CN103729453A - HBase table conjunctive query optimization method - Google Patents
HBase table conjunctive query optimization method Download PDFInfo
- Publication number
- CN103729453A CN103729453A CN201410000720.8A CN201410000720A CN103729453A CN 103729453 A CN103729453 A CN 103729453A CN 201410000720 A CN201410000720 A CN 201410000720A CN 103729453 A CN103729453 A CN 103729453A
- Authority
- CN
- China
- Prior art keywords
- hive
- hbase
- query
- conjunctive
- conjunctive query
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2455—Query execution
- G06F16/24553—Query execution of query operations
- G06F16/24558—Binary matching operations
- G06F16/2456—Join operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2453—Query optimisation
- G06F16/24532—Query optimisation of parallel queries
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention provides an HBase table conjunctive query optimization method which includes the following steps: enabling an Hive to be combined with an HBase, achieving HBase table conjunctive querying through an HQL language provided by the Hive, then conducting optimization by setting parameters having influences on a bottom layer MapReduce task, and improving the conjunctive query performance. Compared with the prior art, the HBase table conjunctive query optimization method has the advantages that unnecessary programming trouble is reduced, parallel processing can be carried out on conjunctive query tasks, the query efficiency is improved, the practicability is high, and popularization is easy to achieve.
Description
Technical field
The present invention relates to computer information technology field, the method that a kind of HBase table conjunctive query is specifically optimized.
Background technology
Along with popularizing of large-scale internet application, the magnanimity of network information data increases severely, and large data have all produced tremendous influence to state treatment, business decision, personal lifestyle already.In the background of large data, under the epoch, distributed file system, distributed data base etc. is all the technology that is applicable to large data.HBase is different from general relational database, and it is a distributed data base that is suitable for unstructured data storage.Along with HBase continues to improve in performance and stability, HBase is adopted by a lot of major companies.HBase is the scalability distributed data base of a non-NoSQL of relation increasing income.It,, towards row, and is suitable for storing ultra-large type unstructured data.HBase can carry out read-write operation to large data at random.But in order and API that HBase itself provides, only have scan to be used for data query, the function of table conjunctive query is not provided, therefore, the method for a kind of HBase table conjunctive query is very important.
Hive is a kind of data warehouse architecture based on Hadoop.It provides uses simple class SQL query language HQL, developer can self-defined Mapper and Reducer process the complicated work that the built-in Mapper of Hive and Reducer cannot complete.Therefore, the conjunctive query of HBase table is by creating respective external table with hive, automatically generating mapreduce carry out with sql statement.When conjunctive query, a large table is with a little table conjunctive query, result is still very large, can use MapJoin to solve, in MapJoin, the operation of join can be placed on to Map and hold, carry out the transmission of Reduce, it utilizes memory headroom that little table is completely read in internal memory and completed again.
In last optimization module, how many very big computing time and efficiency of affecting of Reducer quantity, Reduce number is few, will greatly reduce combined pressure, and rationally controlling Reduce number becomes the emphasis of optimizing execution efficiency.Can be according to the manual performance that Reduce number affects bottom MapReduce executive routine that arranges, thus performance greatly improved.
Based on this, now provide a kind of HBase table is shown to associated conjunctive query optimization method with Hive.
Summary of the invention
Technical assignment of the present invention is to solve the deficiencies in the prior art, and a kind of HBase table method that conjunctive query is optimized is provided.
Technical scheme of the present invention realizes in the following manner, the method that this kind of HBase table conjunctive query is optimized, and its concrete query script is:
One, first according to tableau format in HBase, create corresponding Hive external table;
Two, after Hive external table creates successfully, between Hive table and HBase table, set up incidence relation, the row family of the row of Hive table and row type and HBase table and row determiner are set up associatedly, and mutually communicate by both external api interfaces own;
Three, enter query steps, according to the HQL statement carrying in Hive by two forms in Hive according to condition conjunctive query, with the MapReduce of bottom, process conjunctive query task, inquire the record satisfying condition;
Four, Query Result is left in HBase table.
