The virtues of backup disaster recovery

The virtues of backup disaster recovery

• 1.
  The Virtues of BackupDisaster Recovery April 2017
• 2.
  INTRODUCTION For years, companieshave realized the loss of data can be expensive. As a result, companies consistently backed up important data. Organizations have taken pains to secure backups, including offsite backups. This ensures data is protected against a large scale disaster such as a fire or tornado. As more functions of business have become dependent on computer technology, organizations are investing in a backup disaster recovery plan. While this continues to take into account the financial implications of lost data, it also takes into account the expense of downtime when some or all of an organization cannot function during the process of traditional data restore. This downtime has direct costs, such as employees unable to function, as well as indirect costs, such as the loss of clients or a degraded reputation due to a lack of response to requests. This white paper will show how a backup and disaster recovery plan takes into account the loss of downtime and how it can also save your organization frustration. BACKGROUND Since the late 1980s software has been around to backup digital files. This prevents data loss when a file is accidently deleted, when a file becomes corrupted, or when the platform hosting the data has a failure. There are, however, other considerations for backups, especially those of shared systems. Servers are the most common. Failures of servers have the potential for a much larger impact on an organization. While servers are being replaced and data is being restored, there is a loss of productivity since employees are unable to access the data before a server is restored. Restoring a server from a hardware failure has the added complication that rarely does anyone, companies, government entities, non-profit organizations, educational institutions, and the like, keep spare server hardware on hand. Add to that computer retail establishments don’t stock server hardware either. That means getting replacement parts is at least an overnight wait, if not longer. That downtime can cripple an organization.
• 3.
  When server hardwarearrives, it may take several hours to reinstall Windows (or whatever your operating system), install the various programs it was running, and configure all its settings. These all must be done before you’re ready to restore the data needed by your users. In cases involving databases, which includes email servers, the software you’re installing must be the exact version with the exact same updates. Settings must also be precisely the same. Otherwise it won’t work. Getting all these versions and settings precisely the same is not a trivial task. To aid with this is a special backup called an image – a copy or picture of everything on the hard drives on your server. An image backup will eliminate many of the steps otherwise required to install Windows programs and adjusting precise settings. Restoring an image backup to its hardware means the new hardware needs to be similar to the old hardware. There may be machinations to be able to restore an image. This includes starting your server from a CD and loading the right hardware drivers – all necessary steps before an image can be restored. If you have never restored from an image backup, you will likely run into obstacles causing delays. While having an image backup is important, that alone is not enough ensure server functionality is restored in a timely fashion. SOLUTION An important key to a solid backup and disaster recovery is the plan. An image is the best solution. Many small to medium-sized businesses (SMBs) chose not to use an image, rather rely on direct backups of databases and files, to save on the costs of storing the backups. Although it is possible to do this, it means having precise notes of everything installed: Windows version and all updates, SQL or Exchange and all updates, precise notes on the configuration and keeping notes up to date as updates are applied. Although there are higher upfront costs for an image based backup, the savings in hard and soft costs is, in the event of a restore, significantly higher. Another helpful piece of the plan is software that can modify the hardware drivers automatically (hardware drivers being the software that connect your server to its specific hardware platform). The original hardware platform maybe different from a restored environment. Thus when starting the server for the first time, it’s important that the software will be able to change the hardware drivers on the fly.
• 4.
  Some software solutionscan take a backup image and start it in a virtual environment, running it as a gust on an existing server or a virtual server in an Internet hosted site. Some solutions that create this image, ready to run in a virtual environment, go a step further. On a regular basis, such as nightly or weekly, the backup software can automatically start/load/run the image just to confirm that it works. Standard backups should also include a backup image that is stored on an Internet site. In the case of a site disaster, such as a fire or tornado, the image can still be made live online, in a virtual environment, but in this case running on an Internet cloud site. It may be a little slower, users connecting to that server over the Internet, but the sacrifice of speed is well worth it for the other benefits gained. One of the biggest vendors of cloud computing resources is Amazon. Many companies regularly use Amazon cloud storage space. In the event of a site disaster, an image can be turned into a virtual server, running from a virtual environment on Amazon’s computing resources. This allows you to restore the functionality of critical resources very quickly. CASE IN POINT Emerson Technologies is an IT consulting firm for organizations of all sizes, non-profit organizations and for profit companies. One such client had four physical servers. In the fall of 2016, this company’s SQL database server had multiple simultaneous hardware failures that impacted half of the organization. That server was backed up with StorageCraft ImageManager. It maintains a current image of the system with versioning on its files. The backup data was being kept on a Network Attached Storage unit (NAS) and was further uploaded to Amazon cloud space. When the failure occurred, Emerson Technologies determined which other platform had hardware resources available to run the SQL database server in a virtual environment. It took an hour to copy the backup image of the server to the available hardware platform. From there we started it running as a guest. The SQL database server functioned in this virtual environment without any trouble until replacement hardware was ready days later. Downtime for such an event was quite minimal; less than two hours rather than 2 to 4 days. The client easily saw the value in the backup disaster recovery plan we had in place.
• 5.
  CONCLUSION There is moreto having a solid backup solution than just software running and telling you that backups have been successfully executed. An offsite component is essential – now easier than ever with increasingly faster Internet connections. Just as necessary is a plan of action for both backups and restoration. Of paramount importance is to test your backups with a full restoration of your system. This is the only way to identify problems with your process and insure your plan will work when needed. When your systems are down is not the time to learn that your process doesn’t quickly and fully work. Testing the process by running a restore should be done regularly – monthly or quarterly. When creating your backup plan be sure to consider not only the value of lost data but also the value of downtime. A solid IT strategy should take all these factors into account. It’s not just a backup plan, it’s a backup disaster recovery plan.