Restore protection letter
Everything went smoothly the first time. The backup went through cleanly. A restore of the SAP productive systems that had just been set up also delivered spotlessly clean results. "All was well" was written under the report and the backup was sent to the archive.
One year later, however, the emergency had occurred - only a restore could help now. At first, everything looked fine. At least, no abnormalities had been observed during the regular backups before.
However, the first restore was promptly followed by a rude awakening. Small errors had crept in during the data backups and were updated over time. In the end, nothing fit together cleanly.
Parts of the restored master and transaction data on customers, prices, orders and delivery times had become equally unusable. Business operations were no longer conceivable.
In such cases, external SAP experts are often called in: The consultants from Grandconsult, together with their colleagues from All for One Steeb, were busy from Friday to Sunday restoring the faulty data sets.
The entire weekend was spent feverishly analyzing, retracing, testing and recalculating before business operations could restart punctually on Monday morning.
Backup is not everything
Such events are by no means isolated incidents from an eventful consultant's life. It is true that backups are part of the basics of every IT operation, just like regular checks of the data backups that have been performed.
In practice, however, the picture is clearly different. Not so much when it comes to backup, but above all when it comes to restore. A lack of resources, for example in personnel, compute or storage, new technologies such as Hana, expanded application landscapes and, above all, Big Data, the massive increase in the daily volume of data, are increasingly ensuring that the decisive steps, the validation of the current data backups, fall by the wayside.
Neither CEO nor CFO nor CIO naturally like to be reminded of such sore points. "Close your eyes and get through it, you'll be fine". is still the motto all too often. And when something does happen, it is usually only talked about behind closed doors - if at all.
Such failures must not leak into the public domain, and reliable figures on the frequency of such events are correspondingly rare. The reasons for so much "discretion" cannot be explained by soft factors such as loss of reputation alone. Increasingly, there are more tangible issues at stake.
Financial institutions, for example, have to take an ever closer look at the IT security of their corporate customers seeking credit. Basel 2 virtually demands it. Those who have to reveal weaknesses here face significantly higher costs for corporate financing.
Regular validation
Backup alone is not everything, however, because without appropriate validation, any backup is worthless in an emergency. In phases of "diving through", even obvious findings take time before they have found their permanent place in IT service operations.
Together with NetApp, the causes, effects, findings, and experiences of such emergency deployments were analyzed in detail, and the appropriate solutions have now been molded into a restore protection plan.
The basic idea is simple: No backup without regular validation, because this is the only way to effectively safeguard IT operations. The Restore Protection Plan therefore comprises three coordinated service levels: "Consult & Design", "Build & Implement" and "Run".
The starting point is regularly the precise recording of the data backup procedures practiced to date. From this, the requirements for the exact scope of services of the restore protection plan are derived and the target processes for backup and restore are designed in detail.
The "Consult & Design" phase optionally includes the selection of a suitable provider, who provides the required backup and restore operating environment from his cloud as a turnkey service, so to speak. In practice, this enables considerable cost, quality and scaling advantages to be achieved.
Moreover, with orchestration expertise that spans on-premise, private and public cloud resources, demo environments can be quickly established at this stage, providing a good basis for the subsequent "Build & Implement" steps.
Focus Restore Processes
Depending on the customer application, the focus here is on services such as the provision of the IT infrastructure, the setup for the initial data transfer, the establishment of continuous backup replication at the specified periodicity and, above all, the actual implementation of the restore processes.
On this basis, including successful tests, the restore protection certificate is later formally accepted and transferred to regular operation. In this "run phase", external service providers such as Grandconsult, together with All for One Steeb, take over the regular implementation and validation of the restore processes.
This is done in two stages: For the purely system-side restore, the backup and log files are always transferred to the restore environment first. The log files are deployed in the same step.
This simultaneously lays the foundations for application-side validation. Is the "Database Verify" really okay? Can the "System LogOn" be successfully executed on the test-recovered system? Do selected transactions and reports deliver the same results as on the production systems?
Only when all these questions have been clearly answered with "yes" is the approval given. Depending on the protection agreement, these steps are repeated regularly. The frequency with which the validations are carried out depends on individual requirements.