Each territory in Hive table is present in HBase, and the RowKey in HBase corresponds in Hive corresponding for selecting a territory to come with key, and row family is mapped to other all territories in Hive.
The detailed content of described query steps is: the size of first checking two tables, if one of them table is little table, when another table is large table, with MapJoin, inquire about: in MapJoin, the logic of each MapReduce task be reducer can buffer memory join sequence in except the record of all tables of last table, by last table, result serializing is arrived to file system again, it is in the end the little table of join to be placed on to the left side of inquiry that maximum table is write, in the process of implementation, that little table is read in internal memory completely, consumption by larger memory headroom is cost.
The beneficial effect that the present invention compared with prior art produced is:
The method that a kind of HBase table of the present invention conjunctive query is optimized is by integrating Hive and HBase, HBase table is associated with Hive table, the conjunctive query of HBase table is by creating respective external table with hive, by the HQL statement carrying in Hive, conjunctive query being processed by bottom distributed computing framework MapReduce; This query script is parallel computation, on conjunctive query basis by cluster in parameter optimization improve search efficiency; Practical, be easy to promote.
Accompanying drawing explanation
Accompanying drawing 1 is Hbase table conjunctive query structural drawing in the present invention.
Accompanying drawing 2 is flowchart of HBase table of the present invention conjunctive query.
Accompanying drawing 3 is schematic diagrams of Hive conjunctive query module in the present invention.
Embodiment
Below in conjunction with accompanying drawing, the method for a kind of HBase table conjunctive query optimization of the present invention is described in detail below.
As shown in accompanying drawing 1～Fig. 3, the method that a kind of HBase table provided by the invention conjunctive query is optimized, first completes according to HBase table and Hive table relating module and conjunctive query module the conjunctive query function that HBase itself does not provide.It realizes principle is to process according to the MapReduce task of Hive bottom the degree of parallelism that its query function can raising program be carried out, and finally according to the command line parameter carrying in Hive, affects the performance of MapReduce, thereby has greatly improved performance.Its concrete query script is:
One, first according to tableau format in HBase, create corresponding Hive external table;
Two, after Hive external table creates successfully, between Hive table and HBase table, set up incidence relation, be that HBase table is shown in relating module with Hive, the realization of the integration function of Hive and HBase is to utilize both mutually to communicate by external api interface own, and intercommunication is mainly to rely on hive-hbase-handler.jar tool-class mutually.The integrated HBase of Hive can effectively utilize the storage characteristics of HBase database, as row renewal and column index etc.In integrated process, maintain the consistance of HBase jar bag.The integrated HBase of Hive sets up mapping relations between Hive table and HBase table, and the row columns of Hive table sets up associated with row type column types with the column families of row family and the row determiner column qualifiers of HBase table.Each territory in Hive table is present in HBase, it is territory use of selection that RowKey in HBase corresponds in Hive: key comes corresponding, row families (cf :) are mapped to other all territories in Hive, classify (cf:c1) or (cf:c2) etc. as, can the data of using are corresponding with type by the value in Hive table.
Three, enter query steps, according to the HQL statement carrying in Hive by two forms in Hive according to condition conjunctive query, by bottom distributed computing framework MapReduce, process conjunctive query task, making query script is parallel computation, thereby raising search efficiency, inquire the record satisfying condition, the method for this HBase table conjunctive query optimization mainly by HBase, show to show associated, conjunctive query with Hive and optimization module realizes.
Four, Query Result is left in HBase table.
Each territory in Hive table is present in HBase, and the RowKey in HBase corresponds in Hive corresponding for selecting a territory to come with key, and row family is mapped to other all territories in Hive.
Described query steps is: in conjunctive query module, according to the HQL statement carrying in Hive by two forms in Hive according to condition jion, with the MapReduce of bottom, process conjunctive query task, inquire the record satisfying condition.In this processing, can avoid directly writing MapReduce program, and can reach the effect of parallel processing simultaneously, improve search efficiency.
