APPARATUS AND METHOD OF MANAGING DATABASES OF ACTIVE NODE AND STANDBY NODE OF MAIN MEMORY DATABASE MANAGEMENT SYSTEM

• US 20130151895A1
• Filed: 04/26/2012
• Published: 06/13/2013
• Est. Priority Date: 12/09/2011
• Status: Active Grant

• Current Assignee: Altibase Corp.
• Original Assignee: Altibase Corp.
• Inventors: Lee, Seung Won
• Granted Patent: US 8,862,936 B2
• Time in Patent Office: Days
• US Class Current: 714/19
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: G06F 11/1658 Data re-synchronization of a redundant component, or initial sync of replacement, additional or spare unit
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups
• CPC Class Codes: G06F 2201/855 Details of asynchronous mirroring using a journal to transfer not-yet-mirrored changes

Abstract

Databases of an active node and a standby node of a main memory database management system (MMDBMS) are managed so as to prevent loss of a transaction caused by failure of any one of the active node or the standby node. The MMDBMS is configured to prevent data mismatch between the active node and the standby node when failure of any one of the active node and the standby node occurs. In case of failure of one of the nodes, log information from the other node is obtained to recover the failed node.

Claims

1. An apparatus for managing a database of an active node of a main memory database management system (MMDBMS) that is replicated from the active node to a standby node, the apparatus comprising:

a memory logging unit that records a database log to which a sequence number is sequentially allocated in a log buffer of a memory of the active node;

a replicating unit that replicates the database log to the standby node;

a disk flushing unit that periodically flushes the database log to a disk of the active node; and

a recovering unit that, when failure of the active node occurs;

transmits a sequence number of a last recovered log that is finally recovered to the standby node,

receives at least one replication log belonging to a transaction comprising a lost log recorded after the last recovered log from the standby node, and

recovers the database of the active node based on the at least one replication log.

2. The apparatus of claim 1, wherein the storing of the database log is performed independently from committing of a transaction comprising the database log and replicating of the database log.

3. The apparatus of claim 1, wherein the sequence number of the database log is allocated to sequentially increase, and the lost log is a database log having a sequence number greater than a sequence number of the last recovered log.

4. The apparatus of claim 1, further comprising a recovery information managing unit that allows a sequence number of a database log corresponding to the replication log flushed to the disk of the standby node to be periodically reported thereto from the standby node, and responsive to failure of the standby node, recovers a database of the standby node by replicating database logs recorded after a database log having a sequence number that is finally received to the standby node.

5. A computer-implemented method of managing a database of an active node of a main memory database management system (MMDBMS) that is replicated from the active node to a standby node, the method comprising:

recording a database log to which a sequence number is sequentially allocated in a log buffer of a memory of the active node;

replicating the database log to the standby node;

periodically storing the database log in a disk of the active node; and

receiving, from the standby node, a sequence number of a database log corresponding to a replication log flushed to a disk of the standby node.

6. The method of claim 5, further comprising:

responsive to a failure of the active node, transmitting to the standby node, a sequence number of a last recovered log that is finally recovered;

receiving, from the standby node, at least one replication log belonging to a transaction comprising a lost log recorded after the last recovered log; and

recovering the database of the active node based on the at least one replication log.

7. The method of claim 6, wherein the sequence number of the database log is allocated to sequentially increase,

wherein the lost log is a database log having a sequence number greater than a sequence number of the last recovered log.

8. The method of claim 5, further comprising:

responsive to a failure of the standby node, replicating database logs recorded after a database log having a sequence number that is finally received to the standby node.

9. A non-transitory computer readable storage medium storing a computer program product including computer instructions configured to cause a processor of a computer to manage a database of an active node of a main memory database management system (MMDBMS) that is replicated from the active node to a standby node, the computer program product comprising:

a memory logging unit configured to record a database log to which a sequence number is sequentially allocated in a log buffer of a memory of the active node;

a replicating unit configured to replicate the database log to the standby node;

a disk flushing unit configured to periodically flush the database log to a disk of the active node; and

a recovering unit configured to;

responsive to failure of the active node, transmit a sequence number of a last recovered log that is finally recovered to the standby node;

receive, from the standby node, at least one replication log belonging to a transaction comprising a lost log recorded after the last recovered log; and

recover the database of the active node based on the at least one replication log.

