By David Fox
Modern LTO tapes and drives provide exactly what is needed for backup of large quantities of data: speed and capacity.
However, unlike hard drives and memory sticks, LTO isn’t plug-and-play. Software is required to make a workflow with which you can secure your data via a ‘backup’.
In this article, we’ll look at the principles behind LTO backup software, so that product offerings from various different vendors can be effectively understood and compared.
We won’t look in detail at LTO tapes and drive hardware. Other than to state that at the time of writing, the current generation is LTO-8 and a tape has a capacity of 9TB and a read/write speed of 360MB/second. Compression may increase both of those figures.
4 Steps to Using LTO Backup Software
1. Backup sources and schedules
As with any data backup, it is first necessary to specify what it is that you wish to backup. Additionally, it is necessary to indicate how frequently you would like the backup to take place.
In most cases the ‘what’ likely consists of all the data your business can’t afford to lose, and then ‘when’ will be every day.
Nothing here is specific to LTO backup software and would be available in any backup tool you might pick.
2. LTO hardware & tape/media management
Broadly, LTO tape hardware falls into two categories, standalone (or tabletop) LTO drives or LTO libraries (or jukeboxes/autoloaders). The former requires the manual insertion of tapes, the latter comprise many tapes and LTO drive(s) in a unit. Tape libraries have robotic mechanisms that move tapes, as controlled by the LTO backup software.
When configuring backup software, one is not generally concerned with individual tapes. Rather one is interested in some grouping of several commonly called a ‘storage pool’ or ‘storage set’.
Different software vendors will use different terms for the same concept, we will use ‘pool’ here. A pool with one LTO-8 tape will have a capacity of 12TB. Add 9 more tapes and our 10-tape pool now has 120TB of backup capacity.
LTO Backup software will allow the creation of these pools and allocation of individual tapes into them to add capacity. The allocating of tapes requires that they are physically available to be written and identified. When using tape libraries, barcodes are attached to the spines of tapes to allow automated identification.
When we wish to perform a backup, the target storage is addressed via the pool, rather than individual tapes. In addition, the LTO backup software will be able to send the necessary commands to the tape library to have the required tape moved into the tape drive as required.
As one tape fills up, it can be replaced with another from the pool. Contrast with the use of a single standalone drive where the software must prompt the user for which tape to insert into the drive.
3. Indexing of backed up data
Another important job performed by backup software is keeping track of files being backed up. This usually happens by means of an embedded database, updated while backups are taking place, and made available to browse when data needs to be selected for restore.
This database stores information about each file being written to LTO tape. Filename, modification-date, size, original location etc. All this information about the files is necessary to detect if a file has changed the next day when the backup next runs. The index will also track which LTO tape a file has been saved on, and specifically where on the tape (a block number).
The index database itself should be protected by itself being backed up. Perhaps both to the LTO tapes themselves, and another computer. The index is important since it is required to make sense of the data that has been saved to the tapes. In the event of a loss of hardware due to an accident or disaster, the index must first be recovered from the tapes before data can be restored.
4. Restoring files and folders
To be able to recover either the entire backup or a subset (perhaps a single file that someone deleted), it’s necessary to use the index to locate what is needed. The backup software will generally present the index in such a way that it can be browsed (just like a filesystem) or searched. There is also a time element when looking inside the index, a given folder will have been regularly saved, and so it’s necessary to view the contents of that folder at different points in time.
Ultimately, this browsing, searching and locating the correct point in time will allow a selection to be made, of files that the user wishes to ‘restore’ back to disk storage. The restore task can then be executed and the LTO backup software will automate or prompt for tape insertions as required to recover the data.
Conclusions
LTO tape offers excellent speed and capacity, but to be used effectively for data backup, additional software is required.
Such software will allow users to create automated workflows that regularly save copies of important files/data to tape. The tapes themselves can be managed automatically, and the provision of a database, allows indexing of everything that has been saved. The restoration of data is then a simple matter of interrogating this index and deciding what needs to be restored.
Archiware P5 Backup is such a product that employs all the methods and techniques explored in this article, with excellent LTO support across single drives through small and large tape libraries.