Its detailed content is: in conjunctive query module, first check the size of two tables, if one of them table is little table, when another table is large table, can inquire about with MapJoin.In MapJoin, the logic of each mapreduce task be reducer can buffer memory join sequence in except the record of all tables of last table, then by last table by result serializing to file system.This realization contributes to reduce at reduce end the use amount of internal memory.Therefore, that maximum table being write in the end to (otherwise can because buffer memory be wasted a large amount of internal memories) is the little table of join to be placed on to the left side of inquiry, in the process of implementation, is that little table is read in internal memory completely, and the consumption by larger memory headroom is cost.
Optimize in module, what when Hive realizes conjunctive query, use is the MapReduce program of bottom, however the efficiency how many very big impacts of reducer quantity are calculated, the number of Reduce is few, to greatly reduce combined pressure, therefore, controlling Reduce number is to optimize the major tasks of execution efficiency.The quantity of Reduce can be rationally set according to the size of the size of cluster and operation table, thereby reach the object of optimization.
According to Distributed Architecture, can improve degree of parallelism, then according to the Hadoop cluster scale of building, the relevant parameter configuration in suitable parameter and Hive is set, as mapred.reduce.tasks, when magnitude setting is large, can reach the object of parallel processing task.By rationally arranging of these parameters, can improve performance.
Embodiment: first dispose distributed type assemblies environment according to the step of official website, the hardware environment in this cluster is as shown in table 1.Operating system is centos6.3.Then correct hdfs, mapreduce, hbase and hive are opened and are served according to normal sequence in server cluster.
| Machine models | Cpu model | Core quantity | Internal memory | Disk size | Machine quantity |
| 5280m3 | Xeon(R) CPU E5-2620 0 2.00GHz | 24 | 96G | 6050G | 11 |
The form of source data text is as shown in table 2 in this example, and in file, comprising data is 1,000,000,000, and the size of file is 11G.File is uploaded in HDFS, then file is imported in HBase table.In HBase, set up in addition a little table, these two tables are carried out to conjunctive query.The structural drawing of HBase table conjunctive query as shown in Figure 1, first associated according to the structure of showing in HBase and the foundation of Hive table, then in Hive, carries out conjunctive query, is finally met the Query Result of conjunctive query condition.
| rowkey | uuid + "_" + timeStamp |
| C1 | String.valueOf(random.nextDouble()).substring(0, 4) |
| C2 | String.valueOf(random.nextDouble()).substring(0, 4) |
| C3 | uuid |
| C4 | timeStamp |
In the conjunctive query of HBase table, particular flow sheet as shown in Figure 2.First HBase table and Hive wish being inquired about carries out associated, sets up external table in Hive, and specific instructions is as follows:
#hive
CREATE?EXTERNAL?TABLE?table1(keys?tring,value1?string,value2?string,value3?string，value4?string)
STORED?BY?'org.apache.hadoop.hive.hbase.HBaseStorageHandler'
WITH?SERDEPROPERTIES("hbase.columns.mapping"=":key,info:c1,?info:c2,?info:c3,?info:c4?")
TBLPROPERTIES("hbase.table.name"="table1");
Wherein table1 is the large table in HBase, uses the same method and sets up the external table table2 in Hive, and these two tables are that conjunctive query is prepared.Then set up the tableresult that deposits Query Result, it is the internal table of Hive, and this table is for output is written in HBase, makes Hive table and HBase table produce synchronous effect.
