NON-DISRUPTIVE DATA MOVEMENT AND NODE REBALANCING IN EXTREME OLTP ENVIRONMENTS

• US 20120158805A1
• Filed: 12/16/2010
• Published: 06/21/2012
• Est. Priority Date: 12/16/2010
• Status: Active Grant

• Current Assignee: Sybase Incorporated (SAP SE)
• Original Assignee: Sybase Incorporated (SAP SE)
• Inventors: Cheng, Xun, Wang, Yanhong, Shang, Heping, Shepherd, Stephen C., ANDRE, Mihnea, Wildermuth, Elton Philip
• Granted Patent: US 9,075,858 B2
• Time in Patent Office: Days
• US Class Current: 707/827
• CPC Class Codes: G06F 16/24552 Database cache management
• CPC Class Codes: G06F 16/278 Data partitioning, e.g. horizontal or vertical partitioning

Abstract

Systems and methods for non-disruptive cache startup, shutdown, data partition movement, and node rebalancing for a data fabric within a data grid in a database environment. The data grid is a clustered in-memory database (IMDB) cache comprising one or more data fabrics including multiple IMDB cache nodes. The method takes a time tag and moves data from source to destination nodes while allowing updates to data on the source nodes. The method propagates updates occurring after the time tag to destination nodes from source nodes and switches applications and ownership to the destination nodes. Another method slices data for a partitioned fabric and stores slices on nodes. Another method starts a data fabric by moving data from a grid backend to the nodes. In an embodiment, another method rebalances nodes in a non-disruptive manner. When a fabric is shutdown, the method saves data from the nodes to the grid backend.

Claims

1. A method for non-disruptive movement of data partitions in a data grid within a database environment, the method comprising:

creating a time tag indicating the state of data at a point in time within a plurality of in-memory database (IMDB) system nodes organized as a data fabric of the data grid;

moving data pages from a plurality of source nodes to a plurality of destination nodes, wherein changes by applications to data pages on the plurality of source nodes are allowed during the moving;

propagating data changes from the plurality of source nodes to the plurality of destination nodes;

freezing data page changes on the plurality of source nodes;

instructing applications to perform future data page changes on the plurality of destination nodes; and

switching data page ownership to the plurality of destination nodes.

2. The method of claim 1, wherein the freezing further comprises:

allowing current application transactions against the plurality of source nodes to complete;

committing the allowed transactions on the plurality of source nodes;

blocking new application transactions targeting the plurality of source nodes; and

propagating data changes resulting from the committed transactions to the plurality of destination nodes.

3. The method of claim 1, wherein the propagating further comprises:

propagating data changes occurring after the time tag from the plurality of source nodes to the plurality of destination nodes.

4. The method of claim 1, wherein creating the time tag comprises:

creating a marker in log records of the plurality of IMDB system nodes indicating the time tag; and

recording the marker.

5. The method of claim 1, further comprising, prior to the freezing:

determining if there are additional changes to data pages are to be propagated; and

upon determining that there are additional changes to data pages to be propagated, repeating the propagating.

6. The method of claim 1, wherein the determining comprises determining if an amount of additional changes to data pages to be propagated is below a pre-determined threshold.

7. The method of claim 6, further comprising:

upon determining that the amount of additional changes to data pages to be propagated is greater than the pre-determined threshold, repeating the propagating.

8. The method of claim 6, wherein the pre-determined threshold is a tunable value.

9. The method of claim 8, wherein the tunable value is a number of data manipulation language (DML) statements to be executed on the plurality of destination nodes.

10. The method of claim 5, further comprising:

determining if a current amount of additional changes to data pages to be propagated is greater than or equal to a prior amount of additional changes;

upon determining that there the current amount is less than the prior amount, repeating the propagating;

upon determining that there the current amount is greater than or equal to the prior amount;

reducing access by the applications to the plurality of source nodes; and

repeating the propagating.

