International Business Machines Corporation

Yan Wang Stein

Abhinay Ravinder Nagpal

Sandeep R Patil

Dhaval Kiritbhai Shah

John Gerard Musial

Gaurav Mehrotra

Harshwardhan S Mulay

Jay A Athalye

Dwip N Banerjee

Rohit Raut

Systems and methods to manage database data are provided. A particular method includes automatically identifying a plurality of storage devices. The storage devices include a first device of a first type and a second device of a second type. The first type includes a solid state memory device. The method may further identify a high priority data set of the database. A rebalancing operation is conducted that includes moving the high priority data set to the solid state memory device and substantially evening distribution of other data of the database among the storage devices.

Rebalancing operation using a solid state memory device

The present invention includes a computer-implemented method including, classifying a data set in response to metadata corresponding to one or more data files located on a single database; and creating a data file topology comprising a data file identifier, a data file location and a data file type. The method may also include receiving a predetermined rule directory comprising a set of features corresponding to one or more file systems; and in response to the data file topology and the predetermined rule directory, reorganizing the data set such that at least a portion of the data set is moved to one of a set of new file systems having a predetermined optimized characteristic.

Intelligent redistribution of data in a database

A dataset is identified as a heat-intensive dataset based, at least in part, on the dataset being related to heat generation at a source storage device exceeding a heat rise limit. The source storage device hosts the heat-intensive dataset and the heat-intensive dataset comprises non-executable data. A first cooling area of a plurality of cooling areas is selected to accommodate the heat generation based, at least in part, on cooling characteristics of a plurality of cooling appliances of the plurality of cooling areas. The source storage device is associated with a second cooling area. A target storage device associated with the first cooling area is determined. The heat-intensive dataset is moved from the source storage device to the target storage device.

Cooling appliance rating aware data placement

A method and system for automatically and transparently managing a directory in a database system to preferentially use of objects in an in-memory database instead of copies of the objects in disk-based databases, in order to speed up application response times. The methods and systems selectively add or update LDAP entries to leverage advantage of the database objects in the in-memory database, so that the LDAP entries point to the in-memory database instead of disk-based databases.

System and method for managing directories for a database system having an in-memory database

For distributing data requests between data centers, a cost reduction module selects a data center from a plurality of data centers to fulfill a data operation request corresponding to mirrored data stored at the plurality of data centers. The selection is based on data center cooling efficiency factors comprising a data storage device efficiency percentage. A response module sends the data operation request to the selected data center.

Efficient use of mirrored storage clouds

Adaptive index density in a database management system is provided, which includes receiving a number of partitions for an index for a database table, the index subject to creation. The adaptive index density also includes selecting a column from the database table, the column selected based upon an estimated frequency of execution of database queries for the column. The adaptive index density further includes calculating an estimated cost of executing each of the database queries for the column, and determining data to reside in each of the partitions of the index responsive to the estimated cost.

Intelligent adaptive index density in a database management system

A plurality of data queries that operate on a heat-intensive dataset is determined. A plurality of access plans that correspond to the plurality of data queries is determined. An overall power consumption of the plurality of access plans is computed. It is determined that the overall power consumption exceeds a power consumption threshold. A first cooling area of a plurality of cooling areas to accommodate the heat generation based, at least in part, on cooling characteristics of a plurality of cooling appliances of the plurality of cooling areas is selected. A target storage device associated with the first cooling area is determined. The heat-intensive dataset from the source storage device is moved to the target storage device.

A method, a system, and a computer-readable storage medium are provided for data management. The method may comprise identifying a predefined set of data storage parameters and a predefined set of data scrubbing parameters. The method further may comprise determining the predefined set of data scrubbing parameters for first data in a first data storage based on the predefined set of data storage parameters for the first data and performing data scrubbing for the first data using the determined predefined set of data scrubbing parameters. Furthermore, the method may comprise comparing first data in the first data storage and corresponding second data in a second data storage. Upon determining a mismatch between the first data and the second data, the method further may comprise synchronizing the first data with the second data as a result of the comparison.

Disk scrubbing

Systems and methods for authentication in a data management system are provided. In one embodiment, the method comprises identifying a plurality of data ciphers based on a pre-defined set of properties associated with a plurality of data sources; and generating an authentication response having at least a subset of the identified plurality of data ciphers, wherein the authentication response authenticates access to at least some data sources from among the plurality of data sources.

Authentication in data management

An apparatus, system, and method are disclosed for distributing data requests between data centers. A request module receives a data operation request from a client. The data operation request corresponds to mirrored data stored at a plurality of data centers. A metadata module maintains a set of data center cooling efficiency factors for each of the plurality of data centers. Each data center cooling efficiency factor comprises data relating to operational costs of a data center. A cost reduction module compares sets of data center cooling efficiency factors for each of the plurality of data centers to select the most efficient data center from the plurality of data centers. A response module sends the data operation request to the selected data center.

Apparatus, system, and method for efficient use of mirrored storage clouds

A method, a system, and a computer program product are provided for data management. The method may comprise identifying a predefined set of data storage parameters and a predefined set of data scrubbing parameters. The method further may comprise determining the predefined set of data scrubbing parameters for first data in a first data storage based on the predefined set of data storage parameters for the first data and performing data scrubbing for the first data using the determined predefined set of data scrubbing parameters. Furthermore, the method may comprise comparing first data in the first data storage and corresponding second data in a second data storage. If a mismatch is determined between the first data and the second data, the method further may comprise synchronizing the first data with the second data as a result of the comparison.

Growing community of inventors

San Jose, CA, United States of America

Yan Wang Stein