Meta Model Driven Data Base Replication and Synchronization

• US 20150142734A1
• Filed: 07/13/2012
• Published: 05/21/2015
• Est. Priority Date: 05/24/2012
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Mueller, Elisabeth
• Granted Patent: US 10,089,374 B2
• Time in Patent Office: Days
• US Class Current: 707/610
• CPC Class Codes: G06F 16/254 Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
• CPC Class Codes: G06F 16/256 in federated or virtual databases
• CPC Class Codes: G06F 16/258 Data format conversion from or to a database
• CPC Class Codes: G06F 16/27 Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor

Abstract

The present invention relates to a method of executing an export process for an automated meta model driven export of entity parts from a source data base (10), wherein the meta model specifies an abstract model of the source data base (10) and comprises a description of an entity part and/or a reference relationship between entity parts being resolvable from a referencing entity part to a referenced entity part. The method comprising the steps identifying (S14) a set of entity parts being identifiable outside the source data base (10) by a public key, an internal key, or a mixture of a public key and an internal key as first level entity parts, supplementing (S16) the set of first level entity parts by a set of second level entity parts being referenced by the first level parts according to at least one reference relationship described in the meta model, and writing (S18) export data for the set of first level parts and the set of second level parts into an export file. The present invention also relates to a method of executing an import process for an automated meta model driven import of entity parts from a source data base (10) to a target data base (12).

Claims

1. A method of executing an export process for an automated meta model driven export of entity parts from a source data base, wherein the meta model specifies an abstract model of the source data base and comprises a description of an entity part and/or a reference relationship between entity parts being resolvable from a referencing entity part to a referenced entity part, the method comprising:

identifying a set of entity parts being identifiable outside the source data base by a public key, an internal key, or a mixture of a public key and an internal key as first level entity parts;

supplementing the set of first level entity parts by a set of second level entity parts being referenced by the first level parts according to at least one reference relationship described in the meta model; and

writing export data for the set of first level parts and the set of second level parts into an export file.

2. The method of claim 1, further comprising a step of defining the export process as a task file storing a list of identifiers of first level entity parts and a reference to a meta model file describing the meta model which is applicable to the export process.

3. The method of claim 2, further comprising the steps of:

initializing an empty repository as a staging memory for the export process;

retrieving the identified first level entity parts in sequence from the source data base and adding every retrieved first level entity part to the repository; and

linking every added first level entity part to its meta model description.

4. The method of claim 3, further comprising the step of supplementing the repository by:

a step of scanning the first level entity parts added to the repository;

a step of resolving, for every scanned first level entity part, its reference relationships in a recursive manner to identify related child entity parts;

a step of evaluating, for every child entity part, whether it is already added to the repository; and

a step (S38, S40) of adding every child entity part to the repository when it is not yet added.

5. The method of claim 4, wherein the reference relationship in the meta model is selectable as a 1:

1 reference being resolvable from a referencing entity part of the source data base to one referenced entity part of the source data base and/or a 1;

n child reference being resolvable from a referencing entity part of the source target data base to at least one referenced entity parts of the source data base, and wherein

the step of resolving, for every scanned first level entity part, its reference relationships in a recursive manner to identify related child entity parts is executed such that a 1;

1 reference is resolved over one level of hierarchy and such that a 1;

n child reference is resolved over at least one level of hierarchy.

6. The method of claim 3, wherein the repository comprises a map of keys and related data for every entity part type and the step of adding an entity part to the repository is executed by

identifying an entity part type of the entity part;

identifying the map corresponding to the identified entity part type or creating a new map to cover the identified entity part type when it is added for the first time to the repository; and

entering the keys and related data into the identified or the newly created map.

7. The method of claim 6, wherein the entity part data of a referenced child entity part comprises at least a reference to a parent entity part referencing the child entity part.

8. The method of claim 4, further comprising the step of applying a filter reflecting entity part conditions prior to the addition of an entity part to the repository such that the entity part is added to the repository when its entity part conditions fit the filter.

9. The method of claim 4, further comprising the step of excluding at least one predetermined entity part from the export file and including only a public key to the entity part into the export file.

