Lowering Overall Costs By Implementing A Long Term Data Preservation Strategy

Data growth rates, the high operational costs of managing more storage than ever before, and a national power grid that is not likely to keep up with power demands in the near future is creating a perfect storm that is driving enterprises to look at new approaches to data management. The concepts and technologies are here today that will allow enterprises to achieve an order of magnitude reduction in the costs and power to manage storage by harnessing information about data usage patterns in their environments and making evolutionary changes to their data management processes based on these patterns.

First, if enterprises are not already tracking this data, they need to begin tracking and analyzing file access by age. As they begin to better understand the variables driving file access over time, enterprises can introduce additional measures (such as frequency of access) but file age is a simple, approachable metric to use to begin to address this problem now.

Second, enterprises must resolve the problem of easy access to data that may no longer reside on primary storage. While tape can provide very low cost storage and an almost limitless capacity, it introduces long access latencies that generally require manual intervention, and suffers from other problems such as recovery reliability and obsolescence as tape formats and technologies are replaced by newer, incompatible solutions. With litigation rates rising rapidly, and many enterprises involved in multiple lawsuits per year, ediscovery is quickly becoming a compelling economic reason to ensure on-line searchability of older data. Running discovery against tape-based data sets costs roughly an order of magnitude more than running it against on-line resources per lawsuit. The fact that tape is not accessible on-line is one of the factors that is contributing to primary storage "bloat" in enterprises of all sizes.