Energy concerns impact secondary storage

By Heidi Biggar, Mary Johnston Turner, John McKnight

While much of the recent attention paid to data–center power and cooling concerns has focused on servers and primary storage systems, a recent Enterprise Strategy Group (ESG) Research survey of 398 North American IT decision–makers found that energy–related issues are also affecting organizations’ secondary storage purchase decisions. This is particularly true for large organizations that generate and manage the greatest volumes of data (see figure).

As the figure demonstrates, power and cooling concerns are leading many organizations to consider the physical footprint, energy efficiency, and capacity reduction features of their secondary storage systems. As a proof point, 34% of respondent organizations with more than 25TB of total storage capacity said that the secondary storage system’s physical footprint was now a more important consideration in the purchasing process, while 32% say they are now more likely to weigh the energy efficiency of the storage system itself.

In addition to a closer evaluation of storage system attributes, survey respondents report interest in new data–protection technologies and processes that can also reduce power and cooling requirements. For example, 26% of respondents with more than 25TB of capacity said they had evaluated capacity reduction technologies (e.g., data de–duplication) and 15% said they were evaluating new secondary storage alternatives (e.g., MAID, or massive array of idle disks) directly as a result of power and cooling concerns. Capacity reduction technologies minimize disk capacity requirements, which translates to a smaller overall storage footprint and less electricity needed to power and cool a storage system. Secondary storage alternatives such as MAID match drive activity to I/O activity. Rather than spinning drives continuously as with traditional storage systems, MAID systems spin drives only when needed. All things being equal, less electricity is needed for power and cooling.

Rising electricity costs, issues with limited power supply or floor space, growing data volumes, and longer data retention periods also affect small organizations, but generally to a much lesser degree than they do large organizations. For example, only 14% of small organizations (i.e., those with less than 5TB of capacity) said that physical footprint was an important consideration in the purchasing process, and only 6% and 4% of these lower–capacity organizations said they had evaluated capacity reduction technologies and storage systems alternatives (e.g., MAID), respectively.

Perhaps the most compelling evidence of the relationship between storage capacity and power and cooling concerns can be seen in the finding (see figure on p. 1) that only 34% of the organizations with more than 25TB of capacity said power and cooling concerns have had no impact on their secondary storage purchasing decisions, compared to 59% of organizations with 6TB to 25TB and 70% of respondents with less than 5TB. As data volumes continue to grow, even those organizations with relatively small capacity requirements today will soon face the same power and cooling challenges as their larger counterparts.

Financial services leads the way

After analyzing this data by industry, ESG found that financial services organizations are more likely to consider energy–related factors when evaluating secondary storage systems than other industries:

  • One–third (33%) of financial services organizations said energy efficiency has become an increasingly important consideration in the purchasing process (see figure);
  • Nearly one–third (30%) of financial services organizations said the physical footprint of the storage system was an important factor in their decision–making; and
  • Nearly one–quarter (23%) of the financial services organizations surveyed said they have evaluated capacity reduction technologies (e.g., data de–duplication) specifically due to power and cooling concerns.
Click here to enlarge image

This data is not surprising. In addition to the power and cooling loads generated by their advanced computing infrastructures, financial services organizations are under more pressure than ever when it comes to information storage and management. These firms are generating huge volumes of data and are being asked to store this data for longer periods of time due to regulatory compliance, corporate governance, and e–discovery requirements. In addition, financial organizations are often clustered in energy–scarce metropolitan areas (e.g., New York, San Francisco, London, Tokyo, etc.) and tend to be early adopters of new IT technologies. All in all, it is not surprising that financial organizations are looking to data de–duplication, MAID, and other technologies to help minimize their energy footprints. ESG believes both IT vendors and end–user organizations in other, less data–intensive industry verticals can and should learn from the experiences of these financial services early adopters with respect to their strategies for reducing data–center power and cooling requirements.

Research implications

Rising energy prices and growing environmental concerns will drive awareness of power and cooling costs associated with all elements of the data center in the coming years. Organizations seeking to cut costs and improve the energy efficiency of their IT infrastructure should look beyond servers and primary storage systems to also evaluate

  • The energy efficiency and operational power costs of secondary storage systems;
  • The physical footprint and heat generated by secondary storage systems as they relate to floor space, density, and data–center power and cooling load and costs; and
  • Data de–duplication and other data reduction technologies that can slow system growth and keep power and cooling consumption level or shrinking.

ESG’s research reveals that leading–edge financial services organizations are already well–aware of the savings and opportunities associated with detailed power and cooling evaluations and are changing their storage evaluation and purchasing criteria accordingly. Other industries should learn from these organizations—applying best practices to their environments where appropriate.

Heidi Biggar is an analyst; Mary Johnston Turner is a senior analyst, and John McKnight is research director at the Enterprise Strategy Group (www.enterprisestrategygroup.com).

This article was originally published on May 01, 2008