In Hive, the complex query statement of two tables is to realize according to the HQL language providing in Hive.As shown in Figure 2, this process is to carry out conjunctive query according to two tables in Hive to concrete schematic diagram, and this query script is in fact MapReduce task processes, and the speed that this MapReduce processes has directly affected the efficiency of inquiry.The query statement of this process is as follows:
INSERT?OVERWRITE?TABLE?tableresult?Select?*?from?table2?left?semi?join?table2?on(table1.value1=table2.value1)；
In optimizing module, first before inquiry, according to the scale of hardware environment and cluster, rational parameter is set, and mapred.reduce.tasks parameter is set in hive, meet the efficiency that mapreduce program is carried out, by set mapred.reduce.tasks=300 is set.In this cluster, the complex query time shortens of two tables more than one times, from 31 minutes of starting by 14 minutes optimizing.This result absolutely proves the efficiency that how many very big impacts of reducer quantity are calculated when Hive realizes conjunctive query, therefore, according to the quantity of the control Reduce of the size reasonable of the size of cluster and operation table, is very important.The method can realize the conjunctive query of HBase table by Hive, one of its advantage is can reach the parallel computation effect of MapReduce program when need not write complicated MapReduce program; Two of advantage is by rational cluster parameter is set, to reach the object of optimization, improves search efficiency.
The foregoing is only embodiments of the invention, within the spirit and principles in the present invention all, any modification of doing, be equal to replacement, improvement etc., within all should being included in protection scope of the present invention.
Claims (3)
1. the method that the conjunctive query of HBase table is optimized, is characterized in that its concrete query script is:
One, first according to tableau format in HBase, create corresponding Hive external table;
Two, after Hive external table creates successfully, between Hive table and HBase table, set up incidence relation, the row family of the row of Hive table and row type and HBase table and row determiner are set up associatedly, and mutually communicate by both external api interfaces own;
Three, enter query steps, according to the HQL statement carrying in Hive by two forms in Hive according to condition conjunctive query, with the MapReduce of bottom, process conjunctive query task, inquire the record satisfying condition;
Four, Query Result is left in HBase table.
2. the method that a kind of HBase table according to claim 1 conjunctive query is optimized, it is characterized in that: each territory in Hive table is present in HBase, RowKey in HBase corresponds in Hive corresponding for selecting a territory to come with key, and row family is mapped to other all territories in Hive.
3. the method that a kind of HBase table according to claim 1 conjunctive query is optimized, it is characterized in that: the detailed content of described query steps is: the size of first checking two tables, if one of them table is little table, when another table is large table, with MapJoin, inquire about: in MapJoin, the logic of each MapReduce task be reducer can buffer memory join sequence in except the record of all tables of last table, by last table, result serializing is arrived to file system again, it is in the end the little table of join to be placed on to the left side of inquiry that maximum table is write, in the process of implementation, that little table is read in internal memory completely, consumption by larger memory headroom is cost.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410000720.8A CN103729453A (en) | 2014-01-02 | 2014-01-02 | HBase table conjunctive query optimization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410000720.8A CN103729453A (en) | 2014-01-02 | 2014-01-02 | HBase table conjunctive query optimization method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103729453A true CN103729453A (en) | 2014-04-16 |
