In 
As enterprise data volumes continue to grow in colossal manner, you, as an IT manager, are challenged to keep the storage infrastructure responsive to information and compliance needs both from within and outside the organisation.
ITNext attempts to help you find the way out. But first, a few simple questions:
Does all data in your organisation have equal storage importance and retrieval priority?
Do you have no budget constraints when it comes to buying additional fibre channel (FC) disk arrays?
Has the economic slowdown also slowed down your storage requirements?
Of course, a common answer to the queries would be a clear no. Now that we have done that, lets see how it can be of help.
Add, add, add?
Traditionally, that is what we have been doing: we have bulk-stored enterprise data on arrays of disks and kept adding disks when the data inflow exceeds the installed capacity.
After a while, the process starts appearing as mundane as adding chairs in the back of a conference room as new attendees keep pouring in.
And you thought data storage was about information lifecycle management (ILM). Yes, it is indeed about management! In fact, the task is getting more and more complex by the day, not to mention that adding disk arrays linearly is going to make this complexity only more baffling.
So what is the solution? Before coming to that lets first visit the answer to question number 1: not all data are equal, it says.
Why do we then treat all data as almost equal? We store all enterprise data on the same storage media without discrimination. Agreed, the prevailing data storage practices are an outcome of tradition than of your conscious choice. But then, probably it is time for one to take more conscious decisions, proactively. We need to segregate data by its sensitivity, access priority, compliance requirement, and other parameters. We need to be cost-conscious and prudent in choosing the storage media.
Getting started
Running a detailed audit of the storage infrastructure and needs in your enterprise should get one started, the objective being to discover the following:
1. What is the total installed capacity and how much of that has been utilised?
This is not just to estimate the capacity that would be available but also to gauge how much capacity is being utilised on a cumulative basis versus annually. In other words, this is to get a ratio of the archived data versus current data.
This first level of data segregation gives you a simple yet effective basis for determining how much of higher-speed storage media would be a must, as the rest of the data would go to reside on a lower-speed media.
2. Which all storage media have been deployed and what is the break-up?
This will give you an estimate of how much storage media is available across high, medium and low-speed segments. This, along with data segmentation, should get you started with mapping the high-priority data to high-speed storage media and help note the need for procuring any additional media.
Add new comment