METHOD AND SYSTEM FOR CREATION AND DYNAMIC UPDATING OF BEST DATA ARRANGEMENT IN DIGITAL DATA STORE SYSTEM

• US 20090327217A1
• Filed: 06/24/2009
• Published: 12/31/2009
• Est. Priority Date: 06/26/2008
• Status: Active Grant

• Current Assignee: Tamiras Per Pte Ltd LLC (Intellectual Ventures LLC)
• Original Assignee: HyperRoll Israel Limited (Oracle Corporation)
• Inventors: Golan, Shahar, Belyaev, Victor, Sadetsky, Marina, Shaked, Guy, Braverman, Vladimir, Matias, Yossi
• Granted Patent: US 8,364,697 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/245 Query processing

Abstract

The current invention is a method and a system for analyzing templates of queries, with different levels of precision. When the system of the current invention takes a decision to arrange data, these templates are used, and a single data arrangement object (HRBlob) is designed which may serve a single template or several templates and not just a single query. This approach has an important impact on data arrangement performance and on queries performance, since this approach may help many more queries to be handled using less data arrangement objects.

Claims

1. A system comprising two modules, which perform automation procedures of the data store arrangement and queries treatment in a data store system;

wherein said automation procedures are based on creation and continuous updating of a hierarchical set of HRBlobs and said two modules are;

a. an Analytical Dynamic Mechanism (ADM) module, which comprises components adapted to be able to decide how to act on each query, to collect the statistics from both said complementary system and said existing system, to manage the configuration of said complementary system, to update said hierarchical set of HRBlobs, and to process instructions from the supervisor;

said components of said ADM comprising;

a Query Analyzer;

a Statistics Collector; and

a Decision Maker; and

optionally comprises one or both of the following components;

a Queries Cache; and

a Query Router;

b. a Data Arrangements Module (DAM), which comprises components adapted to be able to handle the data in the Data Store of said existing data store system and said HRBlob set, to load data from said Data Store into said complementary system, to perform the configuration and the updates of said HRBlobs according to instructions received from the ADM, to extract the relevant HRBlob to provide the response to said query, and to provide query execution statistics;

said components of said DAM comprising;

a Control Module;

a Query Handler;

a HRBlobs Storage component;

a HRBlobs Handler component;

a HRBlobs Arrangement Engine component;

a Base Data Loader component; and

an Execution Statistics Provider component.

2. A system according to claim 1, wherein said system is a complementary system, which is added onto an existing data store system using the existing interface.

3. A system according to claim 1, wherein said system is an integrated system, which is integrated into a data store system.

4. A system according to claim 1, wherein, when said system is activated for the first time, the supervisor provides an initialization of the HRBlobs.

5. A system according to claim 1, wherein said system begins working without any prior knowledge of the data store contents and any existing summary tables.

6. A system according to claim 1, wherein each HRBlob consists of any subset of data store data records, with any subset of measures, and any subset and any combination of metadata dimensions, arranged in any order and in any level.

7. A system according to claim 1, wherein every HRBlob is designed to provide an effective answer to one or more specific query templates;

wherein each user query is composed of requested dimensions in a certain order and the required measures.

8. A system according to claim 2, wherein the supervisor can address instructions directly to the Data Arranger of the data store system, but the queries will be processed only through the Data Arranger Module of said system.

9. A system according to claim 1, wherein, if the Data Store is dynamic and can change both in its content and in its properties wherein, then said system will automatically recognize and adapt itself to said changes by updating the HRBlobs set.

10. A system according to claim 9, wherein the updates are executed online either prior to a query, or while processing a query, or after the query was executed.

11. A system according to claim 9, wherein the updates are executed offline.

12. A system according to claim 1, wherein:

the Query Analyzer analyses each query that is routed to the ADM, said analysis including content (syntactical) analysis and query properties for statistical collection;

the Statistics Collector collects and provides statistics on the content of the Data Store of the existing data store system and information on the dimensions and measures of each query and its environment properties, said Statistics Collector also collects statistics on each query execution, HRBlobs usage, and the Queries Cache usage;

the Decision Maker composes the initial system configuration, makes decisions concerning update of the system, and makes estimations of user benefit;

the Query Router decides for each given query whether to route the query to the DAM for answering the query using an existing HRBlob or to route the query to said Data Store to process the query in the same way as in said original data store system; and

said Queries Cache receives each query and decides if this query result should be cached for future use, wherein the decision is based on a caching policy, which is chosen according to user system limitations and can be automatically changed by said system according to usage effectiveness.

13. A system according to claim 1, wherein:

the Control Module initiates and controls most of the actions of the components of the DAM;

the Query Handler receives a query from the ADM and transfers it to the HRBlobs Handler;

said HRBlobs Handler either extracts a prepared in advance HRBlob from the HRBlob Storage component, or if a new HRBlob is needed, transfers the appropriate instructions to the Arrangement Engine;

said Arrangement Engine creates said needed new HRBlob according to instructions received from said Control Module, wherein said instructions are derived from the Initial System Configuration and/or from offline/online update instructions;

the Base Data Loader loads fact data and dictionary data from the Data Store of the existing data store system into said Arrangement Engine;

said HRBlobs Storage component stores HRBlobs created by said arrangement engine; and

the Execution Statistics Provider compiles and transfers statistics of query executions to the Statistics Collector in said ADM.

14. A system according to claim 1, wherein a query from a user is converted in the Support Mechanisms of the data store system to the syntax of said system according to claim 1 and then transferred to the ADM wherein, if there is an existing HRBlob to answer the query, the Query Router in said ADM will transfer said query to the DAM.

15. A system according to claim 14, wherein if there is no existing HRBlob to answer the query, a new HRBlob is created to answer the query.

16. A system according to claim 14, wherein if there is no existing HRBlob to answer the query, said query is routed to the Data Store of the data store system to be processed in the normal way.

17. A system according to claim 1, wherein, if there is no Query Router in the ADM or it is not being used, then the Control Module in the DAM receives instructions for the current query handling from the Decision Maker component in said ADM and accordingly addresses the relevant components in said DAM.

18. A system according to claim 1, wherein queries are processed online.

19. A system according to claim 1, wherein queries are processed offline.

20. A system according to claim 1, wherein an initial HRBlobs System configuration comprising a potential benefits report, an arrangement execution report, and a potential queries execution report is given to the supervisor without being physically implemented;

thereby allowing said supervisor to decide if the current HRBlobs configuration, either in its entirety or partially, is useful for his users or not.

21. A system according to claim 1, wherein queries are considered to be inherently nested queries and an algorithm in the ADM module looks at each incoming query and analyses each sub-query, at any level, of said incoming query individually as a separate entity.