10. A method of executing an import process for an automated meta model driven import of entity parts from a source data base to a target data base, wherein the meta model specifies an abstract model of the source data base or the target data base and comprises at least a description of an entity part and/or a private child reference being resolvable from a parent entity part of the source data base or the target data base which is identifiable outside the source data base or the target data base by a public key, wherein the resolution is to at least one referenced entity part of the source data base or the target data base and is identifiable only internally in the source data base or the target data base, and wherein the private child reference is used to override a pre-existing aggregation of entity parts in the target data base being referenced from an entity part of the target data base having the same public key as the parent entity part in the source data base, the method comprising:

parsing an import file to build an initial graph of entity parts according to the meta model in a repository as staging memory for data import into the target data base;

transforming the initial graph in the repository into an import graph of entity parts, said transforming comprising;

i) identifying a next entity part in the initial graph being identifiable outside the source data base by a public key as next first level source entity part;

ii) searching the target data base for an entity part which has a public key corresponding to the public key of the next first level source entity part as next first level target entity part;

iii) deleting (S60, S62) entity parts being referenced by the next first level target entity part in the target data base when the referenced entity parts are identifiable internally in the target data base and a private child reference of the meta model indicates that such entity parts are to be deleted in the target data base prior to data import; and

iv) completing the next first level target entity part and its referenced entity parts with data from the target data base; and

updating the target data base on the basis of the import graph of entity parts.

11. The method of claim 10, wherein

the meta model further specifies a 1:

1 reference being resolvable from a referencing entity part of the source data base or the target data base to one referenced entity part of the source data base or the target data base, and/or a 1;

n child reference being resolvable from a referencing entity part of the source data base or the target data base to at least two referenced entity part of the source data base or the target data base, and

references are differentiated as public references or transient references being changeable upon import into the target data base, wherein

the step of completing the next first level target entity part and its referenced entity parts comprises the steps of;

resolving public references of the next first level target entity part from the repository or from the target data base;

resolving 1;

1 references of the next first level target entity part; and

proceeding recursively with the resolution of public references and 1;

1 references from the next first level target entity part to its referenced entity parts.

12. The method of claim 11, wherein the step of completing the next first level target entity part and its referenced entity parts comprises the steps of:

propagating references being identifiable internally in the target data base and corresponding to transient references parsed from the import file from the target data base into the import graph.

13. The method of claim 10, further comprising a step of creating an import sequence for entity parts in the import graph in accordance with types of references according to a rule that child entity parts follow parent entity parts.

14. The method of claim 13, further comprising the step of reordering the import sequence such that entity parts which are referenced are listed prior to entity parts referencing entity parts.

15. The method of claim 10, wherein the step of transforming the initial graph further comprises the step of creating internal keys for entity parts to be newly imported into the target data base.

16. The method of claim 10, wherein each entity part in the initial graph is characterized by entity type, at least one attribute, and a list of references to other entity parts or children and the step of completing the initial graph further comprises the step of creating transient attributes for entity parts being newly inserted to the target data base and filling the transient attributes according to rules specified in the meta model.

17. The method of claim 10, wherein the step of updating the target data base on the basis of the import graph comprises the steps of creating database statements modeling the import graph for subsequent storage thereof in the target data base.

18. An apparatus for executing an export process according to an automated meta model driven export of entity parts from a source data base, comprising:

a meta model administration unit adapted to manage a meta model specifying an abstract model of the source data base and comprising a description of an entity part and/or a reference relationship between entity parts being resolvable from a referencing entity part to a referenced entity part;

an entity part processing unit adapted to identify a set of entity parts being identifiable outside the source data base by a public key, an internal key, or a mixed public and internal key as first level entity parts;

a database modeling unit adapted to supplement the set of first level entity parts by a set of second level entity parts being referenced by the first level parts according to at least one reference relationship described in the meta model; and

an export file generation unit adapted to write export data for the set of first level parts and the set of second level parts into an export file.

19. The apparatus of claim 18, further comprising a task file reading unit adapted to read a task file defining the export process through a list of identifiers of entity parts and through a reference to a meta model file describing the meta model which is applicable to the export process.

20. The apparatus of claim 19, wherein the data base modeling unit further comprises:

a repository memory; and

a repository memory control unit adapted to

initialize an empty repository as a staging memory for the export process;

retrieve the identified first level entity parts in sequence from the source data base and to add every retrieved first level entity part to the repository memory; and

to

link every added first level entity part to its meta model description.

21. The apparatus of claim 20, wherein the data base modeling unit is further adapted to supplement the repository memory with export data and comprises:

an entity part scanning unit adapted to scan the first level entity parts added to the repository memory;

a reference resolution unit adapted to resolve, for every scanned first level entity part, its reference relationships in a recursive manner to identify related child entity parts; and

an entity part addition unit adapted to evaluate, for every child entity part, whether it is already added to the repository memory and to add every child entity part to the repository memory when it is not yet added.

