Model 3 addresses how to reduce the impact of outages with a fault tolerant architecture and GDPS. Always be prepared to restore service as quickly as possible and anticipate failures because they are going to happen using the right technology implementation. Designing a fault tolerant architecture can help prevent failures from causing a single point of failure. These include having a Parallel Sysplex System, cloning most hardware and software components, and planning for workload to move from one logical partition to another. Geographically Dispersed Parallel Sysplex, GDPS, automates management of the central processing complex and site recovery actions. Combining both Metro Mirror, which is a function of the external storage unit where the target copy of a volume is constantly being upgraded to match changes that are made to a source volume, and Global Mirror, which provides data replication over extended distances between two sites for business continuity and disaster recovery that will aid in reducing the duration of a disk outage. There are four GDPS Metro configurations. A GDPS Metro Single Site Workload environment typically consists of a Multisite Sysplex with all production systems running in a single site, normally Site1, and the GPS controlling system in Site2. Multiple site differs from a single site workload in that production systems are running in both sites. Multi-site workload is usually a result of higher availability requirements and Parallel Sysplex and data sharing are core components of such an environment. For the third site, BRS, or Business Recovery Services, all these systems in the GDPS configuration, including the controlling systems, are in a sysplex in the same site namely Site1. This sysplex does not span the two sites. The second site, Site2 might be a client site or might be owned by a third party recovery services provider. GDPS Metro dual-leg configuration uses the IBM MTMM Disk Mirroring Technology to provide a dual-leg configuration which maintains two synchronous secondary copies of data to provide an additional level of data resiliency. Geographically Dispersed Parallel Sysplex Metro supports both planned and unplanned outages, which helps maximize application availability and providing business continuity. Now, what are the benefits of GDPS Metro? Resiliency to support multiple disk replication architectures, high availability, and disaster recovery at any distance from a single point of control in a matter of seconds. Flexibility to gain data consistency across both z/OS, z/VM and guests, plus open system data. GDPS also uses open architecture and is application independent. Automation of operational tasks for availability and recovery with scripts. Monitoring of the environment from a single point and in real time to detect any deviation that could impact resiliency. GDPS CA further extends this concept with the ability to switch the workload between two sysplexes separately by virtually unlimited distance for both planned and unplanned outage situations. Low recovery times are achieved by having a sysplex active at all times in two different locations as long as the software-based replication can be provided with network connectivity between these sites. Take some time to view the different GDPS configurations on this slide. Moving on, for the third deployment model let's consider the following items. The connectivity between distances depends on the distance connecting multiple sites. IBM FICON switches provide short distances, whereas longer distances would require channel extenders. Also, the distance is a factor. The further the sites are from each other also adds to the time for the data to be transferred. When running in Parallel Sysplex mode, all of these systems must run the same GDPS Metro configuration. This ensures that critical events are captured and properly resolved. When there are multiple systems involved, there must be at least one z/OS controlling system. For a better redundancy, two is always better. GDPS Metro with HyperSwap can achieve zero data loss due to the fact that when updates are made to the primary disk, the updates are automatically forwarded to the secondary backup disk. When planned outages are scheduled, the graphical user interface can assist and speed up the process of stopping images, updating server time protocol roles, creating new coupling datasets, activate capacity backup if the need arises, IPL the production system on the other site, and respond to all IPL messages. The Multi-site workload slide explains the benefit of having two separate sites. Here, GDPS is controlling system K1 running in Site2, and also controlling system K2 running in Site1. K1 is the controlling system ready to assist in HyperSwap and failover automation if there is a problem with Site1. Coupling facilities are not required but do improve availability for structures. All of the backup logical partitions are defined as inactive. If an outage occurs at either site, the GDPS Controlling System decides how to react to the failure, determining the necessary recovery actions, and managing the recovery. Now, the great thing about Parallel Sysplex is that it's possible to build an environment with no single points of failure. All of the members have access to all of the critical applications and data. The picture on this slide shows Site1 having a problem, and all processing must go to Site2 as quickly as possible.
