Ensuring Business Continuity in a SAP Environment

Disaster Recovery is about protecting data. Datacenter availability involves protecting both the data and the application.

Vishal Dhupar

Planned or unplanned, downtime signals the certainty of real, quantifiable, losses for businesses that rely on SAP applications. After all, for companies around the globe, SAP applications are fundamental to their daily operations. Any disruption in the critical business processes to which SAP applications are linked could result in the loss of millions of dollars or more every hour.

Research suggests that many companies experiencing a major disaster are at risk of going out of business within the next two months.

As a result, a growing number of organisations are recognising the importance of developing a business continuity plan that protects their SAP environment. To be effective, this plan must address the three primary areas of concern within SAP—availability, manageability, and support for data centre availability —through a combination of end-to-end technology solutions and practices.

It’s important to understand what is being implied when we say datacentre availability in the context of this article. Disaster Recovery (DR) is about protecting data.  Typically companies back up their data to tapes or disk and recover it to a back up server.  This may also entail replicating data from one storage system to a storage system in another location. 

Datacenter availability is the protection of both the data and the application itself.  The key differentiator is that datacenter availability recognises the importance of keeping an organisations applications up and running and protects the application by allowing the application to be failed over from one server to another server regardless of whether or not the server is part of a local, stretch or remote cluster. 

Environmental Challenges

SAP applications must be available not only on a daily basis but they must also be protected against loss of service and data resulting from a disaster or widespread, regional disruption.

However, the SAP environment is highly complex, with multiple services deployed across multiple servers for an SAP application instance. Each service—from the messaging service to the enqueue service, database instance, and Network File System (NFS) or CIFS services—represents a potential single point of failure. Redundancy must be built at the component level for each layer, which, in turn, increases the complexity and challenges of managing the environment.

Also, while maintaining an alternative location that is geographically separate from the primary production environment is an effective business continuity option, it can also be costly. Traditional SAP recovery solutions often require that identical, expensive hardware be deployed at all locations. Secondary sites often must also be maintained at the same update and patch levels as the primary site.

Switching an application to the disaster recovery site is also cumbersome and challenging, often requiring IT to move tapes and personnel from one facility to another, rebuild servers, load operating systems, and back up software. Also, at the secondary site, IT personnel must be available to manually walk through complicated, urgent, but potentially error-prone procedures for restarting interdependent SAP components.

Available, Manageable, Recoverable

In an SAP environment, ensuring business continuity starts locally. For example, local redundancy can be achieved by utilising a single interface for starting, stopping, monitoring, and maintaining SAP application services. Change can be managed by proactively moving application services to enable dynamic maintenance and testing.

Local availability can be improved with automated, application-specific monitoring and failover when problems do occur and fast reconfiguration when the problem is resolved. Local clustering and the virtualisation of critical services can enable the organisation to consolidate servers, make better use of existing resources, and protect against potential single points of failure. And policy-based storage management can significantly improve storage utilisation and enhance productivity.

Once the local availability infrastructure is in place, organisations can then leverage their existing architecture and build an alternative site for remote datacenter availability. Adding global clustering and data replication enables organisations to quickly and accurately switch over an entire SAP application between geographically dispersed datacenters.

A growing number of SAP availability solutions eliminate the cost concerns typically associated with having duplicate data centers. For example, solutions that do not require secondary sites to be identical in computing resources to primary sites obviate the need for organisations to invest in expensive, matching servers.

Moreover, organisations can realise significant cost savings by using their secondary site for other, less essential processes and then simply shut them down before bringing more critical processes online in the event of a disaster.

Global clustering capabilities link together clusters from geographically separate sites and connect SAP applications across them. By providing SAP application failover to a remote site, this capability helps ensure service level protection against the failure of an entire site.

The most advanced global clustering solutions continuously monitor and communicate SAP application events between clusters, thereby making sure that if a system or application service fails, the affected services fail over to another system in the same cluster or, if an entire cluster or site fails, the SAP application itself fails over to the remote cluster. Then, once the application is online, clients are automatically redirected to the new location. This not only ensures the availability of SAP in the event of a disaster, but it also enhances manageability by providing a single point of management for the entire SAP application rather than offering management on a per-SAP-component basis only.

Software-based volume replication solutions further enhance data center availability and protect business continuity by enabling site migration at the click of a button. Replication tools that work over any IP network ensure maximum efficiency and flexibility, while solutions that support a variety of replication configurations—including one-to-one, one-to-many, and many-to-one—help streamline business continuity efforts and ease management.

For global failover scenarios, organisations often use volume replication tools that support asynchronous replication. With asynchronous replication, replication operations queue for network availability, which produces the highest performance but introduces the potential for data loss.

A number of replication tools also offer synchronous replication, which guards against data loss but may impact performance. However, a small number of replication solutions today offer enhanced asynchronous replication by enforcing write order on the replicated site, which ensures data integrity and consistency and guarantees that data will be recoverable at the secondary site.

Best Practices

Technical solutions to business continuity needs help automate critical procedures, reduce administrative costs, increase efficiency, and minimise error. These technologies can be complemented by a business continuity plan that incorporates best practices for assessing risk factors, implementing a datacenter availability strategy, testing without disrupting production servers and validating the technologies and activities established to protect essential business services.

A business continuity plan provides a structured approach for the development of the architecture, processes, and practices for both identifying and reaching recovery time and recovery point objectives.

To help build the most comprehensive and effective strategy and plan, organisations can leverage the expertise of certified business continuity professionals who can provide benchmarks, identify bottlenecks, and recommend solutions that address the requirements of auditors, shareholders, C-level executives, and other corporate officers.

Protecting the availability, manageability, and recoverability of SAP applications is critical to ensuring business continuity. SAP applications form the foundation of an organisation’s most vital processes. As a result, downtime is simply not an option.

By leveraging advanced technologies that provide global clustering and replication of the SAP environment across geographically distributed, heterogeneous sites, and by complementing these technologies with sound best practices, organisations can not only keep their businesses up and running, but growing as well.

— By Vishal Dhupar, Managing Director, Symantec India