Information life-cycle management (ILM) and hierarchical storage management (HSM) have improved considerably and are achieving many of the objectives required of data archiving. Although there are key differences, both are focused on how to archive data efficiently. Typically HSM focuses on applying policies to the storage devices, while ILM focuses on the information.
Today, ILM delivers the ability to manage large data growth cost effectively through policies and procedures and then to migrate less-used/old data to less expensive storage devices. Often, ILM is used by large organisations who need to manage vast amounts of data. Most importantly, ILM also delivers the ability to discover and use data as and when required.
There are two primary methodologies to ILM - partitioning and archiving. These can be used in conjunction with each other. First, data is partitioned so that it can be searched quickly and easily andused to derive a business benefit. A potential challenge of partitioning is that as the data grows, so do the storage costs. On the other hand, archiving focuses on moving less frequently accessed data to less expensive storage, helping to improve database performance while reducing storage, management and infrastructure costs.
Data more important than people
Why is this so important? Without question, data is the most important asset for all organisations – more important even than people. Consider: if your CEO, CIO, Sales Manager were to leave the business tomorrow, would you go out of business? The answer is no. However, lose a month’s data, and it might be yes.
Smaller businesses often do not use ILM products owing to their cost. They rely on backup products to archive their entire data set, typically to tape. Once archived, there is a firm hope that the tapes never have to be used, as the restore process is slow, complex and fraught with challenges.
However, good news comes from recent advances in backup, recovery and archival products for medium businesses. The emerging real-time recovery solutions deliver enterprise class functionality at SMB prices. This enables companies to deliver an almost immediate ROI on their investment in backup, recovery, archive and disaster recovery products.
New real-time recovery technologies (now in their fourth generation), work at the sector level, rather than the file or block level. This is crucially important because a real-time recovery technology backs up a sector only once, and does not do so again unless that sector changes. Typically this results in much smaller backups and less disk use.
Additionally, as the incremental backups are usually tiny, backups can now occur as frequently as every 15 minutes with minimal impact on servers, users or networks.
This provides a substantially higher degree of protection and recoverability if a document gets damaged, deleted or corrupted. Furthermore, real-time recovery technologies are heavily focused on the recoverability, rather than the backup process itself.
Migrating / archiving old data
Tape has been the primary archival mechanism for years owing to its low cost compared to disk. However, the cost of disk continues to decrease, further driving away from dependence on tape.
Additionally, if archives created using real-time recovery technology are stored on disk at remote sites, or in the cloud, the archive can also double as a second disaster recovery site. If the archives are based on exact state imaging, the backup archive can be spun up into temporary virtual machines within minutes. Tape is incapable of doing this.
So, how does this impact the lifecycle of a document? Even at the final resting place of documents (the archive) – the documents should still be accessible, available and recoverable. It is this that real-time recovery delivers.