It's common for data backup operations to grow with operating environments, applications or specific business applications. However, backup operations that haven't been examined in more than two years are likely in need of modifications. There are several strategies that can help streamline your backup and recovery process. However, a lot depends on the type of operation, the size of the company, number of locations and your business requirements. In other words, one size rarely fits all.
Look first to operational and data classification issues, because these should drive the backup and data retention policies across your environment. The following recommendations flow in a top-down structure, starting with the items that affect the most data and have the largest impact for the least amount of cost.
Align data protection with business needs
The most important aspect of protecting data is ensuring that the level of protection matches the business requirements. This includes what type of data, how often it's backed up, how long the data is retained and other aspects that can affect the service level. Service levels for data protection are typically measured with recovery point objective (RPOs) and recovery time objectives (RTOs). These measurements indicate how much data can be lost, and how long it takes to recover information. The goal isn't to minimize these, but rather to align them with the business needs.
It's common for IT to provide similar levels of service across multiple applications and data types. However, this strategy will deliver too low a service level for business critical data, and too high a level for less important data. Some data may need minimal protection levels, and some may need no protection at all. By first understanding what service levels are required, it's possible to reduce the total amount of data being protected, providing better protection for important data without increasing operational or capital costs.
The use of data retention policies, and more importantly, the use of policies that are consistent across the enterprise, can go a long way towards streamlining the backup and recovery processes, too. Performing a full backup on home directories five times a week may have a business requirement, but often it occurs simply because "that's what we've always done."
Centralize your backups
This includes centralization within a single site, across sites, across operating environments and any other arbitrary barrier. The backup and restore audit should look across the entire enterprise and include all platforms, applications, business applications and business groups. Minimize or eliminate equipment at remote sites, especially equipment used solely for backups. Diagnosing, maintaining and supporting remote equipment is difficult under good circumstances. Eliminating equipment can go a long way toward streamlining the overall data protection process.
Adopt a disk-based backup strategy
Much of the time spent on backup procedures involves managing tapes, including loading, unloading, labeling, moving tapes offsite and bringing older tapes back onsite for reuse. Using disk storage to improve backup operations can help alleviate the issues around media management. Disk storage for backup and restore is often deployed as a VTL or as a disk cache in a D2D2T strategy. Disk array technologies that provide point-in-time copies of data are ideal for backup purposes, and may be leveraged for restoration as well. Ensuring application consistency with the copy of data has been a more difficult issue, but one that's also being addressed by a number of technologies, vendors and product offerings. These options will likely increase the speed of backups -- and especially recovery -- but often ignore the larger issues of determining what should be backed up and how often. Getting the business requirements right up front will help streamline the rest of the process, ensuring that data that's placed on secondary disk storage belongs there in the first place. While disk is more costly to purchase, eliminating tape partially or entirely can streamline the backup and recovery processes. However, it is essential that current, offsite copies of data are maintained. This may occur via VTL or disk replication, electronic vaulting or tape for offsite storage of data.
Using disk to enhance backup and restore operations isn't a panacea, and can increase operational complexity. The cost of disk storage may be justified through the delivery of higher service levels, including reduced backup time and faster restoration.
Consider a backup service
You might also consider using an online backup tool, such as EMC Corp. Avamar and MozyEnterprise, Iron Mountain Inc. LiveVault and Seagate Technology EVault, along with other alternatives. For backup sets less than 1 TB, this option is likely to be a good alternative. For data sets that are larger, online backup can be a good option, but variables such as rate of change, connection speeds and other factors become important to determine whether or not it is viable. Companies that have numerous operational centers storing data, or those that have minimal IT staffs often find that online backup is a good option.
About this author:
Russ Fellows is a Senior Analyst with Evaluator Group. He is responsible for leading research and analysis of product and market trends for NAS, virtual tape libraries and storage security. He is also the primary analyst for coverage of selected open-systems arrays and virtualization products.