I plan to keep this simple and straight. No gimmicks :)
What came first chicken or the egg? I am kidding. I don't know that answer yet but if we relate the same to enterprise data we know the answer. Its information life cycle management. Data is born. It attains puberty, goes through the usual adolescent pangs, becomes young and lively, finds more data which is related to it, multiplies, matures , ages, becomes old and then is no longer relevant (and then dies).
Well hold on to that thought I said dies but does it really? People have an habit of digging up old graves and you would want people to have an option of digging it up. So that's fair and all that. But how do we do it.
How can we make sure that we keep data when is needed, keep it well massaged while its active and even when it is no longer needed (becomes old) make it feel all comfy and give it some pension and make sure that all's well with it so that when you need some advice you can always go back to it. That's why I love old people. They might not be quick or beautiful but they have the experience and you can learn from it. Same holds good with data. More older the data , more accurate is the analysis and pattern derivation. But wait before that lets talk about how data ages and how it matures.
When the data is new and fresh its the most volatile and active. It needs more attention, grooming and every one tries to access it. So its imperative that this data is stored in a highly reliable, resourceful environment where applications need not bother about latency, performance or any such negative buzz. That is what is production data. Tiered storage is the way to go to leverage optimum balance between performance and storage and associated costs.
On the left, you have one to two years worth of “current” financial data in your production database. Your monthly and quarterly period processing performs well, as do your yearly comparison reports.
In the middle, you have another few years of financial data in an active archive database that provides full access to your data with mid-level performance.
You can archive even older data to a flat-file format to take advantage of more cost-effective storage options, such as HP StorageWorks™, EMC Centera™ or NetApp NearStore® with SnapLock™. If you need access to data stored in one of these storage alternatives, you should be able to access the data directly and/or selectively restore some of the data to the active archive database.
Finally, for the oldest data that you must retain for a decade or two or more, having the ability to store these flat files in an offline archive tape may help reduce costs even further.This diagram, of course, would be adjusted to meet your company’s unique storage requirements and strategies.
IBM Infosphere Optim provides the ability to store archived data on the most cost-effective storage alternative, according to its business value and access requirements. Every company values data differently. Understand how your data is valued and spend as much, and only as much, as you need to provide data access.
I will discuss more details about how IBM Optim can help in data retention in the next part of this series.
Hope you enjoyed reading this blog and do send me your valuable comments. In case you would want me to blog on something specific to IBM Optim do let me know and I would attempt to include that in my future blogs.
Well this is much food for thought for now. Till next time its me Girish, signing off.