Changing Economy of Storage?

After listening to the Wikibon discussion on FCoE yesterday, it really spurred me to piece together a number of things that have been posted as of late relating to storage access and the economies of storage.

I have been in the storage business for longer than I care to discuss, but, it has always revolved around the economies ($ per something, that something usually being capacity), with the economies of reality trailing the actual sale…and subsequently requiring further purchases (or in the cae of a generous vendor, free stuff….that usually doesn’t happen unless it’s t-shirts, mugs or coffee cards), to sustain the “real” need, and that is usually a performance related function (getting that capacity to do something in a timely fashion for the application or user).

In his blog, Tommyt (@StorageTexan) addressed a number of things, but this one specifically addresses where the industry is now regarding the building of a resilient/performing storage solution and moore specifically, PSA describes the realities around the needs that applications have for “data access” and getting that data in a timely fashion.

Rogerk (@storagewonk) has blogged about the confusion in comparing storage vendors and the “race for features”, while applications like databases, OS’s and hypervisors are putting most of those same features core to their function, where they rightly belong as the application or controlling entity is really the “boss”, as it were.

Davidb (@DavidBBlack), the dynamic speaker from Oak (we discovered this so well at Nerdfest 2010) and techno dweeb extrordinaire elucidated the gap that is ever increasing regarding computation and the access of data. This gap will drive the core capabilities of storage (storing, protecting and delivering data when and how fast it’s needed), as Tommy spoke to in his PSA discussion above.

Steves (@mensch45) illustrated how Xiotech is trying to change the game further, by focussing on the P of PSA, and ensuring the rest (storae and protect) happens in an even more reliable fashion while plugging more directly into the application and it’s control/needs.

Rob Peglar (@peglarr) spoke of the current trend in the industry of  “feature creep” and the storage vendors trying to constantly invent the new “feature” they can “sell” on/to, while making everyone want that feature like Apple has sold the world on IPOD’s (yes, I love my IPOD, that is a feature we all NEED); but do we really NEED all those other “features” that the applications/OS’s/backup vendors are integrating anyways?

Where is the rightful place for any of those features? Once reviewing what some of the great collection of minds at Xiotech are creating, you will see that letting the application rule has a lot of merit.

2 Responses to “Changing Economy of Storage?”

  1. March 2, 2010 at 4:04 pm

    Your points are well made. Any tiering option that simply looks at data access as an indicator of value (block level tiering) simply means that valuable data, when it is needed, will likely be on the wrong tier of storage. The only way to ensure valuable data will be where it is needed, when it is needed, is to turn off the tiering feature for that particular volume. Thus the Oracle database that is used by payroll every two weeks will not be relegated to Tier 3 disk and Tier 3 RAID during the two weeks it is not utilized, only to be moved up in the tiering three days after payroll is run. Proper tiering can only be made at the application or filesystem level where company management arbitrarily dictates data value and then enforces it with application policies that ensure data is where it should be at all times.

    This definition is straight from the Storage Network Industry Association (SNIA) ILM description. SNIA defines ILM as “the policies, processes, practices, and tools used to align the business value of information with the most appropriate and cost-effective IT infrastructure from the time information is conceived through its final disposition. Information is aligned with business processes through management policies and service levels associated with applications, metadata, information, and data.” Bob Rogers, Co-Founder, SNIA ILM initiative (ILMI).



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 242 other followers


%d bloggers like this: