Home | Uncategorized | Enterprise Computing: A Myopic View of Unified Storage

Enterprise Computing: A Myopic View of Unified Storage

0 Flares Twitter 0 Facebook 0 Google+ 0 StumbleUpon 0 Buffer 0 LinkedIn 0 Filament.io 0 Flares ×

I’ve only just stumbled on Kostadis’ video on Unified Storage.  In case you’re not familiar with the concept, here’s his definition:

  1. A piece of hardware that has CPU, Memory and disk
  2. That supports FC, iSCSI, CIFS and NFS
  3. That has a common management console for all storage functions that are not protocol specific
  4. Has a single replication mechanism that is independent of protocol

Quite how and why this narrow definition should define something that is classed as unified, I’m not sure, however if you watch the video you’ll see it’s a direct attack at Netapp’s arch nemesis, EMC.  Does this mean Unified Storage is a real concept or just an attempt to have a dig at the competition?

Here’s a few thoughts:

  1. How many (persistent) storage devices do you know that don’t contain CPU, memory and disk?
  2. What have FICON, ESCON, Infiniband, SCSI, AoE and FCoE (and others) done to deserve exclusion from the term “Unified”?
  3. Why permit exclusions from a common management console?  Surely common means it does everything?  Otherwise it’s not common.
  4. Why is  a single replication mechanism so superior?  Isn’t choice a good thing?

There are other storage vendors out there.  Some of them even use the Unified moniker too – with just cause.  So here’s my more generic definition of Unified Storage:


  1. A hardware device dedicated to data storage.
  2. A device which supports all commonly used protocols, including Mainframe and Open Systems standards.
  3. A device which enables *all* configuration to be performed through a single management interface (regardless of whether multiple interfaces are supported)
  4. A device which provides consistent performance/throughput, regardless of the protocol used.
  5. A device which scales.


Now does that actually exist?

About Chris M Evans

Chris M Evans has worked in the technology industry since 1987, starting as a systems programmer on the IBM mainframe platform, while retaining an interest in storage. After working abroad, he co-founded an Internet-based music distribution company during the .com era, returning to consultancy in the new millennium. In 2009 Chris co-founded Langton Blue Ltd (www.langtonblue.com), a boutique consultancy firm focused on delivering business benefit through efficient technology deployments. Chris writes a popular blog at http://blog.architecting.it, attends many conferences and invitation-only events and can be found providing regular industry contributions through Twitter (@chrismevans) and other social media outlets.
  • http://ewan.to Ewan

    I think your definition is much better, but don’t think it does exist right now.

    When it does finally arrive, I’d like to add an extra item to the list:

    6. Provides appropriate storage to an application based on user-defined SLAs, not manual LUN placements.

    Once you’ve got those 6 requirements, I think you’ll have got very near storage management nirvana.

  • S.

    If knowledge is not required then we’re out of work! :)
    Sun unified storage claims to do #6 (based on ZFS)


  • http://blogs.netapp.com/extensible_netapp kostadis


    As always your free to define unified storage any way you want.

    But I think you’re being unfair to me.

    I explained the value of my definition here:


    And I explained why having a single replication mechanism is valuable.

    I think you get confused with the words single and the words flexiblity.

    Having a single mechanism that works the same across all protocols does not imply a lack of flexiblity.

    I also think you misunderstood my point about management.

    Clearly setting up NFS or CIFS has to be different, but provision the space for a CIFS and NFS export does not have to be different. That’s what i meant.


  • Mark

    What is the point of the first requriement? Even a hard drive has memory, a CPU (microcontroller), and of course a disk. This is kind of a silly requirement. If the “disk” part can be any flash storage, then my cell phone and even my watch meet the criteria.

  • Chris Evans

    Good point Ewan. Storage should be provided on a service basis in the future rather than require local/configuration knowledge.


  • Chris Evans


    I’ve always found that when we solve one problem, another one arises to take its place….


  • Chris Evans


    Exactly! I guess I was being sarcastic/ironic by changing the definition.


0 Flares Twitter 0 Facebook 0 Google+ 0 StumbleUpon 0 Buffer 0 LinkedIn 0 Filament.io 0 Flares ×