Virtualization to the Rescue

Dave Graham

Subscribe to Dave Graham: eMailAlertsEmail Alerts
Get Dave Graham via: homepageHomepage mobileMobile rssRSS facebookFacebook twitterTwitter linkedinLinkedIn

Blog Feed Post

The Cloud Provider Strata Post (IRM, Nasuni, Nirvanix)

Since it’s been roughly (i don’t know) several months since i last wrote (and yes, i’m doing the e.e. cummings thing by only using lower case letters), i decided it was time to motivate the sedentary gray matter between my ears and actually put finger to keyboard to…pontificate, really over the recent events in the Iron Mountain stepping-out-of-the-cloud debate.  Coupled to that was a challenge from Val B @ NTAP to expound upon my assertion that Iron Mountain, Nirvanix, and Nasuni were all stratified differently in the cloud space.  So, without further ado, here’s my thoughts.

(brief editor’s note here:  I’m really not doing a ton of research into specific features/functionality here.  just reporting what my end users are seeing and what reality, as unfortunate as it may be, dictates)

Nasuni.  Nasuni is a MSO offering where the “manager” is Nasuni and the offering is a limited scale scalable, semi-SMB 2.0 compliant set of virtual machines.  I’m not trying to disparage what they’re doing because honestly, it’s a great product when positioned correctly.  However, if you’re looking for performance (e.g. speeds/feeds), you’re going to be out of luck.  Classically, this is a Small-Medium business product. The decoupling of Nasuni from the backend service is an interesting question to ask them;  typically, a customer runs their Amazon account directly through the service, paying “all in” to Nasuni.

Nirvanix.  Nirvanix is a hybrid of a cloud offering plus perimeter edge (or gateway style) deployment model.  As such, Nirvanix most CLOSELY mirrors what IRM was trying to do, but they’ve been moderately more successful.   I say “moderately” in that I’ve run across exactly 0 (zero, for those who need it spelled out) folks who actually use their product.  Moderately from the sense that, in my not-so-humble-opinion, their CloudNAS is a wasted effort when compared to Nasuni.  So, with Nirvanix, you’ve got backhaul (e.g. the “public” cloud portion) via their SDN and on-premises via hnodes and/or CloudNAS.

Iron Mountain. Iron Mountain approached the cloud like another business model to be tackled.  as a “master” of backup, the assumption was made that the same model could be made to work using the existing customer base plus some on-premise hardware that would accelerate the writing to the cloud for digital asset management (assets being loosely defined as backup/archive). to further ‘extend’ functionality here, Iron Mountain brought to bear two discrete clouds; one for their archiving services and one for the backup services.  The problem here is in the approach; rather than create a SOA-based architecture, IRM went for the retrofit…and lost.  It’s unfortunate, to be sure, but certainly not unexpected given the momentum in the cloud storage space.  With application level “integration” (albeit loosely), IRM essentially looked like a scaled-down version of Nirvanix with software capabilities.  Target market here was everyone, SMB, Commercial, Enterprise with (my opinion) the biggest folks involved being commercial and enterprise, not SMB.


The cloud is constantly evolving and frankly, we’re only at the tip of the “cloud darwinism” cycle.  many more CSSPs will be subject to closure and it’ll be the strongest that will survive, both from standpoint of application integration and perimeter reach.

UPDATE:  it’s also important to look at who controls the data.  IRM was on-pre + cloud w/applications.  Nirvanix is application agnostic but self contained.  Nasuni is on-prem with no owned cloud but multiple backends…

Enhanced by Zemanta

Read the original blog entry...

More Stories By Dave Graham

Dave Graham is a Technical Consultant with EMC Corporation where he focused on designing/architecting private cloud solutions for commercial customers.