The results are recorded in a separate report. Maintenance of the restore environment, as well as adjustments to new requirements, also takes place in this phase. Technologically, the backup and restore environments are based on solutions from NetApp, such as SnapVault and SnapMirror.
This means that data backups can be carried out, swapped out and restored in parallel and without disrupting IT operations during ongoing day-to-day business. NetApp Private Storage, Cloud ONTAP, Flex Clone and FlexClone Split are used to validate the backup.
Which operating scenario?
Expand the backup environment on-site in your own data center? Outsourcing backup to the service provider in the private cloud with or without additional inclusion of public cloud resources? Or even a dynamic combination of everything?
The best operating solution in each case always requires a detailed assessment of the initial situation. This is by nature individual. In general, the following can be stated: The expansion, operation and adaptation of an in-house backup environment, including regularly validated restores, tie up considerable human resources over the long term. In addition, there are fixed costs.
This is not the case when outsourcing to an external provider. Here, billing is on-demand, i.e., usage-based. In addition, regular, reliable service level agreements guarantee the promised services.
Service providers are opening up completely new avenues with the orchestration of dynamic IT scenarios. Such operating solutions are aimed at interaction between on-site IT operations, for example in the customer's data center, and IT operations from the private cloud, for example from the all-for-one Steeb data center.
In the process, public cloud resources are increasingly being added for certain purposes outside of business-critical SAP application operation. This may come as a surprise in the wake of events such as the NSA affair or the current Safe Harbor.
Questions about data sovereignty, data autonomy or data security, for example to defend against cybercrime, are always an integral part of any overall consideration of the value contribution that every CIO and his team must make to business development in times of digital transformation.
Safety aspects are therefore increasingly being examined soberly and analytically as part of a qualified technical and legal discussion. A particularly well-developed legal framework in Germany serves as the background here.
The entire data pool is located in the private cloud, for example, in the data centers of a particularly trustworthy service provider. Their level of protection often exceeds the IT security level of in-house operations due to complex certifications such as ISAE 3402 Type II or ISO 27001.
The compute resources selectively switched in from the public cloud do not hold any data permanently. Such resources are only used temporarily at runtime. This allows databases to be restored, applications to be started up and consistency checks to be performed. Data sovereignty thus remains untouched and always with the customer.
Public cloud providers have become much more sensitive when it comes to data security. On-site IT production facilities in Germany are also increasingly the order of the day. In such dynamic IT operating models, many other services can be used from the cloud.
The spectrum ranges from systems that can be switched on and off very quickly, for example for repair, project, sandbox, test, demo or training purposes, to backup assurance systems that serve as a fallback position in the event of disaster recovery, so that downtimes for the IT systems are effectively minimized.
Backup under new auspices
The advance of Hana and SAP Cloud Solutions in conjunction with conventional SAP landscapes is becoming a litmus test for many conventional IT operating solutions. However, only a few top specialists have yet mastered end-to-end virtualization, automation and scaling within industrialized IT operations from the private cloud under the many new auspices when taking a closer look.
What initially began with large corporations is now increasingly finding its way into the SME sector. Demand for consulting services is growing accordingly. Trusted advisors with a holistic consulting approach to strategic technology issues are therefore in particularly high demand, including the design, virtualization, flexibilization and orchestration of specific reference architectures for the operation of Hana, on-going support for IT operations with service level agreements, and additional management and process consulting.
While Grandconsult advises major corporations on their data centers and operates a Joint Research & Development Lab at SAP in Walldorf with partners such as NetApp and Cisco, All for One Steeb is one of the leading private cloud providers for SAP operations in the managed services market.
Webinars: Learn more about the Restore Protection Letter
Backup alone is not everything, because without appropriate validation, all backup is worthless in an emergency.
This is precisely where the Restore Protection Letter from Grandconsult and All for One Steeb comes in. A series of webinars will provide more information about typical deployment scenarios in the company's own data center, in the private cloud and in the public cloud. Dates and registration at www.all-for-one.com/events
