Storage virtualization: Why aren't the big guys talking more about it?

Storage virtualization: Why aren't the big guys talking more about it?

Summary: One of the fastest growing segments of the virtualization software market between 2006 and 2007 was storage virtualization software. As organizations adopt a more virtualized approach to their IT infrastructure, it makes sense that they would also separate the storage function from other parts of the infrastructure.

SHARE:

One of the fastest growing segments of the virtualization software market between 2006 and 2007 was storage virtualization software. As organizations adopt a more virtualized approach to their IT infrastructure, it makes sense that they would also separate the storage function from other parts of the infrastructure. Why, then, do the big folks in virtual machine software spend so little time rounding out their stories?

Let's run through some of the players in the virtual machine software market and review where they stand on the subject.

  • Citrix/XenSource mentions virtual storage as an important part of a virtualized environment but, refers to its friends and partners in that arena rather than spending much time on the topic itself.
  • Microsoft, on the other hand, a large network of partners and alliances in this area. It's pretty clear that Microsoft is also letting others play the music here.
  • Sun Microsystems offers virtual machine software and an array (pun intended) of storage systems and software. They're one of the few "one stop shops."
  • Virtual Iron spends a great deal of time pushing the "server consolidation" button and talks up its relationship with FalconStor and Snapserver as an important part of the story.
  • In the case of VMware, it's pretty clear that their parent/largest stockholder/good buddy EMC's major added value is storage virtualization. It's quite possible that VMware has chosen to focus its marketing resources on its virtual machine software, virtual machine migration software, management software for virtual resources and leave the focus on virtual storage to others.

As organizations develop their plans for a more virtualized environment, they would be wise to consider a number of topics, not just virtual machine software. The topics that ought to be included at how peope are going to access virtual resources, how applications can be "freed" of being tied to a specific instance of an operating system, where and how processing is going to be accomplished, where and how applicaitons and data are going to be stored, how resources are going to connect to one another and, most importantly, how the organization is going to be able to manage this increasingly complex environment.

Topics: Storage, Hardware, Software, Virtualization

About

Daniel Kusnetzky, a reformed software engineer and product manager, founded Kusnetzky Group LLC in 2006. He's literally written the book on virtualization and often comments on cloud computing, mobility and systems software. In his spare time, he's also the managing partner of Lux Sonus LLC, an investment firm.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

4 comments
Log in or register to join the discussion
  • IBM's P-series virtualization in AIX 6

    You should have a good look at IBM AIX 6. Storage and network virtualization already exists with Virtual IO servers in the P5 series servers.
    Application level virtualization occurs in Workload Partitions (WPARs). This frees the application from the LPAR (logical partition) holding the operating system.
    Norm_z
  • Simple...

    Virtualization raises the support structure exponentially in terms of expertise and complexity. The return is minimal and doesn't apply to all situations. While the return may seem good or exceptional if presented in the right method, it's true return is diminished by the complexity it adds, and the downtime accrued by simple single point failures.

    Virtualization is a very good tool, but is still a long way from being a viable option for each and every case.
    Narg
    • So wrong...

      You couldn't be more wrong about the ROI of storage virtualization and partially wrong about it being a viable option in each and every case.

      Point #2 first: You are correct - for some very small shops or shops with very simple storage needs, a lot, if not most, SV products don't bring much to the table when you look at their cost. Also, if you're in a one man shop then yes, it could be too complex for one support person to handle. However, SV products and technologies have been around for many, many years and have filled just about every niche out there. Many vendors (NetApp, FalconStor, etc.) concentrate on making their products extremely easy to use.

      Point #1: The ROI for SV products is measured in different ways. Some bring HA to the table, others focus on maximum utilization, most have some sort of (remote) backup & recovery, point-in-time, etc., etc. To say that SV's true return is diminished by single points of failure indicates a deep misunderstanding or lack of knowledge of the technology and products. If you approach storage virtualization and networking with the same care and attention you give your servers and ethernet networking, you will NOT build single points of failure into your infrastructure.

      I will partially concede your 'complexity' point - guys like me have made a good living in the storage business for many years because sometimes storage networking and virtualization can be hard. But like I said earlier, a lot of SV vendors recognize this and purposely design their products to be simple to use.

      Do some research, explore the SAN & SV worlds, talk to shops of all sizes and I think you'll see that most SV technology provides a sizeable ROI and doesn't have to be overly complex. Then ask yourself this question: How could virtualization vendors stay in business if they didn't provide value and ROI on their customer's investments?
      psyclone56
      • No technology can be applied in all cases

        Thanks for the run down describing your thinking on the topic. As one might expect, I don't believe that any specific technology is applicable in all cases, for everyone, everywhere. That certainly wasn't the thought behind the post.

        The post was meant to offer one possible vision of the future and how it is quite possible that IBM was thinking along the same lines when it developed that product.

        Dan K
        dkusnetzky