News Stay informed about the latest enterprise technology news and product updates.

Back up expert fields sWM users' questions

Users tossed their toughest backup problems into Liam Scanlan's court. Were this backup expert's solutions slam dunks? Get into the game with this Q&A.

Even though more than half of all routine data backups fail, most businesses don't act on the backup problem until disaster strikes.

A survey by Milford, Mass.-based research firm Enterprise Storage Group revealed that 40-60% of daily backups contain errors. Surprisingly, this high margin of error doesn't seem to alarm IT managers, said Bocada Inc. backup expert Liam Scanlan.

"Few enterprises take measures to identify causes for routine failures until after they experience a catastrophic failure," said Scanlan, product development vice president for Bocada, a maker of backup and reporting tools. Rather than wait until it's too late, Scanlan advises businesses to create policies and practices that will ensure back up reliability.

Scanlan's advice was in great demand during a recent searchWindowsManageability Live Expert Discussion Day. Here are his responses to the questions that IT professionals submitted to the Management Tools Discussion Forum during the event.

User: Which backup application is the most reliable?

Each of the dozens of backup software products have their individual strengths. Many companies discover that their optimal backup strategy includes the use of more than one backup product. Some products are designed for large amounts of centralized data, others are optimized for a large number of broadly distributed protected resources. There is not any "most reliable" or "best" backup product. Knowing your own requirements ahead of any product selection will make that selection process easier, and may significantly increase your chances of implementing a backup solution that works well for you.

User: What are the requirements administrators should consider?

  1. What am I trying to backup? Is it an integrated application like an SAP system, or is it hundreds of Word documents on a server?
  2. Will the data be in use or changing as the backup process is taking place?
  3. Where does the data live? On remote servers, or locally?
  4. Do you have a storage area network (SAN) already in place? (Some backup products have a reputation for taking better advantage of SANs)
  5. How valuable is the data?
  6. Do you expect to do many recovers? How fast will the data be needed?

User: What tips do you recommend to make the backup process more cost-efficient?

You must first define your business requirements, including tolerance for risk. The more accurately and quickly you need to be able to restore data, the more redundant and expensive the storage solution required. Generally, the more complex and expensive the storage solution, the more difficult and expensive the day-to-day management is. Human resource is often the greatest, and least measured, expense. Daily evaluation of the reliability and efficiency of your data backups will greatly reduce the overall cost of your storage solution, no matter how small or large your environment. This creates an up front commitment and investment, but it is nominal compared to the cost of recreating even 20MB of data (estimated to cost more than $100,000).

User: Are there tools that will help administrators know how efficiently back up is being performed?

There are many tools that help evaluate your backup and overall storage infrastructure for reliability and efficiency issues. Some solutions can even help you automatically optimize the use of your storage devices and drives.

User: My boss wants me to implement a disaster recovery strategy. Where do I start, and how can I achieve this?

I am a great believer in dry-runs. I would want to be able to completely reconstruct a representative portion of the entire system that would need to be reconstructed in the event of a complete catastrophe. Going through this exercise can give you a great idea about how long it might take to do the entire job, expose dependencies and permit you to train others on the recovery in the event you are unavailable. The more "real" this dry-run is, the more insight you will get and the more comfort you and your boss will have that your plan will work.

User: How frequently should dry-runs be performed?

Recovering small amounts of data now and again is much different than reconstructing the entire IT infrastructure. In addition to random monthly restores, we recommend daily evaluations of backups to identify and correct errors, and quarterly restores of select significant systems in their entirety. All of these elements are important because the challenges following a complete disaster involve not just restoring live data, but also the PCs, servers, and, in some circumstances, every nut and bolt down to the wires under the floorboards. Added to that, the people who put that together in the first place may not be available to help you.

User: What are some statistics on when backup products are the cause of backup failures, as opposed to human error?

In my experience, well in excess of 50% of failures are caused by issues that cannot be controlled by the backup software. Of the remaining, the great majority are caused by humans. Even the remaining small percentage of potential errors are often not attributable to the products themselves, but are due to the incredible complexity of the IT environment. No matter what solution you deploy, proactive management is required to keep your storage solution up-to-date with your changing IT environment. If managed carefully, companies can greatly reduce the number and impact of failed backups.

User: What are some examples of how backup errors occur?

Common examples of issues that cannot be controlled by backup software are tape or media errors, network errors and corrupted or locked file errors. Errors caused by humans are: failure to run a backup, failure to schedule a backup, failure to properly register a protected resource or add it to a backup schedule, failure to proactively manage errors and system efficiency, and even failure to timely upgrade systems. Configuration errors, underpowered hardware and lack of seamless compatibility with certain networks, operating systems, applications and file types, and storage devices greatly impact the reliability and efficiency of certain backup solutions.


Find the rest of the Discussion Day questions and answers here

Read about Bocada

Dig Deeper on Windows Server storage management

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.