Data Reduction for Archive


Every modern data center has an archive class of data whether they know it or not. This is the storage used to retain and protect data that is no longer hot, not even warm, but which may be required in the future for anything from resumption of operations to analytics, regulatory compliance, or even legal protection. The most common way to handle archive data is via object-based storage solutions, but the deduplication provided by object-based storage systems is limited to a per-object basis.

Take this example:

  1. A user has a set of 26 documents (A-Z respectively), each of varying size, that they insert
    into a zip file.
  2. This file is then saved as an object in a cloud storage system.
  3. Another user in the same organization has the same files, but inserts them into a zip file in the opposite order (Z-A).

These two objects (zip files) will NOT deduplicate; they are different.  So the organization is left storing 52 files (2 copies of 26 files).  Now magnify the problem by the thousands of users in a large enterprise and you see the issue.

Permabit Solution

Data reduction, the combination of deduplication and compression, is the answer to the problem of archive storage bloat. Additionally, Permabit provides unique data Optimizer technology which transparently breaks up objects on logical boundaries to realize data deduplication even in the situation described in the example above.  After optimization, block-level deduplication can identify duplicate chunks of data on 4 KB boundaries so that each unique chunk of archived data is stored only once. And once data deduplication is performed, compression can be applied to unique blocks remaining in order to reclaim additional space.

Permabit’s Albireo products are available that provide all three of these technologies: data optimizer, data deduplication, and HIOPS compression for use on premises as well as in hybrid cloud and public cloud environments.