Archive indexing engine

• US 8,051,045 B2
• Filed: 08/31/2005
• Issued: 11/01/2011
• Est. Priority Date: 08/31/2005
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP AG (SAP SE)
• Inventors: Vogler, Hartmut K.
• Primary Examiner(s): MORRISON, JAY A
• Application Number:
  US11/217,212
• Publication Number:
  US 20070050333A1
• Time in Patent Office: 2,253 Days
• Field of Search: None
• US Class Current: 707/661
• CPC Class Codes: G06F 16/22 Indexing; Data structures therefor; Storage structures

Abstract

Methods and apparatus, including computer program products, for archiving data from a database. One method includes identifying a data record to be archived; determining the contents of an archive record, the archive record having values for a first plurality of attributes in the data record; storing the archive record in a data archive; determining the contents of an index record, the index record comprising values for a second plurality of attributes in the data record; adding the index record to a dictionary-based archive index with a reference to the location of the archive record in the data archive; deleting the data record from the database; accepting a query for a desired archive record; and performing a search of the archive index to find the desired archive record.

Claims

1. A method comprising:

identifying a data record for deletion from a database and storage in a data archive, the data record comprising a plurality of data record attributes, each of the plurality of data record attributes comprising a value that comprises at least one term;

creating an archive record that comprises a first subset of attribute values of the plurality of data record attributes and an index record that comprises a second subset of attribute values of the plurality of data record attributes;

storing the archive record in a data archive that is stored separately from the database;

adding a reference to a location of the archive record in the data archive to the new index record;

adding the new index record to a dictionary-based archive index that is stored separately from the database, the dictionary-based archive index comprising a plurality of index records and a dictionary, the adding of the index record to the dictionary-based archive index comprising identifying every term of the second subset of attribute values of the plurality of data record attributes and adding each of the terms to the dictionary except for those terms that are already in the dictionary, wherein at least one index record of the plurality of index records in the dictionary-based archive index comprises one term from terms stored in the dictionary, references to locations of multiple archive records in the data archive that contain the one term, and information regarding locations of the one term within the referenced multiple archive records, the location information including indications, for each of the multiple of archive records referenced by the at least one index record, respective attributes of the each of the multiple archive records in which the one term is found;

deleting the data record from the database;

determining whether an attribute value of the plurality of attribute values is required for frequent user read access; and

if the attribute value is not required for frequent user read access, deleting the attribute value from the dictionary-based archive index.

2. The method of claim 1, further comprising:

accepting a query from a user for a desired archive record;

performing a search of the archive index to find the desired archive record; and

returning a reference to the location of the desired archive record in the data archive in response to the query.

3. The method of claim 1, further comprising:

accepting a query from a user for a desired archive record;

performing a search of the dictionary-based archive index to find the desired archive record; and

returning one or more of the values from the index record for the desired archive record in response to the query.

4. The method of claim 1, further comprising:

accepting a query from a user for a desired archive record;

performing a search of the archive index to find the desired archive record; and

returning one or more of the values from the desired archive record in response to the query.

5. The method of claim 1, wherein the second subset of attribute values is fully included within the first subset of attribute values.

6. The method of claim 1, wherein one or more of the first subset and the second subset of attribute values comprises values for all attributes in the data record.

7. The method of claim 1, wherein:

the second subset of attribute values at least includes the first subset of attribute values; and

accepting a query for a desired archive record comprises accepting a query to find the desired data record in the dictionary-based archive index based on the values for the second subset of attribute values;

the method further comprising;

accepting a query from a user for a desired archive record;

performing a search of the archive index to find the desired archive record; and

returning as a response to the query a responsive subset of the values for the second subset of attribute values for the desired data record.

8. A computer program product, encoded in an information carrier, operable to cause data processing apparatus to perform operations comprising:

identifying a data record to be archived, the data record comprising a plurality of data record attributes and originally residing in a database;

creating an archive record, the archive record comprising a first subset of the plurality of data record attributes, the first subset comprising at least some of the plurality of data record attributes;

storing the archive record in a data archive, the data archive being maintained separately from the database;

creating a new archive index record, the new archive index record comprising a reference to a location of the archive record in the data archive and a second plurality of attributes of the plurality of data record attributes, the second subset comprising selected attributes from the plurality of data record attributes, the selected attributes being those identified as necessary for access by users of the database;

adding the new archive index record to a dictionary-based archive index, the dictionary-based archive index comprising a plurality of archive index records, being stored separately from the database, and comprising a dictionary storing every term used in the plurality of index records;

deleting the data record from the database;

determining whether an attribute value of the plurality of attribute values is required for frequent user read access; and

if the attribute value is not required for frequent user read access, deleting the attribute value from the dictionary-based archive index.

