Dependent on regulation and policy, tape may be required, although it can be completed at a central site. Elimination of tape brings many efficiencies, cost savings, performance, and reliability that traditional tape backup and restore have consistently fallen short on delivering.
De-duplication is often an appliance within the enterprise environment. Some common concerns about a non-appliance-based de-duplication include:
De-duplication is most efficient when de-duplication occurs for all data in the environment. For example:
• Some solutions perform data de-duplication dependent on the storage volume or host. If similar data is stored in two different volumes, de-duplication of the same file in different volumes may not occur.
• Compression must occur on de-duplicated data to increase efficiency.
• Performance overhead must be considered and monitored closely. Many vendors claim a 3 to 4% performance degradation when de-duplication activities are occurring, although many customers have reported up to 30 to 40% performance the degradation.
Archiving is a process primarily driven by automated workflow and business defined roles in managing data within the enterprise storage environment. Redundant file elimination and redundant block elimination are essential to a successful archiving solution. The ability to store “like files” as a single instance has great cost savings associated with the technology.
Archiving is a complementary technology to data de-duplication. Remember that archiving will address storage needs for active data, whereas data de-duplication addresses the backup data.
Archiving functions on a technology that places “stubs” to find the archive data. For example, I have a file called Sparky.Doc, and it is stored on the primary storage. When stored in an archiving solution, a single copy of Sparky.doc is retained, while duplicate copies of the file are saved as “pointers” to the original file.