Eliminate unnecessary data from your production storage. This typically involves moving the unnecessary data to a different tier or deleting the data entirely. Use information lifecycle management
(ILM) and data retention policies to identify and eliminate old or unneeded data. We can come up with all kinds of schemes to manage, store and deduplicate data, but ultimately it's all about making key decisions -- what data to keep, what data is needed by the business and where that data should be stored (e.g., tiers).
From an archival perspective, maybe the solution is just to move data from our main storage to an archival platform where it can be infrequently accessed if needed. Content-addressed storage
(CAS) is an important archival option where the information needed to quickly retrieve unstructured data (the index and metadata) is kept in the CAS system itself. Data deduplication
is usually an integral part of archival storage where only a single instance of a file is actually stored on disk. Database-assisted archiving products offer yet another possible solution. If data is no longer needed or has met its retention period, it may be preferable, even necessary, to delete the unneeded data.
Regardless of the platforms or products, you cannot count on technology to make the retention/deletion decisions for you. Make the decisions first at the corporate policy-making level, and then use technology to enforce and assist you with those decisions.
Listen to the Unstructured data FAQ audiocast.
Go to the beginning of the Unstructured Data FAQ Guide.
27 Mar 2007