Disaster recovery for Linux on System z

This topic is intended to give an overview about how to provide disaster recovery and high availability for Linux™ on System z® using Geographically Dispersed Parallel Sysplex™ (GDPS®) and IBM® Tivoli® System Automation.

Detailed information is available in the IBMRedbooks® publication:

IBM GDPS Family - An Introduction to Concepts and Capabilities.

Due to more and more applications running on Linux on System z that are critical for the daily business, a solution for continuous availability, high availability, and disaster recovery is required. The goal is to keep a business critical application running 24 hours a day x 7 days a week even if there are downtimes for parts of the infrastructure. Reasons for downtimes can be planned downtimes such as hardware or software maintenance as well as unplanned downtimes due to, for example, application failures, network failures, hardware failures or disasters.

Geographically Dispersed Parallel Sysplex (GDPS) is IBM's premium business resilience solution for System z environments running workloads on z/OS®. It provides disaster and failure recovery from a single point of control and ensures data consistency across multiple sites. Disaster recovery solutions always need to take into account data replication from a primary storage subsystem to a backup storage subsystem. GDPS/PPRC is the variant of GDPS that performs synchronous data replication using IBM Metro Mirror (aka PPRC) technology.

The function of GDPS/PPRC can be extended using IBM Tivoli® System Automation to provide a coordinated disaster recovery solution for computing environments that comprise workloads running on both z/OS and Linux on System z operating systems where the Linux images storing data on CKD disks may be running in native LPARs or in z/VM® guests. The extension of GDPS/PPRC that provides disaster recovery for both z/OS and Linux is called GDPS/PPRC Multiplatform Resiliency for System z. It is also known as xDR.

IBM Tivoli System Automation additionally allows you to extend GDPS/PPRC Multiplatform Resiliency to provide application level high availability for Linux on z workloads in addition to system level disaster recovery, continuous availability for Linux systems running on System z.