I think many will agree that 2004 was definitely the year of compliance. However, with many organizations scrambling to meet some of the compliance requirements and deadlines (like Sarbannes-Oxley), a number of storage and data backup decisions may not have received all the attention they deserved. For numerous organizations based outside the U.S., 2005 will bring much of the same, as foreign companies publicly traded in the U.S. will also have to be SOX compliant by mid-2005.
For some organizations, assuring compliance simply meant throwing more storage at the problem to avoid having to make any kind of immediate data retention decision. However, keeping more data also means backing up more data, in most cases. 2005 will put back in the spotlight a number of capacity related storage management and data backup challenges that had been relegated to the back burner in favor of the more pressing and omnipresent concerns around compliance. These old storage wounds will impose a need on many to reexamine their corporate storage and retention policies.
We can also expect to see an increase in the demand for application independent data archival solutions. The dependency on software that is as old as the archived data to be retrieved has been at the center of many data backup debates for a long time. With legislation now imposing new rules in terms of availability of records, it will only add to the debate. The simple fact of having a copy of the data on some media will not be sufficient if the records sought cannot be read.
ILM and storage virtualization
While some vendors keep promoting ILM solutions, other vendors are offering storage virtualization. As far as the end users are concerned however, the distinction between both technologies is not going to get any clearer in the coming year. Both solutions attempt to present storage holistically by abstraction of the hardware layer. Both promise a greater return on storage investment. Both are aimed at increasing storage manageability.
Unfortunately, the lack of useful tools to automate data movement will prevent the users from reaping the full benefits of these technologies. It is unlikely that 2005 will deliver automated policy based storage management solutions capable of looking beyond a file's time stamp and offer some form of storage decision intelligence. Organizations will continue to face the daunting task of manually categorizing data for the purpose of storage tier allocation and backups.
Read-only LUNs The growing need for unalterable storage media such as WORM (Write Once Read Many) will continue to be driven by numerous regulatory compliance requirements. This will ultimately force more storage vendors to live up to their D2D backup promises by offering a magnetic disk equivalent to this technology.
However, although some offerings have already hit the market and more will come, we should not expect to see all storage vendors race to develop their own disk based WORM offering. Some vendors will most likely wait and see what kind of demand there will be especially with the growing availability of tape based WORM technology, which offers the possibility to leverage existing storage infrastructure investments.
Disk-to-disk backups will by far continue to be the most discussed backup topic in 2005. Whether it is through replication, streaming backups or tape device emulation, the performance advantage and ever decreasing cost of disk will see D2D backups continue to gain in popularity as more organizations adopt the technology. The only hurdle will remain off-site storage, which is still hampered by network bandwidth cost/availability. Although some vendors are promoting their new and improved, revolutionary data compression algorithms, network bandwidth limitations and costs are still very real. This leads us right into the next topic.
Tape backups Without contradicting the previous statement, backups to tape will remain strong over the coming year. The portability of tape cartridges will continue to make it the media of choice for off-site storage of backup copies or long-term data retention. Many organizations that have made considerable investments in tape storage infrastructures will not be in any rush to part with it unless they are falling short of meeting their recovery requirements. However, it can be expected that a large number of net new backup solutions will include disk as a primary backup media.
The big deception for 2004 Many storage experts had predicted that ILM would be the big storage event of 2004. Not to say that the need does not exist but vendors have not yet been able to offer a solution that can make automated intelligent storage decisions. Although certain vendors made some progress, data categorization and retention remains a manual process for most organization.
About the author: Pierre Dorion is a business continuity consultant with Mainland Information Systems Ltd. in Calgary, Alberta, specializing in business continuity planning, backup and recovery and data availability as well as IT processes. He has extensive experience working with backups and storage management solutions in a variety of operating environments with oil and gas, railway, communication and utility companies. Pierre is also an IBM Tivoli Storage Manager Certified Consultant, is an authorized IBM TSM Instructor and holds an ITIL Foundation Certificate. Focusing mostly on business continuity and IT recovery, he has been a guest speaker on data availability at a number of conferences such as ARMA and CIPS.
This was first published in December 2004