9. The computer program product of claim 8, further operable to cause data processing apparatus to perform operations comprising:

accepting a query from a user for the data record, the query requesting at least one requested attribute not included in the second subset of the plurality of data record attributes;

performing a search of the archive index to find the location of the archive record based on the reference in the new archive index record; and

returning the reference to the location of the archive record in the data archive in response to the query.

10. The computer program product of claim 8, further operable to cause data processing apparatus to perform operations comprising:

accepting a query from a user for the data record, the query requesting one or more requested attributes that are part of the second subset of the plurality of data record attributes;

performing a search of the archive index to find the new archive index record; and

returning the one or more requested attributes of the data record from the new archive index record in response to the query without accessing the archive record in the data archive.

11. The computer program product of claim 8, further operable to cause data processing apparatus to perform operations comprising:

accepting a query from a user for the data record, the query requesting at least one requested attribute not included in the second subset of the plurality of data record attributes;

performing a search of the archive index to find the location of the archive record based on the reference in the new archive index record;

accessing the archive record from the data archive to retrieve the at least one attribute; and

returning the at least one requested attribute in response to the query.

12. The computer program product of claim 8, wherein:

the second plurality of attributes at least includes the first plurality of attributes.

21. The computer program product of claim 8, wherein the adding of the new archive index record to the dictionary-based archive index further comprises:

identifying each term in the second subset of the plurality of attribute values; and

adding each term to the dictionary if the term is not already in the dictionary, wherein at least one archive index record of the plurality of archive index records in the dictionary-based archive index comprises one term from a plurality of terms stored in the dictionary, references to locations of multiple archive records in the data archive that contain the one term, and information regarding locations of the one term within the referenced multiple archive records, and wherein the location information includes indications, for each one of the multiple of archive records referenced by the at least one archive index record, respective attributes of the one of the multiple archive records in which the one term is found.

13. A system comprising:

at least one processor; and

an information carrier storing instructions that, when executed by the at least one processor, cause the at least one processor to perform operations comprising;

identifying a data record to be archived, the data record comprising a plurality of data record attributes and originally residing in a database;

creating an archive record, the archive record comprising a first subset of the plurality of data record attributes, the first subset comprising at least some of the plurality of data record attributes;

storing the archive record in a data archive that is separate from the database;

creating a new archive index record, the new archive index record comprising a reference to a location of the archive record in the data archive and a second plurality of attribute of the plurality of data record attributes, the second subset comprising selected attributes from the plurality of data record attributes, the selected attributes being those identified as necessary for access by users of the database;

adding the new archive index record to a dictionary-based archive index, the dictionary-based archive index comprising a plurality of archive index records, being stored separately from the database, and comprising a dictionary storing every term used in the plurality of index records;

deleting the data record from the database; and

deleting, after a selected period, the values of all but a selected subset of the plurality of data record attributes from the dictionary-based archive index, the selected subset of the plurality of data record attributes comprising an index key to the archived record.

14. The system of claim 13, wherein the operations further comprise:

accepting a query from a user for the data record, the query requesting at least one requested attribute not included in the second subset of the plurality of data record attributes;

performing a search of the archive index to find the location of the archive record based on the reference in the new archive index record; and

returning the reference to the location of the archive record in the data archive in response to the query.

15. The system of claim 13, wherein the operations further comprise:

accepting a query from a user for the data record, the query requesting one or more requested attributes that are part of the second subset of the plurality of data record attributes;

performing a search of the archive index to find the new archive index record; and

returning the one or more requested attributes of the data record from the new archive index record in response to the query without accessing the archive record in the data archive.

16. The system of claim 13, wherein the operations further comprise:

accepting a query from a user for the data record, the query requesting at least one requested attribute not included in the second subset of the plurality of data record attributes;

performing a search of the archive index to find the location of the archive record based on the reference in the new archive index record;

accessing the archive record from the data archive to retrieve the at least one attribute; and

returning the at least one requested attribute in response to the query.