10. The non-transitory computer readable storage medium of claim 9, wherein the sequence number of the database log is allocated to sequentially increase, and the lost log is a database log having a sequence number greater than a sequence number of the last recovered log.

11. The non-transitory computer readable storage medium of claim 9, wherein the computer program product further comprises:

a recovery information managing unit configured to;

receive periodically, from a standby node, a sequence number of a database log corresponding to the replication log flushed to the disk of the standby node, and

responsive to failure of the standby node, recover a database of the standby node by replicating to the standby node, database logs recorded after a database log having a sequence number that is finally received.

12. An apparatus for managing a database of a standby node of a main memory database management system (MMDBMS) that is replicated from an active node to the standby node, the apparatus comprising:

a memory logging unit that allows a database log to which a sequence number is sequentially allocated to be replicated thereto from the active node, and records changes to the database which occur during a transaction performed based on the database log as a replication log to a log buffer of a memory of the standby node;

a disk flushing unit that periodically flushes the replication log to a disk of the standby node;

a log flush reporting unit that periodically reports a sequence number of a database log corresponding to the flushed replication log to the active node; and

a recovering unit that, when failure of the standby node occurs;

receives a recovery-replication log obtained by replicating a database log after a database log having a sequence number that is finally reported from the active node, and

recovers the database of the standby node based on the replication log for recovery.

13. The apparatus of claim 12, wherein the memory logging unit maps a sequence number of the replication log to a sequence number of a database log corresponding to the replication log to obtain a result and stores the result in a transaction table in units of transaction.

14. The apparatus of claim 13, further comprising:

a log searching unit that, when failure of the active node occurs, receives a sequence number of a last recovered log that is finally recovered from the active node and searches for all replication logs belonging to a transaction comprising a lost log after the last recovered log; and

a replication log transmitting unit that transmits the searched replication logs to the active node.

15. A computer-implemented method of managing a database of a standby node of a main memory database management system (MMDBMS) that is replicated from an active node to the standby node, the method comprising:

allowing a database log to which a sequence number is sequentially allocated to be replicated from the active node;

recording changes to the database which occur during a transaction performed based on the database log as a replication log to a log buffer of a memory of the standby node;

storing the replication log and a sequence number of a database log corresponding to the replication log in a transaction table in units of transaction;

periodically flushing the replication log to a disk of the standby node; and

periodically reporting a sequence number of a database log corresponding to the replication log flushed to the disk to the active node.

16. The method of claim 15, further comprising:

when failure of the active node occurs, receiving a sequence number of a last recovered log that is finally recovered from the active node;

searching for all replication logs belonging to a transaction comprising a lost log after the last recovered log based on the transaction table; and

transmitting the searched replication logs to the active node.

17. The method of claim 15, further comprising:

when failure of the standby mode occurs, receiving a recovery-replication log obtained by replicating a database log after a database log having a sequence number that is finally reported from the active node; and

recovering the database of the standby node based on the replication log for recovery.

18. A non-transitory computer readable storage medium storing a computer program product including computer instructions configured to cause a processor of a computer to manage a database of a standby node of a main memory database management system (MMDBMS) that is replicated from an active node to the standby node, the computer program product comprising:

a memory logging unit configured to allow a database log to which a sequence number is sequentially allocated to be replicated thereto from the active node, and record changes to the database which occur during a transaction performed based on the database log as a replication log to a log buffer of a memory of the standby node;

a disk flushing unit configured to periodically flush the replication log to a disk of the standby node;

a log flush reporting unit configured to periodically report a sequence number of a database log corresponding to the flushed replication log to the active node; and

a recovering unit that is configured to;

responsive to a failure of the standby node, receive a recovery-replication log obtained by replicating a database log after a database log having a sequence number that is finally reported from the active node, and

recover the database of the standby node based on the replication log for recovery.

19. The non-transitory computer readable storage medium of claim 18, wherein the memory logging unit maps a sequence number of the replication log to a sequence number of a database log corresponding to the replication log to obtain a result and stores the result in a transaction table in units of transaction.

20. The non-transitory computer readable storage medium of claim 18, wherein the computer program product further comprises:

a log searching unit configured to;

responsive to a failure of the active node, receive a sequence number of a last recovered log that is finally recovered from the active node and search for all replication logs belonging to a transaction comprising a lost log after the last recovered log; and

a replication log transmitting unit configured to transmit the searched replication logs to the active node.