Oracle International Corporation

Ajay Keni

Ramaprakash Hosalli Sathyanarayan

Hari Sastry

Neelima Bawa

Systems and methods for performing a bulk append into an LDAP Server Directory Store. A conversion tool may receive an input file in LDIF format. The conversion tool performs a schema check on each entry in the input file, and creates a plurality of output files in SQL or another LDAP database structured format. One output file is created for each attribute in the input file. In addition, a file for an attribute table and a file for a catalog table may be created. One or more SQL Loading tools then load the output files into the LDAP Server Directory Store in parallel, except for the DN catalog table. After the other output files have been loaded, the DN catalog file is loaded into the LDAP Server Directory Store.

LDAP bulk append

The method, system and computer program product for efficient retrieval of knowledge referral are disclosed. In one approach, the method, system and computer program product modify a query for retrieving an entry from a directory information system, that is stored in a relational database, to retrieve the entry and a knowledge referral.

Efficient retrieval and storage of directory information system knowledge referrals

Systems and methods for providing consistent metadata in a cluster database system are disclosed. Each node contains a number of LDAP server instances, as well as a monitor process. Each node contains a node shared memory, and the cluster contains a cluster shared memory stored on a directory store. When a client alters metadata in one of the LDAP server instances, the LDAP server instances writes the changes to the directory store, and provides update information, such as a version number, to the node shared memory. Other server instances on that node poll the node shared memory to determine whether updates have been made. If updates have been made, the other server instances update their on-heap caches based on the directory store. Each monitor process determines whether changes have been made to the cluster shared memory, and if so, updates the node shared memory based on the directory store.

Growing community of inventors

Fremont, CA, United States of America

Ajay Keni