17. A computer program product, encoded in an information carrier, operable to cause data processing apparatus to perform operations comprising:

identifying a data record for deletion from a database and storage in a data archive, the data record comprising a plurality of data record attributes, each of the plurality of data record attributes comprising a value that comprises at least one term;

creating an archive record that comprises a first subset of attribute values of the plurality of data record attributes and an index record that comprises a second subset of attribute values of the plurality of data record attributes, the second subset including all of the data record attributes;

storing the archive record in a data archive that is stored separately from the database;

adding a reference to a location of the archive record in the data archive to the new index record;

adding the new index record to a dictionary-based archive index that is stored separately, from the database, the dictionary-based archive index comprising a plurality of index records and a dictionary, the adding of the index record to the dictionary-based archive index comprising identifying every term of the second subset of attribute values of the plurality of data record attributes and adding each of the terms to the dictionary except for those terms that are already in the dictionary, wherein at least one index record of the plurality of index records in the dictionary-based archive index comprises one term from terms stored in the dictionary, references to locations of multiple archive records in the data archive that contain the one term, and information regarding locations of the one term within the referenced multiple archive records, the location information including indications, for each of the multiple of archive records referenced by the at least one index record, respective attributes of the each of the multiple archive records in which the one term is found;

deleting the data record from the database; and

deleting, after a selected period, the values of all but a selected subset of the plurality of data record attributes from the dictionary-based archive index, the selected subset of the plurality of data record attributes comprising an index key to the archived record.

18. A method comprising:

identifying a data record to be archived, the data record comprising a plurality of data record attributes and originally residing in a database;

creating an archive record, the archive record comprising a first subset of the plurality of data record attributes, the first subset comprising at least some of the plurality of data record attributes;

storing the archive record in a data archive, the data archive being maintained separately from the database;

creating a new archive index record, the new archive index record comprising a reference to a location of the archive record in the data archive and a second plurality of attributes of the plurality of data record attributes, the second subset comprising selected attributes from the plurality of data record attributes, the selected attributes being those identified as necessary for access by users of the database;

adding the new archive index record to a dictionary-based archive index, the dictionary-based archive index comprising a plurality of archive index records, being stored separately from the database, and comprising a dictionary storing every term used in the plurality of index records;

deleting the data record from the database;

determining whether an attribute value of the plurality of attribute values is required for frequent user read access; and

if the attribute value is not required for frequent user read access, deleting the attribute value from the dictionary-based archive index.

19. The method of claim 18, wherein the determining whether the attribute value of the plurality of attribute values is required for frequent user read access comprises analyzing data collected from queries run by users on the dictionary-based archive index to determine a frequency with which attributes of the plurality of attributes are accessed.

20. A method comprising:

identifying a data record to be archived, the data record comprising a plurality of data record attributes and originally residing in a database;

creating an archive record, the archive record comprising a first subset of the plurality of data record attributes, the first subset comprising at least some of the plurality of data record attributes;

storing the archive record in a data archive, the data archive being maintained separately from the database;

creating a new archive index record, the new archive index record comprising a reference to a location of the archive record in the data archive and a second plurality of attributes of the plurality of data record attributes, the second subset comprising selected attributes from the plurality of data record attributes, the selected attributes being those identified as necessary for access by users of the database;

adding the new archive index record to a dictionary-based archive index, the dictionary-based archive index comprising a plurality of archive index records, being stored separately from the database, and comprising a dictionary storing every term used in the plurality of index records;

deleting the data record from the database; and

applying statistical techniques or artificial intelligence to determine which data record attributes are required for frequent user read access and should therefore remain in the database, and which data record attributes should be deleted because of lower user read access frequencies.

22. A computer program product, encoded in an information carrier, operable to cause data processing apparatus to perform operations comprising:

identifying a data record to be archived, the data record comprising a plurality of data record attributes and originally residing in a database;

creating an archive record, the archive record comprising a first subset of the plurality of data record attributes, the first subset comprising at least some of the plurality of data record attributes;

storing the archive record in a data archive, the data archive being maintained separately from the database;

creating a new archive index record, the new archive index record comprising a reference to a location of the archive record in the data archive and a second plurality of attributes of the plurality of data record attributes, the second subset comprising selected attributes from the plurality of data record attributes, the selected attributes being those identified as necessary for access by users of the database;

adding the new archive index record to a dictionary-based archive index, the dictionary-based archive index comprising a plurality of archive index records, being stored separately from the database, and comprising a dictionary storing every term used in the plurality of index records;

deleting the data record from the database; and

applying statistical techniques or artificial intelligence to determine which data record attributes are required for frequent user read access and should therefore remain in the database, and which data record attributes should be deleted because of lower user read access frequencies.