IT teams have long been trying to economize energy consumption in data centers, but they still have a way to go. Here are eight approaches that IT can use to further green initiatives.

Mary E. Shacklett, President of Transworld Data

June 15, 2021

6 Min Read
Credit: wajan via Adobe Stock

Green IT originated in 1992 when the US Environmental Protection Agency created its Energy Start program that focused on reduced energy consumption in home appliances such as refrigerators and air conditioners. The program was later extended to computer equipment in 2006.

Since then, green computing has been a persistent topic in data center discussions and initiatives, with mixed results. Computer manufacturers have reduced energy consumption of the devices that they produce when the devices go idle but manufacturing the devices themselves remains energy inefficient. The Internet of Things (IoT) relies on sensors that can turn themselves off to reduce energy consumption and prolong battery life when they are not in use -- but at end of life these devices still need to be disposed of. Meanwhile in data centers, most companies have moved to the cloud and increased computer virtualization, but large swatches of computing resources and storage remain as allocated energy consumers long after usage has ceased.

The result is a mélange of successes and shortfalls that were created by green computing not being an utmost IT priority and executive management not being overly convinced of investment returns.

Despite these mixed results, green computing remains a goal for companies and for technology vendors, and the energy findings of a 2007 Storage Networking Industry Association (SNIA) study still stand. In the study, which focused on green data centers, SNIA singled out buildings that host data centers, noting that buildings in the US used 71% of all electricity produced and were responsible for 39% of carbon emissions. Data centers ranked fifth in energy consumption; and among the biggest concerns that data center managers had were heat density/cooling (22%), space constraints/growth (19%) and power density (18%). The IT processing load was identified as the largest energy consumer (46%)

In response, computer manufacturers reduced energy footprints and consumption, and IT took a stab at economizing energy consumption in data centers. But there is still work to be done.

Here are eight ways that IT can use to further green initiatives:

1. Include green computing in vendor RFP discussions

Talking energy consumption with hardware vendors is a natural fit, but what about software vendors? With the introduction of more low-code and no-code tools, one side effect is more application overhead that is generated by surplus code that a native coder wouldn't include. This forces extra processing and energy consumption. If you’re moving to more low-code and no-code tools, discussing this overhead (and how it can be tuned for efficiency) should be a topic with prospective vendors.

2. Identify and eliminate cloud sprawl

With the expansion of citizen development and end users entering their own cloud service contracts, companies are having a hard time achieving efficient, energy-effective cloud usage. Cloud resources that are rarely used should be identified and reviewed to see if they should be de-allocated. The benefit falls directly to the bottom line because it is likely the company will be reducing its cloud spend.

3. Automate virtual system allocation and de-allocation

One advantage of virtualized instances of systems is that they enable application developers to quickly spin up v-systems for testing. Automated system allocation speeds time to market because developers no longer have to wait for a system specialist to do the job for them The catch however, is that once developers are through testing, they forget to de-allocate the virtual test system. These now idle systems consume space and energy, but an automated virtual system de-allocation process can cure this. All IT needs to do is to set a rule, like taking down any virtual instance of a system that hasn't been used for 90 days.

4. Better storage management

Storage is cheap, so the fast way of fixing a storage issue is just to buy more of it. But buying more storage inflates energy bills and is energy inefficient. Because storage is regarded as a cheap and available commodity, storage is seldom on IT strategic plans. It should be. Minimally, CIOs should require an architectural plan for storage that shows which data will be processed on the most expensive  in-memory and solid state storage, and which seldom used data should be relegated to cheap, slow disk and tape drives. Data should be regularly reviewed for usage and possible purges. Older practices such as striping data across multiple disk drives, with each device storing only a fraction of the data it is capable of storing so processing can be faster, should be discouraged.

5. Air cooling efficiencies

There are many vendor options for data center cooling. Of primary importance is selecting a vendor that you feel you can work with, and that really understands the energy and asset preservation needs of your data center.

The data center floor plan should also be reviewed to ensure that maximum cooling is placed in proximity to the most heat-intensive areas of the floor. In all cases, an IoT sensor system should be used to monitor environments.

6. IoT monitoring and disposal

Companies employing IoT at the edges of their enterprises should develop a plan for maintaining sensors in the field or in remote facilities. For energy (and battery) efficiency, an optimal plan would include sensors that automatically turn off or go into sleep mode when they aren’t active. This preserves energy and lengthens the life span of the sensor. At end of life, sensors should be gathered up and disposed of in an environmentally safe manner.

7. Computer recycling and disposal

Relegating older computer equipment to less-taxing functions and then pushing them out altogether is still a common practice in most organizations. What starts out as a beefed-up computer in a super user’s hands becomes an old-gen computer that gets cycled down to someone who just needs to do data entry and modest word processing. From here, the equipment gets auctioned off to employees or donated to a charity. If all else fails, the equipment gets recycled, with many states providing recycling services for this purpose.

8. Energy audits

Companies should arrange to conduct an energy audit of their data centers. A number of professional firms offer this service, and some utility companies do. The audit will likely uncover hidden areas of energy waste or excess consumption that you may have overlooked.

Related Content:

How CIOs Can Advance Company Sustainability Goals

Ways to Put a Lid on Cloud Waste

How to Make Waste-Free Product a Reality in Your Organization

 

About the Author(s)

Mary E. Shacklett

President of Transworld Data

Mary E. Shacklett is an internationally recognized technology commentator and President of Transworld Data, a marketing and technology services firm. Prior to founding her own company, she was Vice President of Product Research and Software Development for Summit Information Systems, a computer software company; and Vice President of Strategic Planning and Technology at FSI International, a multinational manufacturer in the semiconductor industry.

Mary has business experience in Europe, Japan, and the Pacific Rim. She has a BS degree from the University of Wisconsin and an MA from the University of Southern California, where she taught for several years. She is listed in Who's Who Worldwide and in Who's Who in the Computer Industry.

Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.

You May Also Like


More Insights