Unravelling the ILM mystery

Information lifecycle management (ILM) is not a new concept. In fact, it's about as old a concept as data itself. Like many other concepts, however, it has undergone a facelift or two over time.

ILM is a policy or a procedure for handling data from the time it is created, through its useful life, and into storage, archiving and its eventual destruction.

ILM specifies precisely what the value of any given piece of data is and determines how it should be stored based on its value. Highly valuable data, which needs to be accessed regularly, will be stored on high-end storage for the duration of its usefulness. Once it is deemed less valuable, or needs to be accessed less often, it will be moved onto less available storage. Ultimately, it will be archived, usually on tape drives.

It's essential that the data is available when it's needed.

Simon Jeggo, data management sales and marketing manager, IBM

InfoBluePrint MD Bryn Davies says ILM is hierarchical storage management dressed up in a new suit. "ILM as a term being coined has just confused everyone. It has been around forever - records management, enterprise content management, etc. Most companies are doing it one way or another, particularly with paper and unstructured data. ILM is purported to add a more enterprise-friendly view of data management.

"The bottom line is that it doesn't address fundamental problems [like] the alignment of data across the enterprise, master data management, or a single view of the customer. ILM does nothing for that."

IBM data management sales and marketing manager Simon Jeggo says ILM covers virtually all aspects of data "from the time it is created, the uses it is put to, to the time it is destroyed". He says it covers disciplines such as master data management, data governance, data quality, compliance, data archiving, data security records management and data access.

The IT acronym game strikes again. Given the above, anyone considering ILM would be well advised to quiz potential suppliers on exactly how they define it. A quick Web search reveals that, by most definitions, Davies is correct. If one goes by common usage, and the usage employed by most of ITWeb's interviewees, then IBM's definition is indeed correct.

For the sake of simplicity, let's go with IBM's definition as it is the one most of those quoted here seem to have in mind. That said, given the number of acronyms for what amounts to data management in one form or another, why the fuss about ILM?

Says Jeggo: "For large corporations, it is imperative that they have a capable and extensive ILM strategy to ensure the data they use is of good quality and a source of trusted information. It is essential that the data is available when it is needed. This doesn't mean that all data must be online, but it should be accessible in an appropriate timescale."

Says EOH CIO Andrew Murray: "The longer companies fail to effectively deal with their data, the more cumbersome it will become to manage the entire organisation. I think ILM has become a catch-all term for storage.

"Storage capacity has been growing year-on-year, but we're getting to the point where that increase has to stop. Manufacturers are no longer going to be able to create any [more dense] storage. [Increasing storage capacity] is fine when costs are halving, but they're not going to halve any longer and cost is going to become an issue. People need to understand that now and start planning for it now."

KPMG IT advisory director Frank Rizzo says the ILM concept is not new. "It's a cradle to grave process for information, from when it is created or enters the organisation. This implies confidentiality, sensitivity, privacy and being able to access the information at the right time, on demand; all those good 20-year-old concepts.

"The difference now is that it is being legislated in South Africa, most recently through the Protection of Personal Information Act. We're starting to see clients taking this very seriously," he notes.

Although companies may be starting to take ILM seriously, most of ITWeb's interviewees note that uptake has not been great.

Says Magix Integration CEO Amir Lubashevsky: "I don't feel people understand the importance of information in the business. When we try to classify information and how important it is for the organisation, people do not assign the right type of importance, mainly from the perspective that information sits in so many places - structured and unstructured - and people don't really understand how to classify it properly.

"Companies should be looking at data governance, although there are many terms for the same thing. It's really about recognising information as a corporate asset, a business asset, and getting the business people involved. In terms of issues such as data ownership, data integrity, alignment of the meaning of data and terminology across the enterprise, this is where it all starts - as a business issue and not a technical issue."

Says EOH's Murray: "One of the main drivers for ILM is cost and, not so much in this country, around compliance and risk. CIOs need to start looking at what's required, especially now that e-mail is legal tender, and we need to be able to store and retrieve any business mail going in and out of an organisation.

"For the CIO, it is difficult to determine what is and isn't important, so we end up storing almost all of it. We need to catalogue and be able to retrieve data if we are legally required to do so and, at the same time, prove that it has not been tampered with. This means putting in the right processes, the right organisational models and structures in terms of how to manage data. You need to have a strategy to manage that. You need to be able to move data from being highly accessible, to tape, and yet still have it available on-demand when you need it.

"I don't know that ILM is going to be top of CIO to-do lists until we run into a wall in terms of the cost of storage. There is a lot of talk, a lot driven by vendors, but there is no compliance law today, although there are a lot of proposed laws. Unless you're an international company that needs to comply with [Sarbanes-Oxley], or are one of the banks and need to take these things seriously [such as Basel II], nobody has a good storage and ILM strategy. It's all talk and no action."

IBM's Jeggo concurs: "People have projects and are starting on it and looking at it, but I'm not aware of anyone doing what I would term full lifecycle management. A number of people are doing the archiving side; that is pretty common. Some are doing the master data management side, but not for ILM reasons. It's for other reasons."

I don't feel people understand the importance of information in the business.

Amir Lubashevsky, CEO, Magix Integration

Says Murray: "My bugbear is that data should become information and, ultimately, become knowledge to the company. Some organisations are putting in some CRM and BI to do analytics on data and create information, but that's almost not enough. It's how you put the right tools in place to create knowledge out of that; that's the trick I don't think anyone is actually looking at. The data explosion is going to catch everybody. The business is going to question IT and say, 'Okay, we've got all this data; now what do we do with it?' Data has huge value. The question is how to extract that."

Dr Francine Berman of the San Diego Supercomputer Centre has some estimations of her own. "In 2006, 161 exabytes (161 billion gigabytes) of digital data was created from cellphones, computers, iPods, DVDs, sensors, satellites, scientific instruments and other sources, providing a foundation for our digital world.

"Migrating digital content through new generations of storage media, making sense of its content, and ensuring that needed information is accessible now and for the foreseeable future, constitutes some of the most critical challenges of the information age."

Take the volumes of data that corporates have generated on their own in the last 50 years, add the amount of data expected to be generated this year alone, and you have a potential data disaster.