What is Copy Data Management? – Trouble With Tribbles

gilroy-90By John Gilroy, Director, BLT – In 1967 Captain James Kirk was perplexed with rapidly multiplying bundles of fur called “Tribbles.”  Here in 2014, our “Trouble with Tribbles,” involves copies of data. In most environments, you may keep a production copy of your data and then make a few copies for data protection, information sharing, analysis, or perhaps compliance. Hard drives have gotten cheap, virtualization has made them larger and the net effect is that copying data had gotten out of control.

This may seem like a minor concern until you realize that IDC estimates that copy duplication will cost businesses roughly $46 billion in 2014.  A recent study of federal information professions estimates that in 2014, the federal government alone will spend $2.78 billion in storing copies of data.

Ash from Actifio

Ash Ashutosh is the CEO of a company called Actifio whose focus is on improving the ways that large companies manage data.  On July 29, 2014, he sat for an interview on Federal News Radio to talk about the problem and some possible solutions.  The interview began with explaining traditional concepts in managing data, then a description of the offering by Actifio called the Virtual Data Pipeline™.

What about the ACID test?

The discussion began with the classical ways to describe data.  Ever since Star Trek, information technology novices have been introduced to the world of data management with the acronym “ACID.”  Grizzled veterans know that ACID stands for

Atomicity – this concept makes sure the data is stored in a manner where the entire transaction is complete.

Consistency – assures that related data will have no impact on the data stored.

Isolation – defines the relationship with other data, guaranteeing that only completed data will have the ability to alter other data.

Durability – makes sure that the completed transaction in the database is stored in a medium that is safe, like a hard drive.

Essentially, these terms refer to making sure the data is accurate and can be recreated if there is a system failure.  Even if a dutiful system administrator can assure an ACID copy of a database, this does nothing to address the cost of retaining this copy.  Yes, hard drives and inexpensive and virtualization has further reduced cost, there is still a major cost in storage that companies have only recently realized.

Options

The world of database management has many debates about ACID and data fidelity. Rather than delving into the debate concerning consistency models, this discussion highlights the offering from Actifio called the Virtual Data Pipeline. Their approach is to make a “Gold” copy of the data where all the other instances are derived from.  The patented approach reduces storage costs and improves network utilization.

During the interview Ash makes some startling statements.  He refers to the recent study by Meritalk mentioning 40% of federal agencies having four copies of data.  In his recent experience with a Fortune five company, that CIO mentioned over five hundred copies of a database!

There is pressure on all large organizations today to reduce cost of their information technology systems. Instead of seeking out complex bleeding edge systems, a better approach may be to reflect on something as simple as how your company, or federal agency, is handling copy data management.  The solution could be right under your nose, whether it is virtual or not.

 

This guest post from John Gilroy was originally posted on the TechTuesdayBlog.

Recent Posts