Video Screencast Help
Symantec to Separate Into Two Focused, Industry-Leading Technology Companies. Learn more.
Storage & Clustering Community Blog

Storage Optimization - Part IV - Tiering - SmartTier

Created: 26 Mar 2013 • Updated: 11 Jun 2014
Steve_Grimwood's picture
0 0 Votes
Login to vote

As in my series of blogs around Storage Optimization there is a commonality about the recognition of the
"data explosion" conundrum, and how this is a placing an extra burden on an already stretched IT storage team.

I've covered off how Symantec’s Thin Provisioning, Compression and Dedupe technologies can assist
with the ever expanding storage paradigm, now it’s time to change tack.
Today’s post targets getting the best value from storage, and in particular, using the appropriate tier of storage
and specifically Symantec’s SmartTier.

So what Symantec S/W are we talking about? - as ever, Storage Foundation.
SmartTier can offer better data placement on the correct tier of storage and reduce the cost of storage ownership.

What is Tiering ?
Simply put, tiering is selecting the correct tier of Storage for data to reside on, by aligning the data
classification with the appropriate storage requirements.

Is all data equal ? of course, the answer is NO
Data is classified in many different ways, as we can see in the examples below :
Is it current ?
Does it need to be accessed frequently ?
Is it mission critical ?
Does it require High performance Storage?
What are the file extensions - .dbf, .jpg, .txt, .wav ?

If we are able to identify and answer the classification questions then we can match the data with the
tiers of storage, i.e. Tier0 SSDS' | Tier 1 - Premium SAN HDD's | Tier 2 - NAS | Tier 3 - Jbod
The definitions are not important as classifications, but serve as an example, as data importance is specific
to individual companies.

As data evolves and structures change we must bear in mind data classification isn't persistent,
i.e. what's mission critical today, may not be tomorrow. For example, a mobile phone companies billing information
for the current month is very important, however this may not be so for the next month, and thus this makes it a
target for dynamic tiering to re-evaluate classification.

Equally H/W arrays have there own tiering mechanisms. H/W tiering works on IOTemperature, which requires data
that's accessed frequently to be promoted to a higher tier. By the time data is hot, the job may have completed.
H/W Arrays are unable to identify file type classifications, where as Symantec's File System is required for that
function (VXFS) enabling extension based migration.

SmarTier can be used proactively - if you can identify target data then you can proactively move it to the correct
tier prior to the start of a job, and convexley move it at the end of a job.

In summary therefore, Symantec’s tiering differentiators include :
The ability to proactively move data prior to job commencement.
Tiering on many different parameters.
Migration between different array types.
A consistent tiering function across disparate arrays.
Definition of data by extension type, .dbf .redo. .wav
Defining specific data types never to migrate between tiers.

As is the common mantra, Symantec’s values are such that we have no H/W agenda and we realise no benefit from selling more tin, we genuinely want customers to achieve storage optimization benefits.
Therefore if you are looking to use your storage more effectively and gain benefits from adopting tiering,
contact your Sales rep now !