Method and system for fast recovery of a primary store database using selective recovery by data type

• US 5,721,918 A
• Filed: 02/06/1996
• Issued: 02/24/1998
• Est. Priority Date: 02/06/1996
• Status: Expired due to Term

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Nilsson, Sven I., Todorovic, Zoran
• Primary Examiner(s): AMSBURY, WAYNE P
• Application Number:
  US08/597,342
• Time in Patent Office: 749 Days
• Field of Search: 395/618, 395/619
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1471 involving logging of persistent data for recovery
• CPC Class Codes: Y10S 707/99953 Recoverability

Abstract

The present invention provides an electronic data storage and processing system where non-persistent memory such as random access memory (RAM) stores a database with a first memory section storing semi-permanent data and a second memory section storing transient types of data. A third memory section in RAM may be used to buffer database transactions relating to the semi-permanent data stored in the first memory section of RAM. At periodic and appropriate checkpoint time intervals, the semi-permanent data currently stored in the first section of RAM are copied or "dumped" onto persistent (disk) memory. Only those database transactions that affect the semi-permanent data stored in the first section of RAM occurring after the most recent checkpoint "dump" are logged onto persistent memory. Database transactions affecting the transient type of data stored in the second portion of RAM are not logged. A recovery processor recovers from a system failure by reloading semi-permanent data is from the persistent memory into the first section of RAM and executing the log. However, in one embodiment, the recovery processor may leave the data in the second section of RAM in the state in which that data exists after the system failure. Considerable time is saved by not logging transient database transactions or executing a log for those transactions when recovering from a system failure.

Claims

1. An electronic data storage and processing system, comprising:

nonpersistent memory for storing a database including first and second sections corresponding to first and second types of data;

first persistent memory for storing a copy of data currently stored in the first section;

second persistent memory for logging database transactions; and

a data processor copying the first type of data stored in the first section into the first persistent memory at checkpoint time periods and logging in the second persistent memory only those database transactions relating to the first type of data stored in the first section that occur after a most recent checkpoint,

wherein transactions relating to the second type of data stored in the second portion are not logged.

2. The system in claim 1, wherein the data processor stores only data from the first section in the first persistent memory at each checkpoint.

3. The system in claim 1, wherein the data processor stores data from both the first and second sections in the first persistent memory at each checkpoint.

4. The system in claim 1, wherein the first type of data are semi-permanent data that usually change over a time period greater than the checkpoint time periods and the second type of data are transient data that change more frequently than the first type of data.

5. An electronic data storage and processing system, comprising:

nonpersistent memory for storing a database including first and second sections corresponding to first and second types of data;

first persistent memory for storing a copy of the first type of data stored in the first memory section;

a system manager copying data stored in the first section into the first persistent memory at checkpoint time periods; and

a recovery processor for recovering from a system failure by copying the first type of data from the first persistent memory into the first section of the nonpersistent memory,

wherein the recovery processor leaves the data in the second section of the nonpersistent memory in the state that data exists after the failure.

6. The system in claim 5, wherein the database manager determines the consistency of the second type of data stored in the second section, and if the second type of data stored in the second section is inconsistent, an application program using the second type of data in the database recovers consistent data for storage in the second section.

7. The system in claim 5, wherein the first persistent memory stores a copy of the second type of data stored in the second memory section and the database manager determines the consistency of the second type of data stored in the second section, and wherein if the data in the second memory section is inconsistent, the database manager copies the second type of data stored in the first persistent memory into the second section of nonpersistent memory.

8. The system in claim 5, wherein the non-persistent memory includes a third section for buffering recent database transactions relating to data in the first section and the database manager stores in the third section only those transactions relating to the first type of data stored in the first section that occur after a most recent checkpoint, with transactions relating to the second type of data stored in the second section not being buffered in the third section.

9. The system in claim 8, further comprising:

a second persistent memory for logging the buffered transactions, wherein the recovery processor recovers only the first type of data for the first section from the system failure using the logged transactions stored in the second persistent memory.

10. The system in claim 5, further comprising:

second persistent memory for logging only those transactions relating to the first type of data that occur after a most recent checkpoint, with transactions relating to the second type of data not being logged in the second persistent memory.

11. The system in claim 5, wherein the first type of data are semi-permanent data that usually changes only over a time period greater than the checkpoint time interval and the second type of data are transient data that changes more frequently than the first type of data.

12. A method for managing a data handling system that includes nonpersistent memory for storing a database and persistent memory, comprising the steps of:

storing database transactions relating to a first type of data in a first section of the non-persistent memory;

storing database transactions relating to a second type of data in a second section of the non-persistent memory;

copying the first type of data in the first section at predetermined checkpoint time intervals into the persistent memory;

storing in the persistent memory only those database transactions affecting the first type of data stored in the first section that occur after a recent checkpoint time interval,

wherein database transactions affecting the second type of data stored in the second section are not stored in the persistent memory.

13. The method in claim 12, wherein only the first type of data in the first section of the non-persistent memory are stored in the persistent memory at each checkpoint.

14. The method in claim 12, wherein data in both the first and second sections of the non-persistent memory are stored in the persistent memory at each checkpoint.

15. The method in claim 12, wherein the first type of data are semi-permanent data that usually change only over a time period greater than the checkpoint time intervals and the second type of data are transient data that change more frequently than the first type of data.

16. A method for recovering an electronic data storage and processing system from a system failure, comprising the steps of:

storing in nonpersistent memory a database having first and second sections corresponding to first and second types of data;

transferring the first type of data stored in the first section into a first persistent memory at checkpoint time periods; and

recovering from a system failure by loading the first type of data from the first persistent memory into the first section of the database after the failure and leaving the second type of data in the second section of the database in a state that second type of data exists after the failure.

17. The method in claim 16, farther comprising:

logging only database transactions relating to the first type of data in the first section, and

storing in a third section of the database only transactions relating to the first type of data stored in the first section that occur after a recent checkpoint and not storing in the third section transactions relating to the second type of data stored in the second section.

18. The method in claim 17, further comprising:

transferring from the third section into the persistent memory only transactions relating to the first type of data stored in the first section that occur after a most recent checkpoint.

19. The method in claim 18, further comprising:

recovering only the first type of data for the first section from the system failure using the transactions stored in the persistent memory.

20. The method in claim 16, wherein the first type of data are semi-permanent data that usually change only over a time period greater than the checkpoint time intervals and the second type of data are transient data that change more frequently than the first type of data.