Information lifecycle management: destined for a bumpy ride

Graham Titterington, Senior Analyst

Information lifecycle management (ILM) has become a cult in the second half of 2003. It appears to be the result of a business need, but it has a long way to go before it becomes an accepted business tool.

This relatively new discipline recognises that the properties of data change over time - sometimes automatically with time, and sometimes in response to events, such as the appearance of a new version of the data. ILM uses an automated workflow for storing, moving, copying and deleting data. The problem is that to be generally useful, ILM needs to understand something about the data, beyond looking at the routine file attributes.

Management of information requires competence in both the content management and storage areas. Content management brings an understanding of what the data means and of its business significance. Storage provides the means of delivering the overall information system, and elaborate storage technology is required for satisfying scalability and geographic distribution requirements.

This is why some storage vendors are turning to content management vendors for support, through acquisition, partnership or by working with other product teams in the same vendor. Content management systems can generate and associate metadata with data to assist in its management.

Most of the big storage vendors are rushing onto the ILM bandwagon. EMC has been the most prominent, as its initiative has been linked to two high-profile acquisitions: Legato and Documentum. Veritas, IBM, HP and Hitachi Data Systems have all made moves into the ILM space. Hitachi is partnering with IXOS for content management and email management. HP is working with IXOS, Legato, Persist and several email specialists. IBM has developed the Tivoli Storage Manager as its central response to this demand. OuterBay has developed an `application aware' data lifecycle management tool and is working with EMC. Veritas has also developed an ILM product.

However some significant problems remain.

Content management is not yet used widely enough

Content management is only used selectively in those organisations that have adopted it. Legislation and regulation apply to all data held by companies, as do the pressures of managing data storage. Content management processes have to be applied consistently across organisations before they yield their payback, so the adoption of content management requires a substantial investment of time and resources. Whatever ILM technology is put in place has to work effectively in an environment in which much of the data is not tagged or indexed by a content management system. Content management is not a `quick win' technology.

ILM has scalability problems

ILM has to be applied across the enterprise storage networks, including offline storage. The scalability problems are most acute in the information searching area. For example, Veritas's information search function, modelled on the Windows file search, is somehow supposed to work across corporate datastores, including multiple versions of archived data. Experience with Windows file systems shows that this function is not efficient for searching large volumes of data and that a more structured approach, understanding the semantics of the data, is needed.

Integration hurdles are being underestimated

All of the initiatives around today have been driven by storage vendors, with a content management specialist playing a supporting role. The task of fully integrating the two technologies is hard. The design of appropriate metadata structures for data categorisation is difficult, and needs to anticipate the changing business environment throughout the life of the data and of the data management system. Today we only see the first step on the road towards integration, and the steps we see could turn out to be on a dead-end road.

The storage vendors have a poor record of integrating anything. Storage virtualisation, with its modest aim of making all disks appear logically similar, remains an aspiration after years of work. The task of integrating storage infrastructure technology with an understanding of the logical content adds another dimension to the complexity of storage management.

User organisations should now be working to improve their content management processes in order to satisfy today's regulatory environment, and to be able to exploit ILM technology when it becomes mature.

To find out how Ovum can help your organisation understand these issues, see Software@Ovum, you can also contact Graham Titterington directly on gct@ovum.com.