One of the major drawbacks with information lifecycle management (ILM), says Graham Titterington, principal analyst at research group Ovum, is that, despite all the chatter about it, “true ILM across the enterprise does not exist today”. But the parts that are there already provide significant business value. “The benefits [of ILM] for business are immense,” he says, “but big benefits don’t come easy.”
Speaking at Information Age’s inaugural webinar in November, Titterington joined sponsor, storage consultancy GlassHouse Technologies, in examining ILM’s role in managing information assets in an increasingly stringent regulatory environment. In an ILM world, data which is of greatest importance to the organisation is accessed most frequently and is placed in the most reliable and available storage device. Data is moved down the storage hierarchy as its value diminishes until eventual deletion. Currently, business adoption of ILM falls into three distinct categories, said Paul Hammond, managing partner at Glasshouse: those that have bought technology that claims to solve ILM; those that believe ILM could help impose some order on corporate information assets, but do not know where to start; and those that have yet to do anything about implementing ILM.
Point solutions, such as email archiving technologies, can help organisations impose order on specific categories of data, said Hammond, but in dealing with information held in databases, businesses are faced with the dilemma of analysing which records are actually valuable. “It is the data versus information challenge. To get value, you need to work out how records will be defined,” he said.
In part, the growing volume of legislation that is forcing companies to actively manage records is helping businesses classify the type of information they hold, thereby imposing storage policies on that data.
“But keeping the regulators happy is the easy part,” said Titterington. “Just because you’re required to keep some data, there’s no reason to assume that it is that data that is going to be of the most value to the business.”
If businesses are not able to differentiate data on the basis of value to the organisation there may be a temptation to simply store everything. In part, evidence of this can be seen in the spiralling costs of storage at many organisations: for example, private health provider Bupa, estimated that before it began to manage its data, it was creating around 3 terabytes of new data annually.
Those levels are not uncommon, said Hammond, but they usually prompt senior management to question why storage costs are rising so fast. “We have a tendency to be hoarders: everything gets shoved in the attic, so to speak – until the ceiling collapses,” he said.
Responding to questions from the webinar audience, Hammond said he advocates using charge-back mechanisms for storage, as a precursor for full-blown ILM, in order to get business units to realise the impact of storing everything. “They realise that the costs associated with the ‘platinum’ option for storage might not be justified when the ‘bronze’ option will do,” he said.
This can then help the business units become more sophisticated in their thinking about data classification schemes: applying business-based policies to data storage. And ultimately, that is what ILM is about: setting policies and processes in place, and building the architecture underneath to support that.