Information lifecycle management (ILM) has been widely bandied about as the panacea to storage headaches, but common standards need to be set before this new approach to data management can make an impact. With data creation in overdrive and government-led regulations such as the Sarbanes Oxley Act placing more emphasis on certain data, firms are struggling to make sense of ever-growing mountains of data using limited resources. This is what ILM promises to do, by intelligently organising information across a storage system. Spotting an opening, IT vendors - mainly storage players such as IBM, Network Appliance, Hitachi Data Systems and Veritas Software - have all announced ILM initiatives of some sort, and have been jostling for new technologies and bagging new companies to plug holes in their strategies. Data storage kingpin EMC Corp has even gone so far as to realign its entire company strategy with ILM. But is it the silver bullet that vendors are making it out to be? 'The problem for customers is that ILM means something different to each of [the vendors], and potential customers are befuddled by the confusion and hyperbole surrounding it,' Adam Couture, principal analyst at IT research firm Gartner said in a February report. Tom Zack, HDS vice-president, marketing sales and technical operations for Asia Pacific, agreed. 'The overaggressive positioning of ILM by some vendors is pushing this technology rapidly into the 'trough of disillusionment' but we do believe it will eventually come out and become a productive technology,' he said. 'All the industry wants is a common understanding on ILM, and customers are going to wait for the dust to clear,' said Paul Massiglia, technical director for engineering at Veritas. So what exactly is ILM? Again, ask different vendors and you are bound to get different answers. But there is one thing that they all agree on: it is not a product, a new technology or even a new storage solution. It is 'not a 'rip and replace' hardware purchase strategy', said Andrew Hu, Network Appliance regional managing director in Greater China. 'Most customers already have some ILM solution components deployed in their environment.' In essence, ILM is no more than common sense. '[It] attempts to classify various categories of data according to their importance and business value, and establish policies to migrate the data to different storage media accordingly,' Mr Couture said in his report. In this regard, it also differs from ILM's predecessor, hierarchical storage management (HSM), which also dealt with data movement. HSM automates data migration for archival purposes, and it is access-driven. Such a simplistic approach does not meet today's demands 'where valuable data cannot be just determined by the amount of times it was accessed', said Mr Massiglia. In contrast, ILM employs automated data migration (ADM) tools which use intelligent SRM to enable data migration across various storage resources based on user-defined criteria and data value. This is especially important in today's compliance-driven world, where regulations like the Sarbanes Oxley Act determine the number of times data is archived as well as the requirements surrounding the archiving. 'Increasingly, customers interact with their suppliers in various forms that are not suitable for management by traditional database structures,' said James Hung, manager of TotalStorage, Systems and Technology Group at IBM. 'This is already having a serious impact not only on compliance but also on good business management.' Along with disaster recovery and business continuance tools, ILM surpasses HSM to meet today's need - a reason why Gartner calls ILM 'HSM on steroids'. In practice, however, ILM is not going to be easy to achieve. First of all, it needs competing vendors to co-operate and agree on standards. Historically, this always leads to the dragging of feet. 'Standardisation is something that as a big market is problematic and takes a long time,' said Mr Massiglia. Although some success has been made in the arena of common APIs (application program interfaces) and standards, the industry is far from producing products that work or integrate seamlessly. 'I do admit that we are still a long way off from that,' said Ricky Chung, marketing programme manager for Greater China at EMC. However, Mr Massiglia said that 'pressure' to work together came not only from market economics 'but also from the government' enacting regulations. 'I think the government will drive customers to drive vendors to work together, and time to maturity will be compressed,' he said. The second problem is changing the mindset of customers, which, according to Mr Chung, is 'not just related to ILM but everything else in IT'. Consolidation and centralisation - all side effects of ILM - can lead to political innuendo that firms are seldom prepared for. And lastly, ILM has one weakness: Policies. 'Without the right policies, ILM will not be as beneficial to your company. ILM is an evolving process, and for IT to realise the benefits, usage patterns and policies must be constantly reviewed and audited, which is often easier said then done,' said Mr Chung. Even if customers feel they are ready for ILM, they can only implement it in parts, as no single company today offers all the pieces to complete the ILM puzzle. This means a best-of-breed approach is more suitable but will require people to make difficult choices. 'If a customer waits for a total ILM offering to address 100 per cent of their information, they will be waiting for a very long time,' said Mr Zack. Moreover, best-of-breed solutions are complex, which runs counter to the reason for installing ILM in the first place. 'An ILM solution has to be simple to deploy and manage without increasing operational costs,' said Mr Hu. For IT managers, at a time when IT investments are being scrutinised by their finance-minded colleagues, minimising risk remains vital and may dampen enthusiasm for ILM. 'Customers are not looking for anything disruptive,' said Mr Hu. This is the reason why analysts such as Gartner forecast that 'services' will be important for ILM's uptake. 'Services will be key to the success of any ILM implementation,' said Mr Couture. 'Without defined services to link ILM to business value and road map, the chances of widespread ILM adoption are very low,' he said. Nevertheless, the need for ILM is real. 'Already over 80 per cent of a typical organisation's information holdings are a candidate for ILM,' said Mr Hung. 'That sort of need is unlikely to go away overnight.'