Backup redesign: A top priority for IT managers (Part 1)

Person Marking in a Checkbox 5722106Backup redesign continues to be toward the top of most analysts’ lists for 2013 IT priorities. I’ve talked a lot about some of the catalysts behind this trend like data growth, big data, VMware and software defined storage. With IT managers redesigning, the incumbent enterprise backup vendors have a lot of motivation to offer innovative solutions that are a bit ahead of the times. The leaders have all placed strategic bets on what the winning formula will be. I discussed these bets in my post “Forrester’s take on enterprise backup and recovery.”

For its part, IBM is being quick about helping IT managers redesign. The help starts with a clear understanding of the economic benefit a redesign can bring. After all, in today’s environment few IT managers make technology moves simply for the sake of technology. Storage is about economics. I discuss this more fully in my post “Does trying to find a better economic approach to storage give you ‘Butterflies’?” But there is still efficient technology that enables these economic savings, and the person in IBM who is ultimately responsible for the technology in IBM Tivoli Storage Manager (TSM) is the product manager, Dr. Xin Wang.

Businesswoman touch the virtual cloud buttonRecently I spoke with Xin about the important shifts IT managers are facing and how she is helping IT managers reimagine backup.

The Line: Xin, I’m going to start with the “Dr.” part of your title. Should folks call you the Backup Doctor?

Xin: (laughing) Well, I don’t know about that. I’m actually a doctor of Applied Physics. One thing that drove me to earn a PhD and has moved me ever since is that I love to learn. I started my career in IBM hard disk drive research, spent some time as a storage software developer and development manager, and have now been working with backup clients as a product manager for several years.

The Line: Wow, I could probably do an entire post just on your career. But let’s stay focused. What have you learned about the challenges IT managers are facing and this whole backup redesign movement?

Labyrinth - business conceptXin:  It’s interesting. The challenges aren’t secret but they carry big implications for backup. Data is growing like crazy; that’s no secret. But it is now so big that the old method of loading an agent on a server to collect and copy backup data over a network to a tape isn’t keeping up. So IT managers are redesigning.

And what about servers? Servers aren’t servers anymore. Thanks to VMware, they are virtual machines that come, go and move around in a hurry. Traditional backup is too rigid. So IT managers are redesigning.

Administrators are changing too. The generation of backup admins who grew up tuning the environment is giving way to a new generation of backup, VMware and cloud admins who need much more intuitive and automated management tools. And so IT managers are redesigning. (Editorial comment: I discussed the change in administration in my post “Do IT managers really ‘manage’ storage anymore?)

The Line: Okay, I think I’m seeing your trend. IT managers are redesigning. And it seems like you’ve got a clear idea of why. Can we take your list one at a time? I think my readers would be interested in what you are doing with TSM in each of these areas.

Xin: Sure, that makes sense.

Check back for part 2 of the interview in which Xin shares her near term plans for TSM. If you have questions for Xin, please join the conversation by leaving a comment below.


About Ron Riffe

IBM Manager, Software Defined Portfolio. Adoring husband, proud daddy, imperfect but redeemed. I have a thirty year background in the storage industry having held positions as both a consumer and a peddler and in roles spanning from administrator, to thought leader, to strategic planner, to senior manager. I’m also a passionate leader, speaker and author. My perspectives are my own.
This entry was posted in Backup and Recovery, VMware and storage and tagged , , , , , , , , , . Bookmark the permalink.

4 Responses to Backup redesign: A top priority for IT managers (Part 1)

  1. Pingback: Backup redesign: A top priority for IT managers (Part 2) | The Line — by Ron Riffe

  2. Pingback: Backup redesign: A top priority for IT managers (Part 3) | The Line — by Ron Riffe

  3. William Mansfield says:

    Ron, thanks for posting this conversation. What I’m missing from it is much of anything about restore. The ability to reliably restore data from well defined recovery points within a (short) recovery SLA is pretty much an afterthought with most backup products. With TSM, for example, there is no capability for tracking to a restore SLA. A system’s data might be spread across 100’s of tapes, and TSM is perfectly happy. There isn’t even a capability to diagnose the problem – unless you attempt a restore you can’t identify the collection of tapes a restore point’s data are on. I would like to see a much greater use of proactive analytics (Cognos anyone?) to predict the recovery capability for systems of different criticality, and to autonomously adjust backup techniques and storage management to keep within a given recovery SLA.

  4. Xin Wang says:

    William, thanks for your comment on restore and SLAs. I agree with you that recovery service level agreement (and tracking to it) is becoming more important not only for managed service providers, but also for enterprise users that cater to various organizations within the company. Tracking to SLA and proactive analytics to provide guidance on backup and recovery capabilites will be one of the key areas we want to focus on with Cognos or the Operations Center going forward.

    To address your point about restoring data when a system’s data is spread across many tapes, there are a few ways to find out which tapes are used for the system’s data (Query NODEDATA came to mind ) and if the tapes are in normal state or offline. Another way for diagnostics is to use GENERATE BACKUPSET. This isn’t really the intended purpose of the command, but the fact is, it will read all the volumes containing active data for a node as it copies the node’s data into the new backup set. As a preview, this might be one of the best ways to make sure all of the required data is available and accessible short of actually performing a restore. And, as an added benefit, if the backup set is generated to a point in time corresponding to one of those ‘recovery points’ mentioned in your comment, then restore could be performed directly from the backup set, bypassing the need to mount many tapes that the client’s backup data was spread across.

    But your point of tracking recovery SLA and predictive analytics is well taken, and is consistent with our product vision. We do plan to provide prescriptive guidance on the best way to use technology, and to be a trusted advisor that helps customers navigate the complexities of their environments in the future. Simplifying the overall end user experience with analytics and automation is our goal.

Join the conversation!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s