Family
ID=50453527
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410000720.8A Pending CN103729453A (en) | 2014-01-02 | 2014-01-02 | HBase table conjunctive query optimization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103729453A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103942099A (en) * | 2014-04-30 | 2014-07-23 | 广州唯品会网络技术有限公司 | Parallel task execution method and device based on Hive |
| CN104268158A (en) * | 2014-09-03 | 2015-01-07 | 深圳大学 | Structural data distributed index and retrieval method |
| CN104346447A (en) * | 2014-10-28 | 2015-02-11 | 浪潮电子信息产业股份有限公司 | Partitioned connection method oriented to mixed type big data processing systems |
| CN104376047A (en) * | 2014-10-28 | 2015-02-25 | 浪潮电子信息产业股份有限公司 | Big table join method based on HBase |
| CN104376103A (en) * | 2014-11-26 | 2015-02-25 | 浪潮电子信息产业股份有限公司 | Multi-HBase table association method based on snowflake model |
| CN104391957A (en) * | 2014-12-01 | 2015-03-04 | 浪潮电子信息产业股份有限公司 | Data interaction analysis method for hybrid big data processing system |
| CN105550351A (en) * | 2015-12-28 | 2016-05-04 | 中国民航信息网络股份有限公司 | Passenger travel data ad-hoc query system and method |
| CN106528750A (en) * | 2016-10-28 | 2017-03-22 | 无锡雅座在线科技发展有限公司 | Data extracting method and device |
| CN106649503A (en) * | 2016-10-11 | 2017-05-10 | 北京集奥聚合科技有限公司 | Query method and system based on sql |
| CN107203594A (en) * | 2017-04-28 | 2017-09-26 | 努比亚技术有限公司 | A kind of data processing equipment, method and computer-readable recording medium |
| CN108255838A (en) * | 2016-12-28 | 2018-07-06 | 航天信息股份有限公司 | A kind of method and system for establishing the intermediate data warehouse for big data analysis |
| CN105357311B (en) * | 2015-11-23 | 2018-11-27 | 中国南方电网有限责任公司 | A kind of storage of secondary device big data and processing method of cloud computing technology |
| CN111382179A (en) * | 2020-03-10 | 2020-07-07 | 北京金山云网络技术有限公司 | Data processing method and device and electronic equipment |
| CN112069177A (en) * | 2020-08-31 | 2020-12-11 | 银盛支付服务股份有限公司 | Data query method and device, computer equipment and readable storage medium |
| CN112256704A (en) * | 2020-10-23 | 2021-01-22 | 山东超越数控电子股份有限公司 | Quick join method, storage medium and computer |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521246A (en) * | 2011-11-11 | 2012-06-27 | 国网信息通信有限公司 | Cloud data warehouse system |
| CN103440288A (en) * | 2013-08-16 | 2013-12-11 | 曙光信息产业股份有限公司 | Big data storage method and device |
-
2014
- 2014-01-02 CN CN201410000720.8A patent/CN103729453A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102521246A (en) * | 2011-11-11 | 2012-06-27 | 国网信息通信有限公司 | Cloud data warehouse system |
| CN103440288A (en) * | 2013-08-16 | 2013-12-11 | 曙光信息产业股份有限公司 | Big data storage method and device |
Non-Patent Citations (4)
| Title |
|---|
| GEORGE SFAKIANAKIS ETC.: "Interval Indexing and Querying on Key-Value Cloud Stores", 《DATA ENGINEERING (ICDE), 2013 IEEE 29TH INTERNATIONAL CONFERENCE ON》 * |
| KALOR: "hive与hbase的整合", 《博客园》 * |
| 叶文宸: "基于hive的性能优化方法的研究与实践", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 吕明育: "Hadoop 架构下数据挖掘与数据迁移***的设计与实现", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103942099A (en) * | 2014-04-30 | 2014-07-23 | 广州唯品会网络技术有限公司 | Parallel task execution method and device based on Hive |
| CN103942099B (en) * | 2014-04-30 | 2017-11-28 | 广州唯品会网络技术有限公司 | Executing tasks parallelly method and device based on Hive |
| CN104268158A (en) * | 2014-09-03 | 2015-01-07 | 深圳大学 | Structural data distributed index and retrieval method |
| CN104346447A (en) * | 2014-10-28 | 2015-02-11 | 浪潮电子信息产业股份有限公司 | Partitioned connection method oriented to mixed type big data processing systems |
| CN104376047A (en) * | 2014-10-28 | 2015-02-25 | 浪潮电子信息产业股份有限公司 | Big table join method based on HBase |
| CN104376047B (en) * | 2014-10-28 | 2017-06-30 | 浪潮电子信息产业股份有限公司 | A kind of big table join methods based on HBase |
| CN104376103B (en) * | 2014-11-26 | 2017-10-03 | 浪潮电子信息产业股份有限公司 | A kind of multiple HBase table correlating methods based on snowflake model |
