International Business Machines Corporation

Ajay Kumar Gupta

Karl Ostner

In a database system having a primary server side () and a secondary server side (), a high availability data replicator () transfers log entries from the primary side () to the secondary side () and replays the transferred log entries to synchronize the secondary side () with the primary side (). R-tree index transfer threads () copy user-defined routines, the user defined index, and index databases deployed on the primary server side () to the secondary server side () and deploy the copied user-defined routines, reconstruct the user-defined index, and copy data pages on the secondary side () to make the user-defined index consistent and usable on the secondary side ().

Database backup system using data and user-defined routines replicators for maintaining a copy of database on a secondary server

In a database, a high availability data replicator () synchronizes primary and secondary servers (). A smart large object application program interface () resides on the secondary server (). A cache module () creates a memory cache () corresponding to a smart large object responsive to a smart large object read operation requested by a client (). A smart large object read module () executes the read operation without acquiring a lock on the corresponding smart large object. An exception module () sends an exception to the client () responsive to a synchronizing event of the high availability data replicator () modifying said smart large object. On the primary server (), log entries of the modifying operation are ordered so that a log entry updating a large object header of said smart large object is consistent immediately upon execution on the secondary server ().

Growing community of inventors

Fremont, CA, United States of America

Ajay Kumar Gupta