Posted on: 19 06 2023.

Disaster Recovery: Evolution of Solutions and Approaches

Author: Luka Juh, Senior Lead Engineer at Comtrade 360 

Disaster recovery solutions have come a long way since their inception in the 1950s. Back then, backup and disaster recovery were simple and straightforward. Tape drives were used to back up data, and in the event of a disaster, IT staff would physically transport the tapes to a secondary site. As technology has evolved, so has disaster recovery. Today, we have an array of solutions and approaches available for both Windows and Linux servers, ranging from simple file backups to full-scale Disaster Recovery as a Service (DRaaS) solutions. In this article, we will take a closer look at the history and evolution of disaster recovery solutions and approaches, and how they have adapted to meet the ever-changing needs of modern IT Services. 

Disaster Recovery and How Did It All Start? 

Throughout the evolution of disaster recovery solutions and approaches, different eras have presented unique challenges and opportunities. 

Early OS era – In the early days of operating systems, such as MSDOS, the only real option for disaster recovery was using enhanced file copy tools to manually copy the entire system to alternate locations, file by file. 

Early Windows OS era – Disk image and sector-based approaches were quite popular during the early Windows OS era, such as Windows 3.1 and 98. Although they were thorough in maintaining the integrity of the system, the procedure had to be performed offline, which was not acceptable for enterprise purposes. Even sector-based backups could be tricky with emerging disk geometries. As time went on, this method was primarily used for deployment rather than disaster recovery purposes. 

File system backups using OS-owned backup tools, such as MS Backup, also became prevalent roughly at this time, but the real challenge was creating offline bootable media which would drive the recovery phase. 

The advanced OS era, exemplified by Windows 2008 R2, introduced snapshot-based backups for consistent backup versions. Operating system features started to provide special OS modes, which enabled recovery via a live operating system. Additionally, vendors began to prepare versions of offline operating systems, such as Windows Preinstallation Environment. 

The high availability era offers additional options for keeping systems online all the time while still maintaining the possibility of using existing disaster recovery methods to recover individual nodes. 

The virtualization era provides a completely different approach to addressing the backup of virtual machines. While traditional methods can still be used, using the virtualization stack allows virtual machines to be backed up as essentially black boxes at a specific point frozen in time. 

However, backing up virtualization management infrastructure presents a completely different problem that not many can address properly, and typically OS tools dedicated for this purpose are used. 

The cloud era did not contribute as much to disaster recovery itself, but it does provide alternative means for data storage. However, accessing this data during a DR procedure can be a challenge, making all of the cloud services available at recovery time essential. 

Phases of the DR Process 

To ensure a comprehensive disaster recovery experience, the process is divided into separate and distinct steps, including the preparation phase, backup/storage phase, and recovery phase. 

It’s important to note that the disaster recovery process mentioned in the text is specifically designed for physical Windows and Linux servers. This means that the preparation, backup/storage, and recovery phases are tailored to the needs and capabilities of these systems, which may differ from those of virtualized environments or other types of hardware. 

During the preparation phase, disaster recovery metadata is collected and stored for later use in completely restoring the backed-up system. This metadata includes data about the system’s storage (disks, partitions, volumes, etc.), data about the system itself (general computer information, network-related information, etc.), system component information (device driver data including device driver images themselves, etc.), and, optionally, the preparation of the disaster recovery offline/mini operating system image that will carry the entire restore later on. 

The backup/storage phase utilizes the existing backup infrastructure of individual backup products and is essential for storing the previously collected data during the preparation phase. Examples of this include the storing of disaster recovery related data during Configuration Backup, when the backup application deals with non-standard/non-file related types of data that have to be backed up (e.g., Windows registry, NTDS database, etc.) and the storing of disaster recovery related data during System State backup phase (available on newer Windows systems). 

The recovery phase can be performed either on a live system in a dedicated operating system mode, which requires abundant cooperation of the operating system to properly restore inaccessible system files, or via a previously generated offline operating system image (e.g., Windows Pre-execution Environment or other forms of Linux-based offline images, etc.). 

Both disaster recovery approaches have their pros and cons, but the latter approach seems to be typically more reliable, at least as far as Windows platforms are concerned. However, the recovery phase is undoubtedly the most important and most complex phase that requires performing some common tasks, regardless of whether it runs online or offline: 

• Establish a reliable networking environment: A crucial step in the recovery phase is to ensure a reliable networking environment that allows recovery/restore agents to be run. Without this, the recovery/restoration process cannot be initiated or managed. 
• Prepare the temporary environment: The next step is to prepare the temporary environment, which involves setting up the necessary software, configuring the networking settings, and ensuring that the environment is secure and able to run all the essential tools required for data restoration. 
• Prepare the underlying hardware: After the temporary environment is set up, the underlying hardware needs to be prepared. This includes installing any necessary device drivers, updating firmware, and ensuring that the hardware is properly configured for the restoration process. 
• Reconfigure system storage: After preparing the hardware, the next step is to reconfigure the system storage by creating the necessary partitions, formatting the disks, and configuring the volumes to ensure that they are set up correctly for the restoration process. 

• Restore the previously backed up data: Once the system storage is reconfigured, the next step focuses on restoring all the previously backed-up data, including data files, application data, and any system files that were previously backed up. 
• Prepare the recovered hardware for reboot: Finally, it is recommended to prepare the recovered hardware for reboot, especially if it is dissimilar from the original hardware. This may involve updating the registry settings, reconfiguring the boot loader, and configuring any device drivers that are necessary for the hardware to operate properly. 

Conclusion 

The evolution of disaster recovery solutions and approaches has come a long way since its inception in the 1950s. As technology progressed and new eras emerged, disaster recovery methods have adapted to meet the ever-changing needs of IT services. Companies like Comtrade 360 have always been at the forefront, following the latest industry standards and providing cutting-edge services to ensure optimal protection for their clients’ digital assets. 

The current disaster recovery landscape offers a wide range of solutions, including DRaaS, which cater to the diverse requirements of modern IT infrastructures. With a comprehensive process that incorporates preparation, backup/storage, and recovery phases, businesses can ensure a seamless and reliable restoration of critical data and systems. As technology continues to evolve, Comtrade 360 remains dedicated to staying ahead of the curve, offering innovative and adaptable methods to safeguard our valuable digital assets. 

 

Keywords: Disaster Recovery, DRaaS, Windows servers, Linux servers, Backup, File system backups, Physical servers, System storage