Generating a schema of a not-only-structured-query-language database

• US 10,055,429 B2
• Filed: 04/28/2015
• Issued: 08/21/2018
• Est. Priority Date: 04/30/2014
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Jiao, Limei, Wang, Chen, Wang, Lanjun, Zhang, Shuo
• Primary Examiner(s): Alam, Shahid
• Application Number:
  US14/698,106
• Publication Number:
  US 20150317335A1
• Time in Patent Office: 1,211 Days
• Field of Search: 707803
• CPC Class Codes: G06F 16/211 Schema design and management
• CPC Class Codes: G06F 16/2246 Trees, e.g. B+trees
• CPC Class Codes: G06F 16/2379 Updates performed during online database operations; commit processing
• CPC Class Codes: G06F 16/273 Asynchronous replication or reconciliation

Abstract

Generation of a schema of a NoSQL type database, where a set of entry(ies) of a NoSQL type database can be determined, and the record schema is a treelike relationship between key hierarchies with a key as a node in the entry, wherein the key corresponds to an attribute name. For at least one record schema, a node in which there is an attribute name variation is determined based on a similarity of sub-nodes in the treelike relationship, and the record schema is updated according to the determined result. The schema of the NoSQL type database is generated based on the updated record schema.

Claims

1. A method comprising:

determining a record schema of a set of entry(ies) in a Not-only-Structured-Query-Language (NoSQL) type database, the record schema being a treelike relationship between key hierarchies with a key as a node in each entry of the set of entry(ies), wherein the key corresponds to an attribute name;

determining a node in which there is an attribute name variation based on a similarity of sub-nodes in the treelike relationship, and updating a corresponding record schema according to the determined result;

CinP-encoding the record schema of the each entry of the set of entry(ies);

parsing the encoded record schema in the tree structure into a path set, the path including the codes of all nodes which are traversed from a leaf node to a root node;

putting into buckets according to the P-Code of a certain tier;

in the buckets, determining a parent node in which there is an attribute name variation according to the similarity of the sub-nodes in the tier;

updating a source P-Code of part of the determined parent node in which there is an attribute name variation to a target P-Code, and updating the C-Code code of a parent node corresponding to the updated parent node; and

generating a schema for the NoSQL type database based on the updated record schema;

wherein;

the CinP-code of each node includes a code P-Code of the node itself and a code C-Code of the sub-nodes thereof; and

the determination of a node in which there is an attribute name variation includes determining a node in which there is an attribute name variation based on a similarity of sub-nodes in the CinP-encoded record schema.

2. The method according to claim 1, wherein the determination of a parent node in which there is an attribute name variation includes:

for the bucket in which a plurality of paths exist, determining a node whose parent nodes in the plurality of paths have different P-Code;

calculating the similarity of the parent nodes by the C-Code of the determined parent nodes and an equation

3. The method according to claim 2, wherein the generation of the schema of the NoSQL type database includes:

in the path code, extracting a sub-sequence containing only the P-Code, counting a frequency at which the sub-sequence occurs statistically, and merging the same P-Code sub-sequences; and

retaining the P-Code sub-sequence having a frequency exceeding a threshold value G2, and decoding the P-Code sub-sequence according to the code maps of the tiers to generate the schema of the NoSQL type database.

4. A device comprising:

a determiner configured to determine a record schema of a set of entry(ies) in a NoSQL (Not-only-Structured-Query-Language) type database, the record schema being a treelike relationship between key hierarchies with a key as a node in each entry of the set of entry(ies), wherein the key corresponds to an attribute name;

an updater configured to determine a node in which there is an attribute name variation based on a similarity of sub-nodes in the treelike relationship, and updating a corresponding record schema according to the determined result, the updater including;

a module configured to parse the encoded record schema in the tree structure into a path set, the path comprising the codes of all nodes which are traversed from a leaf node to a root node,

