7 Ways to Make Backup and Recovery Easier, by Logan G. Harbaugh
As data growth spirals, getting a handle on your backup and recovery processes becomes even more crucial. Without a proper strategy in place for maintaining and securing stored data, and for reproducing data on demand, you could lose business or incur huge legal penalties.
Here are seven things you can do to make your storage environment easier to back up and restore.
1. Don't use one tool for everything.
The best tool for backing up file servers isn't necessarily the best tool for archiving email, for example. Likewise, SAN-based replication doesn't get too expensive unless you're trying to use it for everything. Take a realistic look at the backup requirements for each of your projects--whether it's disaster recovery/business continuity, data archiving for compliance or legal discovery--and use the most appropriate tools for that particular area.
2. Standardize on a few good tools.
Find and use the tools that work well for your environment and make sure that the versions you use are consistent across all of your data silos and backup servers. Standardizing on hardware can also pay dividends, whether you're standardizing host bus adapters for SAN connectivity or tape drives. Having many different brands and types of tape drives with dozens of types and capacities of tape cartridges isn't uncommon, but it introduces one of the greatest sources of complexity in backup strategies.
3. Standardize directory structures on file servers.
This goes beyond putting user directories in the same place on file servers. Standardize and document database installations, Web servers, mail servers and so on. This makes it much easier to ensure that all critical data is backed up regularly without having to do daily full backups of the whole server drive.
4. Consider disk-to-disk (D2D) backups.
D2D doesn't just provide faster backups and restores; it lets you consolidate multiple backup servers onto one storage system, which can then be more easily managed. Disk-to-disk-to-tape (D2D2T) allows for offsite archiving; you can also replicate your backup store to a remote site depending on your requirements for long-term archiving.
5. Consider storage management programs.
Storage management platforms require a considerable investment, not only in software licensing but in the installation and configuration of the framework. In the long run, however, the reduction in man hours required to keep storage organized and easy to back up, and backups verified and organized, will pay off.
6. Don't let inertia keep you from trying new solutions.
Administrators accustomed to tape backups can progress from tape libraries to virtual tape libraries, not realizing that there might not be a need to simulate tape drives if it makes more sense to back up to disk. Even if you're eventually archiving to tape, simulating tape libraries in the intermediate stages may not be necessary. Likewise, instead of looking for a faster way to do what you're doing now, look around at the alternatives--there's a lot of new storage technology available.
7. Try dividing tasks by requirements to make them easier to manage.
Instead of managing all file server backups in one place and all application server backups in another, try to manage all disk-based backups in one place, all replication-based backups in another, all offsite tape backups in another and so forth. This will let you consolidate and simplify backup targets, rather than treating each backup source as unique.
Backup administrators can be conservative, sticking with existing tools simply because they aren't broken yet and are familiar to them. Unfortunately, they often don't discover that those tools aren't functional until an attempt to restore fails. Don't be afraid to reengineer your backup infrastructure. You may not only save administrative costs and speed up responses to user requests for restored files, but also avoid having to explain to the higher ups why the "old, reliable" backup system wasn't.
--Logan G. Harbaugh
Note: to view this in it's original content, click here.
» print friendly version
Here are seven things you can do to make your storage environment easier to back up and restore.
1. Don't use one tool for everything.
The best tool for backing up file servers isn't necessarily the best tool for archiving email, for example. Likewise, SAN-based replication doesn't get too expensive unless you're trying to use it for everything. Take a realistic look at the backup requirements for each of your projects--whether it's disaster recovery/business continuity, data archiving for compliance or legal discovery--and use the most appropriate tools for that particular area.
2. Standardize on a few good tools.
Find and use the tools that work well for your environment and make sure that the versions you use are consistent across all of your data silos and backup servers. Standardizing on hardware can also pay dividends, whether you're standardizing host bus adapters for SAN connectivity or tape drives. Having many different brands and types of tape drives with dozens of types and capacities of tape cartridges isn't uncommon, but it introduces one of the greatest sources of complexity in backup strategies.
3. Standardize directory structures on file servers.
This goes beyond putting user directories in the same place on file servers. Standardize and document database installations, Web servers, mail servers and so on. This makes it much easier to ensure that all critical data is backed up regularly without having to do daily full backups of the whole server drive.
4. Consider disk-to-disk (D2D) backups.
D2D doesn't just provide faster backups and restores; it lets you consolidate multiple backup servers onto one storage system, which can then be more easily managed. Disk-to-disk-to-tape (D2D2T) allows for offsite archiving; you can also replicate your backup store to a remote site depending on your requirements for long-term archiving.
5. Consider storage management programs.
Storage management platforms require a considerable investment, not only in software licensing but in the installation and configuration of the framework. In the long run, however, the reduction in man hours required to keep storage organized and easy to back up, and backups verified and organized, will pay off.
6. Don't let inertia keep you from trying new solutions.
Administrators accustomed to tape backups can progress from tape libraries to virtual tape libraries, not realizing that there might not be a need to simulate tape drives if it makes more sense to back up to disk. Even if you're eventually archiving to tape, simulating tape libraries in the intermediate stages may not be necessary. Likewise, instead of looking for a faster way to do what you're doing now, look around at the alternatives--there's a lot of new storage technology available.
7. Try dividing tasks by requirements to make them easier to manage.
Instead of managing all file server backups in one place and all application server backups in another, try to manage all disk-based backups in one place, all replication-based backups in another, all offsite tape backups in another and so forth. This will let you consolidate and simplify backup targets, rather than treating each backup source as unique.
Backup administrators can be conservative, sticking with existing tools simply because they aren't broken yet and are familiar to them. Unfortunately, they often don't discover that those tools aren't functional until an attempt to restore fails. Don't be afraid to reengineer your backup infrastructure. You may not only save administrative costs and speed up responses to user requests for restored files, but also avoid having to explain to the higher ups why the "old, reliable" backup system wasn't.
--Logan G. Harbaugh
Note: to view this in it's original content, click here.
» print friendly version
