Jeff Boles, senior analyst with the Taneja Group, discusses the benefits and drawbacks of file virtualization in the SMB space. His answers are also available below as an MP3.

Table of contents:

>>What is file virtualization and how is it accomplished?
>>What products enable file virtualization?
>>What are the other benefits of file virtualization?
>>What are the drawbacks of file virtualization?
>>If you deploy file virtualization, is it hard to back out?

What is file virtualization and how is it accomplished?

File virtualization has been around for several years. One of the things I'd like to clarify is that our perception of the file virtualization market has shifted. If you go back, we were always talking about file virtualization in the context of global name spaces and complex unification of storage resources.

We've since come to our senses and realized that file virtualization is a valuable technology that can enable true data mobility in any size business. It's not resolutely dependent upon global namespaces. So a number of vendors in the marketplace have file mobility-oriented solutions to deal with issues around file migration when you're replacing storage systems that clients depend upon.

You're either trying to replace and migrate to a new storage system, enable better data migration across tiered storage or move data back and forth across different systems and locations as your needs change.

These solutions don't necessarily require global name spaces even though most of these vendors typically bring a global name space-type solution to the table. But, file virtualization today is about migration and data mobility.

Can you talk about some of the products on the market that enable file virtualization?

With the introduction of auto virtualization on the SMB side, a simple and straightforward data solution, there's now a solution for customers of every size. Frankly, the other guys in the market have always been a little bit bigger in their orientation. But those solutions include EMC Corp. Rainfinity.

Rainfinity actually has a suite of three different products to approach the different aspects of file management. One deals with file migration, one deals with file mobility and one deals with global name spacing. So they can address your needs from a variety of directions.

Brocade Inc. has long had a DFS-integrated StorageX solution, and earlier this year, they introduced a granular file movement engine that gives you data mobility without all the dependencies on name space. Acopia, which was recently purchased by F5, has long been recognized as a leader in file mobility, migration and virtualization. So there a number of solutions out there.

What are some of the other benefits of file virtualization?

There's interesting opportunities around file virtualization, particularly in the long term. There's integration of technologies with enterprise applications or business needs in interesting ways. Some of the vendors aren't even on the radar for U.S. customers, particularly SmApper Technologies in Europe, which has more of an application integration orientation.

Some of these vendors are more focused than others on taking information that belongs to an employee that has been fired or moved to a different business unit, and moving it somewhere else. Maybe you need to move it to an archive, maybe you need to just move it to a different location or maybe you need to remove some types of data and add others.

So there's a tremendous integration opportunity with the number of technologies, but when it comes to file virtualization, we're just not seeing people do that in really interesting ways yet. We can even think in a security context, looking way down the road map, in terms of classification-type technologies, understanding the content of the file and then taking some action on it. File virtualization might play a big role there because it could be the gateway to access your file infrastructure, regardless of where the data is.

What are the drawbacks of file virtualization?

In some situations, we still connect file virtualization with global name spaces. That can end up being a complex implementation, depending on what the solution is if you're trying to do both projects at once.

Another challenge is that you really don't know what a particular vendor's approach is to long-term innovation with other services in your business. So you've got to be able to gauge if file virtualization is important to you because you see long-term potential to make your business more efficient around automating data management. Or, is it more important to you in the project context where you want to do a NAS migration kind of thing.v One dimension more than another will affect your product selection. It's all going to be a guessing game about whether the vendor is able to innovate in the long run to support your desires for business process automation or whether they're just going to be a tool set to help you deal with sticky issues.

If you deploy file virtualization and aren't satisfied with it, is it hard to back out of?

Again, that depends on the implementation and that's sort of a buyer-beware situation. You have to look at each of these tool sets when you're using them for tiered storage or if you're using them for business needs responsive of data mobility. You need to be aware of what you're getting yourself into in terms of long-term commitment and always have a plan for backing out, unless you're sold that it is your long-term solution.

So you either need to vet very well upfront to know that you're going to get for performance, ease of use and everything else up front from your file virtualization solution. Or, you just need to enter the solution gradually enough and have a roll back plan to get out. That really comes down to decent management practices around the tool set.

You can create a pretty hairy mess as you move data around if you use file virtualization half heartedly. So you need to have some scope in mind when you engage these tools. Think in terms of only using these tools for data migration and rigorously tiering this type of data, so you know that you can back out if you need to. Anything that you depart from this practice on, document it, be aware of it and know that you can unstructure the environment after you've deployed these tools.

Some vendors are better than others at making sure that they're not forever connected to your infrastructure and that you can easily back out of them. That's worth having as an evaluation criteria and it's definitely a differentiator between some of the solutions.

Jeff Boles is senior analyst with the Taneja Group. Jeff has 20 years of IT experience as an end user, and today remains hands-on with technologies as the director of Taneja's Technology Validation testing services.

This was first published in December 2008

Back to top