11. A non-transitory computer readable storage medium having instructions stored thereon that, in response to execution by a computing device, cause the computing device to perform operations for performing a non-disruptive cache shutdown in a database environment, the operations comprising:

receiving a command to shutdown a data fabric within a data grid in the database environment;

creating a time tag indicating the state of data at a point in time within a plurality of in-memory database (IMDB) system nodes organized as the data fabric;

moving data pages from the plurality of IMDB system nodes to a disk resident database (DRDB), wherein changes by applications to data pages on the plurality of IMDB system nodes are allowed during the moving;

propagating data changes from the plurality of IMDB system nodes to the DRDB;

freezing data page changes on the plurality of IMDB system nodes; and

shutting down the data fabric.

12. The computer readable storage medium of claim 11, wherein the DRDB is hosted on a grid backend server.

13. A computer program product comprising a non-transitory computer readable storage medium having computer program logic stored thereon that, in response to execution by a processor, causes the processor to perform a non-disruptive cache startup in a database environment by performing operations comprising:

receiving a command to startup a data fabric within a data grid in the database environment;

reading metadata within the data grid indicating data partitions associated with the plurality of IMDB system nodes;

moving data pages from a disk resident database (DRDB) to the plurality of IMDB system nodes, wherein changes by applications to data pages on the DRDB are allowed during the moving;

propagating data changes from the DRDB to the plurality of IMDB system nodes;

freezing data page changes on the DRDB; and

instructing applications to perform future data page changes on the plurality of IMDB system nodes; and

switching data page ownership to the plurality of IMDB system nodes.

14. The computer program product of claim 13, further comprising, prior to the reading:

retrieving a time tag indicating the state of data at a point in time within a plurality of in-memory database (IMDB) system nodes organized as the data fabric;

wherein the reading is based on the time tag.

15. The computer program product of claim 14, wherein the point in time is the time of a prior data fabric shutdown.

16. A system for rebalancing data fabric nodes within a data grid in a database environment, the system comprising:

a data fabric comprising at least one first in-memory database (IMDB) system node and at least one second IMDB system node, the at least one first IMDB system node and the at least one second IMDB system node organized as one or more data fabrics for storing data according to one or more data granularity types;

a load determining module configured to determine load values for the at least one first IMDB system node and the at least one second IMDB system node, wherein the load determining module is further configured to identify, based on the load values, at least one more-loaded IMDB system node and at least one less-loaded IMDB system node and;

a rebalancing module configured to move data slices from the at least one more-loaded IMDB system node of the at least one first IMDB system node and the at least one second IMDB system node to the at least one less-loaded IMDB system node to better distribute a data or processing load within the data fabric, wherein the rebalancing module is further configured to allow applications to update data pages on the at least one more-loaded IMDB system node during the moving;

an update catch-up module configured to propagate data changes occurring during the moving tag from the at least one more-loaded IMDB system node to the at least one less-loaded IMDB system node; and

a switching module configured to;

instruct applications to perform future data page changes on at least one less-loaded IMDB system node; and

switch data page ownership to the at least one less-loaded IMDB system node.

17. The system of claim 16, wherein the load values of the IMDB system nodes are determined as a relative load in relation to other IMDB system nodes in the data fabric.

18. The system of claim 16, wherein the load values of the IMDB system nodes are determined as a relative processing load as compared to other IMDB system nodes in the data fabric.

19. The system of claim 16, wherein the load values of the IMDB system nodes are determined as a relative data load as compared to other IMDB system nodes in the data fabric.

20. The system of claim 16, wherein the load values of the IMDB system nodes are determined based on an absolute processing load.

21. The system of claim 20, wherein the absolute processing load is measured as a percentage of central processing unit (CPU) capacity available on IMDB system nodes in the data fabric.

22. The system of claim 16, wherein the load vales of the IMDB system nodes are determined based on an absolute data storage load measured as a percentage of data storage capacity available on IMDB system nodes in the data fabric.