a module configured to put into buckets according to the P-Code of a certain tier,

a module configured to, in the buckets, determine a parent node in which there is an attribute name variation according to the similarity of the sub-nodes in the tier, and

a module configured to update a source P-Code of part of the determined parent node in which there is an attribute name variation to a target P-Code, and update the C-Code code of a parent node corresponding to the updated parent node; and

a generator configured to generate a schema of the NoSQL type database based on the updated record schema;

wherein;

the CinP-code of each node includes a code P-Code of the node itself and a code C-Code of the sub-nodes thereof; and

the determination of a node in which there is an attribute name variation includes determining a node in which there is an attribute name variation based on a similarity of sub-nodes in the CinP-encoded record schema.

5. The device according to claim 4, wherein the module configured to, in the buckets, determine a parent node in which there is an attribute name variation according to the similarity of the sub-nodes in the tier comprises:

a module configured to determine, for the bucket in which a plurality of paths exist, a node whose parent nodes in the plurality of paths have different P-Code;

a module configured to calculate the similarity of the parent nodes by the C-Code of the parent nodes and an equation

6. The device according to claim 5, wherein the module configured to generate a schema of the NoSQL type database according to the at least one updated record schema comprises:

a module configured to, in the path code, extract a sub-sequence containing only the P-Code, counting a frequency at which the sub-sequence occurs statistically, and merging the same P-Code sub-sequences; and

a module configured to retain the P-Code sub-sequence having a frequency exceeding a threshold value G2, and decoding the P-Code sub-sequence according to the code maps of the tiers to generate the schema of the NoSQL type database.

7. A computer program product comprising a computer readable storage medium having stored thereon:

first program instructions programmed to determine a record schema of a set of entry(ies) in a Not-only-Structured-Query-Language (NoSQL) type database, the record schema being a treelike relationship between key hierarchies with a key as a node in each entry of the set of entry(ies), wherein the key corresponds to an attribute name;

second program instructions programmed to determine a node in which there is an attribute name variation based on a similarity of sub-nodes in the treelike relationship, and updating a corresponding record schema according to the determined result, the second program instructions including;

third program instructions programmed to parse the encoded record schema in the tree structure into a path set, the path including the codes of all nodes which are traversed from a leaf node to a root node,

fourth program instructions programmed to put into buckets according to the P-Code of a certain tier,

fifth program instructions programmed to, in the buckets, determine a parent node in which there is an attribute name variation according to the similarity of the sub-nodes in the tier, and

sixth program instructions programmed to update a source P-Code of part of the determined parent node in which there is an attribute name variation to a target P-Code, and updating the C-Code code of a parent node corresponding to the updated parent node;

seventh program instructions programmed to generate a schema for the NoSQL type database based on the updated record schema;

eighth program instructions programmed to CinP-encode the record schema of each entry of the set of entry(ies);

wherein;

the CinP-code of each node includes a code P-Code of the node itself and a code C-Code of the sub-nodes thereof; and

second program instructions include ninth program instructions programmed to determine a node in which there is an attribute name variation based on a similarity of sub-nodes in the CinP-encoded record schema.

8. The product according to claim 7, wherein the second program instructions further include:

tenth program instructions programmed to, for the bucket in which a plurality of paths exist, determine a node whose parent nodes in the plurality of paths have different P-Code;

eleventh program instructions programmed to calculate the similarity of the parent nodes by the C-Code of the determined parent nodes and an equation

9. The product according to claim 8, wherein the generation of the third program instructions further include:

thirteenth program instructions programmed to, in the path code, extract a sub-sequence containing only the P-Code, counting a frequency at which the sub-sequence occurs statistically, and merging the same P-Code sub-sequences; and

fourteenth program instructions programmed to retain the P-Code sub-sequence having a frequency exceeding a threshold value G2, and decoding the P-Code sub-sequence according to the code maps of the tiers to generate the schema of the NoSQL type database.