22. The apparatus of claim 21, wherein the reference relationship in the meta model is a a 1:

1 reference being resolvable from a referencing entity part of the source data base to one referenced entity part of the source data base and/or a 1;

n child reference being resolvable from a referencing entity part of the source data base to at least one referenced entity parts of the source data base, and wherein

the reference resolution unit is adapted to resolve, for every scanned first level entity part, its reference relationships in a recursive manner to identify related child entity parts such that a 1;

1 reference is resolved over one level of hierarchy and such that a 1;

n child reference is resolved over at least one level of hierarchy.

23. The apparatus of claim 20, wherein the repository memory stores a map of keys and related data for every entity part type and the entity part addition unit is configured to

identify an entity part type of the entity part;

identify the map corresponding to the identified entity part type or to create a new map to cover the identified entity part type when it is added for the first time to the repository memory; and

to

enter the keys and related data into the identified or the newly created map.

24. The apparatus of claim 22, wherein the repository memory control unit further comprises a filter unit adapted to apply a filter reflecting entity part conditions prior to the addition of an entity part to the repository memory such that the entity part is added to the repository memory when its entity part conditions fit the filter.

25. The apparatus of claim 22, wherein the data base modeling unit further comprises a compression unit configured to exclude at least one predetermined entity part from the export file and including only a public key to the entity part into the export file.

26. An apparatus for executing an import process for an automated meta model driven import of entity parts from a source data base to a target data base, wherein the meta model specifies an abstract model of the source data base or the target data base and comprises a description of an entity part and/or a private child reference being resolvable from a parent entity part of the source data base or the target data base which is identifiable outside the source data base or the target data base by a public key, wherein the resolution is to at least one referenced entity part of the source data base or the target data base and is identifiable internally in the source data base or the target data base, and wherein the private child reference is used to override a pre-existing aggregation of entity parts in the target data base being referenced from an entity part of the target data base having the same public key as the parent entity part in the source data base, the apparatus comprising:

an import file parsing unit adapted to parse an import file to build an initial graph of entity parts according to the meta model in a repository as staging memory for data import into the target data base;

a graph transformation unit adapted to transform the initial graph in the repository into an import graph of entity parts by performing the steps of;

i) identifying a next entity part in the initial graph being identifiable outside the source data base by a public key as next first level source entity part;

ii) searching the target data base for an entity part which has a public key corresponding to the public key of the next first level source entity part as next first level target entity part;

iii) deleting entity parts being referenced by the next first level target entity part in the target data base when the referenced entity parts are identifiable internally in the target data base and a private child reference of the meta model indicates that such entity parts are to be deleted in the target data base prior to data import;

iv) completing the next first level target entity part and its referenced entity parts with data from the target data base; and

a data base updating unit adapted to update the target data base on the basis of the import graph of entity parts.

27. The apparatus of claim 26, wherein

the meta model further specifies a 1:

1 reference being resolvable from a referencing entity part of the source data base or target data base to one referenced entity part of the source data base or the target data base, and/or a 1;

n child reference being resolvable from a referencing entity part of the source data base or the target data base to at least one referenced entity part of the source data base or the target data base, and

references are differentiated as public references or transient references being changeable upon import into the target data base, wherein

the graph transformation unit adapted to complete the next first level target entity part and its referenced entity parts comprises a reference resolution unit adapted to

resolve public references of the next first level target entity part from the repository or from the target data base;

resolve 1;

1 references of the next first level target entity part; and

to

proceed recursively with the resolution of public references and 1;

1 references from the next first level target entity part to its referenced entity parts.

28. The apparatus of claim 27, wherein the graph transformation unit adapted to complete the next first level target entity part and its referenced entity parts comprises a back propagation unit adapted to propagate references being identifiable only internally in the target data base and corresponding to transient references parsed from the import file from the target data base into the import graph.

29. The apparatus of claim 26, wherein the graph transformation unit further comprises an input sequence generation unit adapted to create an import sequence for entity parts in the import graph in accordance with types of references according to a rule that child entity parts follow parent entity parts.

30. The apparatus of claim 29, wherein the input sequence generation unit is adapted to reorder the import sequence such that entity parts which are referenced are listed prior to entity parts referencing entity parts.

31. The apparatus of claim 26, wherein the graph transformation unit comprises a data transformation unit adapted to create internal keys for entity parts to be newly imported into the target data base.

32. The apparatus of claim 31, wherein each entity part in the initial graph is characterized by entity type, at least one attribute, and a list of references to other entity part or children and the data transformation unit is adapted to create transient attributes for entity parts being newly inserted to the target data base and to fill the transient attributes according to rules specified in the meta model.

33. The apparatus of claim 26, wherein the data transformation unit is adapted to create database statements modeling the import graph for subsequent storage thereof in the target data base.