| CN104376103A (en) * | 2014-11-26 | 2015-02-25 | 浪潮电子信息产业股份有限公司 | Multi-HBase table association method based on snowflake model |
| CN104391957A (en) * | 2014-12-01 | 2015-03-04 | 浪潮电子信息产业股份有限公司 | Data interaction analysis method for hybrid big data processing system |
| CN105357311B (en) * | 2015-11-23 | 2018-11-27 | 中国南方电网有限责任公司 | A kind of storage of secondary device big data and processing method of cloud computing technology |
| CN105550351B (en) * | 2015-12-28 | 2019-05-14 | 中国民航信息网络股份有限公司 | The extemporaneous inquiry system of passenger's run-length data and method |
| CN105550351A (en) * | 2015-12-28 | 2016-05-04 | 中国民航信息网络股份有限公司 | Passenger travel data ad-hoc query system and method |
| CN106649503A (en) * | 2016-10-11 | 2017-05-10 | 北京集奥聚合科技有限公司 | Query method and system based on sql |
| CN106528750A (en) * | 2016-10-28 | 2017-03-22 | 无锡雅座在线科技发展有限公司 | Data extracting method and device |
| CN106528750B (en) * | 2016-10-28 | 2020-05-15 | 无锡雅座在线科技股份有限公司 | Data extraction method and device |
| CN108255838A (en) * | 2016-12-28 | 2018-07-06 | 航天信息股份有限公司 | A kind of method and system for establishing the intermediate data warehouse for big data analysis |
| CN107203594A (en) * | 2017-04-28 | 2017-09-26 | 努比亚技术有限公司 | A kind of data processing equipment, method and computer-readable recording medium |
| CN111382179A (en) * | 2020-03-10 | 2020-07-07 | 北京金山云网络技术有限公司 | Data processing method and device and electronic equipment |
| CN111382179B (en) * | 2020-03-10 | 2023-12-01 | 北京金山云网络技术有限公司 | Data processing method and device and electronic equipment |
| CN112069177A (en) * | 2020-08-31 | 2020-12-11 | 银盛支付服务股份有限公司 | Data query method and device, computer equipment and readable storage medium |
| CN112256704A (en) * | 2020-10-23 | 2021-01-22 | 山东超越数控电子股份有限公司 | Quick join method, storage medium and computer |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103729453A (en) | HBase table conjunctive query optimization method | |
| US11055287B2 (en) | Eigenvalue-based data query | |
| EP3144826B1 (en) | A method and apparatus for representing compound relationships in a graph database | |
| US10169433B2 (en) | Systems and methods for an SQL-driven distributed operating system | |
| Halperin et al. | Demonstration of the Myria big data management service | |
| US9535963B1 (en) | Graph-based queries | |
| US8819335B1 (en) | System and method for executing map-reduce tasks in a storage device | |
| US9330138B1 (en) | Translating queries into graph queries using primitives | |
| US9378241B1 (en) | Concatenated queries based on graph-query results | |
| Song et al. | HaoLap: A Hadoop based OLAP system for big data | |
| US9378233B2 (en) | For all entries processing | |
| US9396018B2 (en) | Low latency architecture with directory service for integration of transactional data system with analytical data structures | |
| US20160103914A1 (en) | Offloading search processing against analytic data stores | |
| CN103425762A (en) | Telecom operator mass data processing method based on Hadoop platform | |
| US10261950B2 (en) | Table as query language parameter | |
| US20180285475A1 (en) | Flexible record definitions for semi-structured data in a relational database system | |
| US20150154259A1 (en) | Sql query on a nosql database | |
| CN106611037A (en) | Method and device for distributed diagram calculation | |
| US10339040B2 (en) | Core data services test double framework automation tool | |
| US9378239B1 (en) | Verifying graph-based queries | |
| CN106844369B (en) | Objectification SQL sentence construction method and apparatus | |
| Naheman et al. | Review of NoSQL databases and performance testing on HBase | |
| CN106528898A (en) | Method and device for converting data of non-relational database into relational database | |
| CN107480202B (en) | Data processing method and device for multiple parallel processing frameworks | |
| US20150269234A1 (en) | User Defined Functions Including Requests for Analytics by External Analytic Engines |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140416 |