1

Data Storage Is So Cheap Its Expensive

The cost for 1GB of magnetic disk storage 20 years ago was $1,000 – now it’s eight cents. 1GB is enough to store about 20 thousand letter-size scanned documents. To store the same number of paper documents would require two four-drawer filing cabinets which would cost about $400. The cost of electronic data storage is five thousand times less than paper storage.

Costs have dropped consistently 40% per year which accounts for the more than 12,000 times reduction in cost since 1992.  The cost for RAID or mainframe disk storage is somewhat greater, but the historical trend for other storage devices has been similar and the forecast for the foreseeable future is that costs will continue to decrease at the same rate. Twenty years from now we will be able to buy one tera-byte of storage for a penny.

Because of the dramatic decline in data storage costs, it has been easy to address the challenges of data growth, and more recently “big data”, by simply buying more storage.  Why worry about managing the growth of data when it’s much cheaper to simply buy more storage? The problem is that the cost of data is not just storage.  Other costs include:

  • eDiscovery efforts to support litigation. It takes effort and can cost millions in legal fees and award damages. The more data you have the more effort it takes to find what you need, the more opportunity it provides to the plaintiff and the more effort it takes to defend your company against legal challenges.
  • Governing data. It takes effort; deciding who is allowed to access what data, when and where; implementing appropriate security and monitoring controls; making copies of data for training, development or performance reasons, and so on. If two people discuss what to do about data for one hour, and each person’s time costs $50 per hour, then the cost of talking about data is $100 per hour. Because information management involves many people at many levels in an organization, and because many meetings are held at each level, the cost of discussing what to do about data far exceeds the cost of disk storage required to store the data being discussed.
  • Risk and compliance costs. These can quickly amount to more than the data storage costs.  The more data you have the more likely it will be lost or stolen and the greater the chance of non-compliance fines or damage to the corporate brand.  The cost of just one data incident can easily be greater than the enterprise budget for data storage.

The cost of data storage has become virtually inconsequential compared to these other costs so we should stop focusing on storage costs. Rather, we should focus on effective data governance and Lean Data Management.

Lean Data Management is a sustainable approach to treating data as an asset across the enterprise.  It is a method of applying lean principles of waste elimination, automation, continuous improvement and staff empowerment to data management and data governance processes. The end result is a data-driven organization with increased organizational agility to react to changes and opportunities faster than the competition. So let’s stop focusing on the cost of storage, and instead focus on the policies, principles and processes to manage and control data.

FacebookTwitterLinkedInEmailPrintShare
This entry was posted in CIO, Data Integration, Integration Competency Centers, Master Data Management, Operational Efficiency and tagged , , , , , , , , , . Bookmark the permalink.

One Response to Data Storage Is So Cheap Its Expensive

  1. Bill says:

    I was a mainframe CICS COBOL programmer in the early 1980s. Storage was a HUGE concern. So much so that all programs and data files were reviewed by a team to find ways to streamline, cut back or save precious space. Today, it is easier to save EVERYTHING. Then, find out what you really need. Data governance and Lean Data Mgt (new words but the same meaning) are still needed in today’s environment. What goes around comes around.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>