ILM step by step

Hardware and software vendors like to define informationlifecycle management, ILM, in terms of their own products. The factis, however, that ILM is simply a philosophy of ensuring that theinfrastructure chosen for the storing of data is periodicallyaligned with and thus relative to the actual business value of thedata throughout its life.

This approach to ILM can be based on a reference architecturethat provides tiers of service of differing capabilities and thusrelated costs. But how will a move from tier to tier be triggered?Everyone talks about ILM, but it seems most implementations are infact DLM strategies – where D is for data. What is the differencebetween data and information? How is the value of data calculated?When is the delta in value sufficient to trigger a migration? Aredifferent strategies needed for structured data, and unstructureddata? Is ILM in fact application specific? What is the impact ofILM-driven data migration on recovery capabilities? Will the costof administering and controlling the ILM environment chew up theprojected savings?

Let’s briefly look at the questions we need to ask of ourselvesand the vendor before embarking on a journey towards ILM.


Just what is the difference between data and information? Howdoes it affect lifecycle management? Data generally needs renderingby an application before the result can be utilized. Information isdata that has been rendered into a human actionable form. What arethe implications?

Valuing Data – $$$ or dreams

This is an area where many vendors take the broadest definitionof value and use subjective measures such as “importance.” Ifmovement of data is to be automated and based on change in value,then value needs to have some empirical base. This empirical basealso needs to be capable of dynamically reflecting current “value”to allow continuous monitoring for change.

Enough change is enough?

Let’s assume that we have been able to calculate andsubsequently dynamically maintain the value of data over time. Nowwe need to ask ourselves what delta in change would be sufficientlysignificant to trigger migration to a more appropriate tier ofstorage. This is perhaps one of the few easy ILM questions toanswer as we have already learned how to do this when we set up theservice provider model.

ILM – camouflage for archiving and HSM

Even though it may not be possible to effectively value datadynamically enough to trigger automated migration, some ILMmigrations can be auto triggered by non-value-based criteria suchas date, age and other standardized metadata components; however,it is difficult to view this activity as anything other thanarchiving in an HSM environment.

Once the issues above are well understood, the implications ofthese issues are apparent and an understanding of the cost benefitof ILM has been agreed on, a CIO can develop a transition plan fordata that fits the ILM parameters as follows:

A five-step transition to ILM

1. Determine a cost model that can show savings that will bedelivered by moving data from point A to point B on condition X.This allows your policies to be developed based on a rational costsaving basis.

2. Determine your strategic archiving policy. What types of datawill move, what will trigger the move (attribute delta), and wherewill it move to based on the attribute?

3. Determine tactical archiving policies. Define for eachspecific data type or even data base or file holder, the exactattribute and its delta metric that will trigger a move to adefined tier or location.

4. Attempt to model the frequency of change and the migrationtime of data along with its impact on production. This will providea basis for the degree of automation possible and possibly even ajustification for investment in automation.

5. Develop standard operating procedures with associatedcompliance, completion and quality metrics to discipline theoperation and place it on a highly repeatable basis, whether it beautomated or manually triggered.

But above all else – know what you want, why you want it, whatit will cost and what it will save you.

You decide. 061275

Related Download
Improving the State of Affairs With Analytics Sponsor: SAS
Improving the State of Affairs With Analytics
Download this case study-rich white paper to learn why data management and analytics are so crucial in the public sector, and how to put it to work in your organization.
Register Now