Sap Se

Ajalesh Puthenparambil Gopi

Naveen Kolli

Axel Herbst

Santosh V

Veit Bolik

Tenzin Choegyen

Sharath Jois

Santosh Vijayan

Naveen K

Vittal Gopinatha Pai

A framework for memory optimization of database indexes, and in particular for aging full-text index data, is described herein. In one embodiment, if, while a database table is aged, there are index tables associated with the database table, the associated index tables are automatically aged. This way, the system memory footprint will be reduced, leading to reduced cost as less system memory is required to perform remaining operations, and leading to increased performance as more system memory is available for other operations.

Memory optimization using data aging in full text indexes

A query is received in a session established between an application and an in-memory database. The query is received from the application. Based on one or more criteria defined for individual data aging parameters in the data aging infrastructure, an aging temperature to be applied to the query is determined. The determined aging temperature is received at the application. The aging temperature is automatically set in the session before executing the query in the in-memory database. Based on the determined aging temperature, a partition is loaded from a secondary memory to a main memory. With the aging temperature as reference, the query is executed to fetch a result from the in-memory database.

Data aging infrastructure for automatically determining aging temperature

Examples of partitioning a group of related database tables are provided herein. A database table in a group of related database tables can be designated as a lead database table. A partitioning field can also be determined by which database tables in the group are partitioned. A data load, with respect to the partitioning field, of the lead database table can be calculated. The data load can include a data distribution across different values of the partitioning field. A group partitioning scheme can be determined based on the data load of the lead database table. The database tables of the group can then be partitioned according to the group partitioning scheme.

Efficient partitioning of related database tables

Systems and methods for creating and populating temporary database tables with intermediate data that can be preserved once a database application session ends and used outside of the instant database operation without storing this data in persistent memory are provided. A local temporary database table may be used only in the local instance of a database application, and a global temporary table makes intermediate data available across multiple applications and multiple application instances.

Global and local temporary database tables

The user of an application may query a data storage system with a request to access a dataset from a database of the system. The system identifies at least one data access rule that is applicable to the dataset, with the at least one data access rule specifying at least one user group authorized to access a restricted portion of the dataset. The system identifies an authorized access group associated with the application user and compares it to the at least one user group authorized to access the restricted portion of the dataset. If the authorized access group associated with the user does not match one of the at least one user group authorized to access the restricted portion of the dataset, the system assembles a modified dataset based on the dataset and the at least one data access rule and transmits the modified dataset to the application.

Data access rules in a database layer

In dynamic data access, a request is received to access data of a core data service view of an in-memory database. It is determined that an aging temperature parameter is specified in an annotation in a core data service view definition. An aging temperature value corresponding to the aging temperature parameter is received as a range restriction. A default access behavior associated with the core data service view definition is overridden. A partition where the aging temperature value lies in a secondary memory is determined. Latest or recent partition in the secondary memory is referred to as a latest partition. Data from the latest partition until the determined partition is accessed in the secondary memory. The accessed data is loaded from the secondary memory to the main memory.

Growing community of inventors

Bengaluru, India

Ajalesh Puthenparambil Gopi