Techdocs Library > White papers > IMS Disaster Recovery with GDPS

Document Author:	Glenn Galler		Document ID:	WP101107
Doc. Organization:	Advanced Technical Sales		Document Revised:	09/30/2007
Product(s) covered:	DB2 for z/OS and OS/390; GDPS; IMS

Abstract: In this document, we discuss the key concepts and definitions to gain an understanding of how IMS works in a Geographically Dispersed Parallel Sysplex (GDPS) Disaster Recovery environment. For any Disaster Recovery solution, the key is to create consistent recovery points at a secondary site. The data at the remote site may agree completely with the primary site or it may lag behind. This can result in a loss of data in the event of a true disaster. The use of Consistency Groups and Dependent Writes for IMS can reduce the amount of data lost at the remote site. In certain mirroring environments, it is possible to reduce the time of the recovery by duplexing the Coupling Facility structures. We will discuss which CF structures can benefit from duplexing and the structures that do not benefit from duplexing in a disaster recovery solution. There are a number of GDPS mirroring configurations and we will focus on five of the most common environments used by IMS customers. With any disaster recovery solution, the key is to test the recovery procedures at the secondary site. This testing ensures that all of the key data sets are available and the Consistent Recovery Points are created correctly to allow IMS to do an Emergency Restart in the shortest amount of time. To conclude, there is a list of customer references, useful websites, interesting publications and one service offering available from IBM Global Services.

Classification:	Software; Solutions	Category:	Backup and Recovery
Solution(s):	Business Innovation and Optimization (BIO)
Platform(s):	IBM System z Family	S/W Pillar(s):	Information Management
O/S:	z/OS	Keywords:	IMS, Disaster Recovery, GDPS, Business Continuity, Mirroring, Replication