Despite all the hype around the use of disk-based data backup and recovery, tape storage is still the primary target for most backups. Not only is the most data stored on tape, but most data is backed up directly to tape without using disk as a buffer.
With this in mind, what does one do to make the performance of their tape drives better?
Know thy tape drive
The most important aspect of obtaining good tape performance is to understand the tape drive to which you are backing up. Modern tape drives are streaming tape drives, which means that they are designed to transfer data at a certain rate, and you need to know what that rate is (or what those rates are) in order to keep that tape drive happy, and keep it streaming during backups and restores.
So start by learning all you can about the tape drives that you use. The first question you need to learn is the tape drive's maximum native (uncompressed) transfer rate. For example, an LTO-4 tape drive's maximum native (uncompressed) transfer rate is 120 MBps. Once you know this value, you must determine what the drive's minimum native transfer rate is. This may take a bit of research, but these numbers should be in the documentation. If not, you may have to contact the company from who you purchased your drives.
For example, LTO-4's minimum transfer rate is about 23 MBps to 25 MBps. That means that an LTO-4 cannot write slower than 23 MBps to 25 MBps. If it appears to be doing that, it is actually writing short bursts of 23 MBps to 25 MBps, then backhitching to write another short burst at 23 MBps to 25 MBps. You should also be aware of any steps in between, as multi-speed tape dri...
To continue reading for free, register below or login
To read more you must become a member of SearchDataBackup.com
ves like LTO-4 often have two to three intermediate speeds between their minimum and maximum rates. Again, this will take more research and hopefully should be found in the documentation. Once you know these numbers, you should be aiming at the maximum transfer rate, but anything less than the minimum rate is going to give you nothing but problems.
Know thy source: data transfer rates and compression rates
You need to know two things about the source: data transfer rates and compression rates. You need to understand data transfer rates for different servers. Higher end servers with higher end storage generally can supply data at much faster rates than lower end servers. You need to be aware of these different data transfer rates as you design your backup architecture. You will only be able to determine these numbers with testing. For example, if you know that a given client's database can only send data at 10 MBps to 20 MBps, connecting it to Fibre Channel for faster backups won't help, as its speed will be limited from the very beginning.
The next thing you need to understand is the rate at which your data will compress. Some data compresses better than other data, and some companies' data compresses better than other companies' data. You need to find out how well the data in your environment compresses. The reason for this is that you must multiply the compression ratio times the minimum and maximum transfer rates mentioned in the first tip to arrive at new minimum and maximum transfer rates. For example, if you're getting 1.5:1 compression in your environment, your 120 MBps tape drive is now a 180 MBps tape drive. The way to determine your compression ratio is to take a look at how much data you're fitting on tape before you hit the Physical End of Tape (PEOT) mark and your backup software marks it full. If you're consistently fitting 1200 GB on 800 GB tapes, then you're getting 1.5:1 and should multiple your native throughput numbers times that number.
Know thy data path
Once you know where the data is coming from and where it's going to, you need to understand the route it's going to take on its way. Will it travel an Ethernet network? Is this a gigabit network, a 10 Gb network or even a 100 Mb network? If you've got a client that can transfer data at hundreds of MBps, and it's trying to send its data across a 100 Mb network, you're not going to get very far. However, that would be the perfect client to move to LAN-free backups by connecting it to Fibre Channel.
Know thy backup application
This goes without saying, but backup applications can be very complicated when it comes to performance tweaking. Learn the ins and outs of the backup product you're saddled with and use your knowledge to make it do what you want it to do. For example, if you're using IBM Tivoli Storage Manager and you're not familiar with colocation groups, you should be. The same as true of Symantec NetBackup, EMC NetWorker, and CommVault and their multiplexing features. Join a mailing list, forum or newsletter and learn what others are doing to get better performance. Search for the word "performance" in the forums at BackupCentral.com to see what other users have already asked and answered about the subject.
Know thy disk salesperson
It's getting harder and harder to make tape drives happy without first sending backup data to disk. If you're buying tape drives these days and don't have disk as part of your purchase, cancel your order and change it to something that includes disk. Believe it or not, moving the data twice (i.e., backup to disk then copying to tape) can actually be faster than moving it once.
Tape is going to be around for a while, so you might as well learn how to make it happy.
About this author: W. Curtis Preston (a.k.a. "Mr. Backup"), Executive Editor and Independent Backup Expert, has been singularly focused on data backup and recovery for more than 15 years. From starting as a backup admin at a $35 billion dollar credit card company to being one of the most sought-after consultants, writers and speakers in this space, it's hard to find someone more focused on recovering lost data. He is the webmaster of BackupCentral.com, the author of hundreds of articles, and the books "Backup and Recovery" and "Using SANs and NAS."
