Time and time again the discussion around Information Lifecycle Management rears its ugly head… whether it is a client, colleague, or partner, someone is sure to raise the question;

“Hey Alex, what’s your take on ILM, or storage management”

… The room goes silent and you can hear the awkward gasps for air. 

Nobody wants to tackle the elephant in the room… Data! It truly comes down to data; Tons of it, mounds of it, hoards of it!

Sure, the hardware vendors seem to have their solutions, and so do the third party software companies.  Everyone has their solutions, lower cost disk, dedup, archive, etc. In addition, I have the utmost confidence that there are quite a few uber smart architects that can whip up a brilliant solution to funnel enterprise data from all tiers of storage, right through the data paths of efficiencies, and end up with a very compelling TCO story that will rock the IT world.

But hold on for a sec… We’re missing something here… Data!

Study after study concludes that the most prolific problem with enterprise storage is the unclassified data.  There are reams, and reams of data stored on common file shares. It grows, and grows, and grows, until it becomes the elephant in the room. This file structure makes it way too easy to save everything we have ever tapped out on a keyboard, along with the ump-teen revisions, to boot. The Windows file directory was never meant to be a document management solution, yet we all use it just as that.   We are all culprits, promoters, and worst of all, “Data Hoarders”. 

Yes, a well thought out architecture could dull the pain.  Popping a 500mg ILM pain reliever would do the job, but it will wear off, and capacity, at one point or another, in the storage architecture chain will max out.  We will have no choice but to tackle that elephant in the corner of the room…Data!

Alex Topitsch

Professed Data Hoarder! 

As I am inundated with cloud computing references everywhere I turn, I wonder if the cloud is cirrocumulus or stratocumulus. Or is the internal cloud cirrocumulus and the external cloud is stratocumulus? Or vice versa? Seriously, with the wide definition of the cloud, is my question that far off?

Let’s go back a few years; talk of a great dynamic datacenter concept where workloads could be dynamically provisioned, monitored, metered, and charged for. Wow, a great concept. Hmm, wasn’t this coined as “Utility Computing”? Sounds familiar, sounds like a concept straight from the clouds. But what happened? The computing power was there, dynamic storage was there, virtual networking was there. So why didn’t utility computing take off? Simply, the customers did not come. The customers would not take the chance with high profile production workloads on a dynamic platform that they could not feel or touch.  By feel and touch, I am not referring to the physical, but rather, lifecycle management.

The workloads needed the cradle to grave management experience to convince customers that utility computing was enterprise grade. Toolsets to self provision, meter, monitor, migrate, scale, charge, and retire, etc. were future concepts that were not available. Thus, utility computing fizzled away.
Flash forward to September 2009, cloud computing is buzzing everywhere! But isn’t this what we were expecting from utility computing a few years back? Or do I have my head in the clouds? Possibly, but if the application lifecycle software stack was at this level of maturity a few years back, utility computing would still be king. But no matter, it has a rebirth as the glorious cloud.

 

Alex Topitsch

Director Advanced Solutions