with the secondary nodes to the coordinator node; and

administering the shared temporary storage space based on the tracking.

2. The method of claim 1 wherein tracking further comprises updating the bitmaps with respect to space allocation activity for chains of reserved units.

3. The method of claim 1 wherein tracking further comprises updating the bitmaps with respect to timed expiration periods.

4. The method of claim 3, wherein the expiration period is reset when a new reservation unit is added to the bitmap.

5. The method of claim 1 wherein administering further comprises utilizing a garbage collection event.

6. The method of claim 5 wherein the garbage collection event handles expired reserved units and deallocated temporary storage space for reuse in the SDC.

7. A shared disk database cluster (SDC) system with temporary storage management comprising:

shared-disk storage; and

a plurality of data processing nodes reserving units on-demand and of variable size from shared temporary storage space of the shared-disk storage, wherein the data processing nodes reserving units controls a size of the reserved units based upon configuration of the SDC, remaining space of the shared temporary storage space, and a number of configured nodes, and wherein the shared temporary storage space exists independent from the reserving units, and administering the shared temporary storage space through tracked utilization of the reserved units, wherein coordinator and secondary nodes of the SDC maintain bitmaps which track the utilization of the reserved units, and each bit in the bitmaps represents a logical disk block and tracks whether the logical disk block is free or in use, and wherein the secondary₁₃ nodes periodically send content of the bitmaps associated with the secondary nodes to the coordinator node.

8. The system of claim 7 wherein the updates to bitmaps include space allocation activity for chains of reserved units.

9. The system of claim 7 wherein the updates to the bitmaps include timed expiration periods.

10. The system of claim 7 wherein administering further comprises utilizing a garbage collection event.

11. The system of claim 10 wherein the garbage collection event handles expired reserved units and deallocated temporary storage space for reuse in the SDC.

12. A non-transitory computer-usable medium having instructions recorded thereon that, if executed by a computing device, cause the computing device to perform a method comprising:

reserving units on-demand and of variable size from shared temporary storage space in a shared disk cluster (SDC), wherein the reserving controls a size of the reserved units based upon configuration of the SDC, remaining space of the shared temporary storage space, and a number of configured nodes, and wherein the shared temporary storage space exists independent from the reserving units;

tracking utilization of the reserved units of the shared temporary storage space, wherein coordinator and secondary nodes of the SDC maintain bitmaps which track the utilization of the reserved units, and each bit in the bitmaps represents a logical disk block and tracks whether the logical disk block is free or in use, and wherein the secondary nodes periodically send content of the bitmaps associated with the secondary nodes to the coordinator node; and

administering the shared temporary storage space based on the tracking.

13. The computer-usable medium of claim 12 wherein tracking further comprises updating the bitmaps with respect to space allocation activity for chains of reserved units.

14. The computer-usable medium of claim 12 wherein tracking further comprises updating the bitmaps with respect to timed expiration periods.

15. The computer-usable medium of claim 12 wherein administering further comprises utilizing a garbage collection event to handle expired reserved units and deallocated temporary storage space for reuse in the SDC.

16. A method comprising:

receiving an initial request, from a first node of a plurality of nodes in a shared disk database cluster (SDC) of nodes performing distributed processing of a query in the SDC, to reserve a portion of a shared temporary storage space, that exists independent from the plurality of nodes, wherein each of the plurality of nodes is allocated at least a portion of the shared temporary storage space;

determining whether the initial request is less than a max request size; and

providing a first discrete reservation unit to the first node based on the initial request, wherein the discrete reservation unit is of a size that is the lesser of the initial request of the max request size, wherein a second node in the cluster is allocated a second discrete reservation unit of a size different than the size of the first discrete reservation unit.

17. The method of claim 16, wherein the initial request is a request for a percentage of the portion of the shared temporary space allocated to the first node.

18. The method of claim 17, further comprising:

receiving a subsequent request from the first node, wherein the subsequent request is for a percentage of remaining free space in the shared temporary space; and

providing a subsequent discrete reservation unit to the first node, wherein a size of the subsequent discrete reservation unit is determined based on the remaining free space, the number of nodes in the cluster, and any current reservation units provided to the first node.