Pass through the method for tables of data shared data between a kind of tenant
Technical field
The present invention relates to big data technical field, shared between specifically a kind of highly practical, tenant by tables of data
The method of data.
Background technology
Multi-tenant technology(English:multi-tenancy technology)Or multiple leasing techniques, it is a kind of software
Architecture technology, it be inquire into realize how in the environment of multi-user shared identical system or program assembly, and still
It can ensure that the isolation of data between each user.
Due to the fever of cloud computing subject under discussion, as how triangular web framework and service offer are more in shared data center
Clients are identical or even the service of customizable for number, and still can ensure the data isolation of client, allow multi-tenant technology into
For aobvious under cloud computing technology.
In cloud computing, the epoch of big data, data are managed, shared and using being an important problem between tenant, such as
How effectively management and the data resource using cloud computing determine the vitality of the upper layer application ecosystem, based on this, now carry
For by the method for tables of data shared data, this method is from the angle that data management, data sharing and data utilize between a kind of tenant
Degree sets out, with it is a kind of gear to actual circumstances, healthy and strong easy-to-use mode is the shared of big data under cloud computing environment and using proposing one
The practical solution of kind.
The content of the invention
The technical assignment of the present invention is to be directed to above shortcoming, there is provided passes through tables of data between a kind of highly practical, tenant
The method of shared data.
It is by the method for tables of data shared data, its specific implementation process between a kind of tenant:
Tenant data source control is carried out first:Under cloud computing environment, tenant by online web page logging data source,
Apply for new data management service, after typing, opening timing device, periodically updates the data;
Carry out data classification:Comb and customize data directory and group list information, tenant closes the tables of data created
Corresponding data directory on connection, easy to tables of data retrieval and browse;
Complete data sharing:The tables of data opening and shares of tenant are simultaneously carrying out application use each other.
The data source is relevant database or HBASE NoSQL databases, and following information is included in the data source:
IP, port, data source types, character set, description, field, table information.
The process that the timing updates the data is:Database, table information and field information under dynamic pulling data source, will
These metadata automatic synchronizations are to data management system, opening timing device, periodically according to the change of real data table dynamic
More new metadata.
Each tenant of data source control uses a set of independent data management service between the tenant, the data between tenant
It is mutually isolated, it is independent of each other.
The data are divided into two kinds of vertical and horizontal:The classification of longitudinal direction is mutually independent data dimension, the data
Dimension includes key element, department;Horizontal data classification is the classification under same data dimension, including personnel under same key element and
Article.
In the data sharing step, after cloud tenant selects the data to be opened, system is serviced automatically by data by ETL
Table is synchronized in Hive, using Hive as data warehouse, carries out the management of data permission, the data of multi-tenant are unified in Hive
Storage;The data opened can be applied using by other tenants;Tenant beyond the clouds combine application tables of data with itself
Tables of data is associated inquiry and analysis by SQL, and the execution authority of SQL is controlled by software systems, according to data application shape
State carries out logic judgment.
By the method for tables of data shared data between a kind of tenant of the present invention, has the following advantages:
By the method for tables of data shared data between a kind of tenant proposed by the present invention, multi-tenant under cloud computing environment is solved
Data isolation between data resource online management, metadata management and tenant;The sharing problem of big data, is high in the clouds between solution tenant
Data mining provides basis;Data application, data are audited and licensed between support tenant, it is allowed to which tenant is existed based on web page
The privately owned data resource of wire management, can carry out data sharing by data application;Shared data is united after the importing of ETL data
One is stored in Hive, and using Hive as data warehouse, the shared of table is carried out by data application between tenant;It is highly practical, it is easy to
Promote.
Embodiment
Below by specific embodiment, the invention will be further described.
The invention discloses by the method for tables of data shared data, under private clound, multi-tenant remotely counts between a kind of tenant
Online data management and the utilization of resources according to multi-source data resources such as source, RDS data sources, tenant is carried out with reference to Hive, Kettle
Between data shared and utilize, for more tissues, the data integration under multidisciplinary and multiservice system and it is shared propose it is one whole
The solution of set.
It, which implements process, includes following three steps:
First, tenant data source control.
Under cloud computing environment, after tenant applies for that a new data management service, application pass through, pass through online web page
Logging data source.
Data source includes the NoSQL databases such as relevant database and the HBASE of mainstream, and data source information includes IP, end
The information such as mouth, data source types, character set, description, after typing, database, table information under dynamic pulling data source and
Field information, these metadata are synchronized automatically between data management system, opening timing device, periodically according to real data table
Change dynamic more new metadata.
Different from general metadata management system, data source control is each rent based on cloud computing environment between tenant
Family can apply for a set of independent data management service, and the data between tenant are mutually isolated, are independent of each other.
2nd, data are classified.
Data directory and group list information are combed and customized according to industry characteristic, and tenant closes the tables of data created
Corresponding data directory on connection, easy to tables of data retrieval and browse.
Data are divided into vertical and horizontal both of which, and longitudinal classification can be mutually independent data dimension, than
Such as key element, department, the classification of horizontal data are classifications under same data dimension, such as the personnel under same key element and article
Deng.
3rd, data sharing.
The tables of data of tenant with opening and shares and can carry out application use each other.Cloud tenant selection will open
Data after, system by ETL service tables of data is synchronized in Hive automatically, using Hive as data warehouse, progress data
The management of authority, the data of multi-tenant, which are unified in Hive, to be stored.
The data opened can be applied using by other tenants.
Tenant is associated inquiry and analysis, SQL with the tables of data of itself with reference to the tables of data of application by SQL beyond the clouds
Execution authority controlled by software systems, carry out logic judgment according to data application status.
Above-mentioned embodiment is only the specific case of the present invention, and scope of patent protection of the invention includes but not limited to
Above-mentioned embodiment, passes through the claim of the method for tables of data shared data between a kind of any tenant for meeting the present invention
The appropriate change or replacement that the those of ordinary skill of book and any technical field does it, should all fall into the present invention's
Scope of patent protection.