Identify and Remove Unused Servers

Many servers aren’t doing useful work, yet they consume electricity 24/7

Surveys of data centers often identify aged servers with no use still running – so-called "comatose" servers. “Decommissioning” involves finding, turning off, and removing these unused servers.  Estimates of the prevalence of comatose servers vary.

  • A 2015 study from the Anthesis Group and a Stanford University Researcher Jonathan Koomey revealed that 30 percent of physical servers in data centers were “comatose,” meaning they have not delivered information or computing services in six months or more.1
  • One New York Times article described a large data center near Atlanta that found more than half of its servers were comatose.2

Comatose Servers

Where “comatose” servers come from

In addition to the one workload, one box approach, (which has led to server sprawl and low utilization of servers) a Green Grid study3 points to the following reasons for the "comatose" server phenomenon:

  • Overworked IT staffs are busy maintaining or upgrading existing devices and deploying new products. Unused servers are not an urgent priority
  • Traditional monitoring tools focus on server availability and performance and can overlook unused servers as a result
  • Concern about honoring Service Level Agreements (SLAs)4 may make IT staff hesitant to turn off any server
  • IT staff may not be diligent about tracking virtual machines. For example, staff may spin up a virtual server for a specific test and leave it running after the test is completed
  • IT staff may like to have spare servers available and running "just in case" there is a spike in user demand, a need to migrate back to the old server, etc.

Identifying Comatose Servers

A survey cited in the same Green Grid study revealed that one-third of IT managers had not yet tried to identify unused servers.  The remaining two-thirds assume a server is unused and can be removed when:

  • No complaints are received following an unplanned outage (1%)
  • No complaints are received after switching a server off (9%)
  • Very low utilization is reported through automated monitoring tools (22%)
  • Application owners, polled periodically, indicate an application is no longer being used (35%).

Savings and Costs

Decommissioning allows you to retire servers and/or defer purchases of new servers, thus directly decreasing electricity consumption. Reducing the number of servers in a data center also allows for a smaller power infrastructure.  As a result, less energy is consumed by power distribution units, UPS systems, and building transformers.   And because every server produces a lot of waste heat, fewer servers means that the data center needs less air conditioning. As a result of these indirect energy benefits, saving one watt-hour of electricity at the server level typically results in an additional 1.9 watt-hours of electricity savings at the facility-level!5

  • According to the Uptime Institute, decommissioning a single 1U rack server can annually save $500 in energy, $500 in operating system licenses, and $1,500 in hardware maintenance costs.6
  • Barclays, a global financial organization, removed 5,515 obsolete servers in 2012, gaining power savings of around 3 megawatts, $3.4 million in annualized energy savings, and a further $800K savings in hardware maintenance.7
  • McKesson, the health care company, decommissioned 586 servers, reducing data center power usage by 931.7 kilowatts and saving $734,550.8

Tips and Considerations9

  • Seek and obtain upper-management support. This will help IT staff shift their existing workloads to accommodate a decommissioning effort.  Also, communicating total cost of ownership and energy efficiency messages from the top down can help alleviate potential end-user frustration at any resulting service impacts.
  • Take inventory. It is easy to lose track of servers as time goes by, as mergers and acquisitions take place, as people procure servers without going through central IT, or as servers are retired or repurposed. The first step toward consolidation is identifying all servers in the organization. Take an inventory of all systems and servers, their associated application workloads, and note the equipment type, location, and Service Level Agreements (SLAs).  Server usage can change often, so establish a policy for regular SLA reviews and inventory updates.
  • Leverage DCIM systems.  Automated monitoring tools such as Data Center Infrastructure Management (DCIM) solutions can identify servers with very low CPU utilization, indicating that the server may be unused. 
  • Look at connectivity to servers.  Specifically, examine the amount of inbound network activity related to secondary and tertiary applications, such as traffic coming in from the backup server, domain controller, antivirus server, etc. If that’s all the traffic that a server shows, then it’s likely that it is not in use.
  • Mitigate risks.  Virtualization, with the ability to deploy a virtual server in minutes, can offset the risks of mistakenly decommissioning server that was only temporarily unused.
  • Have a strategy for server reuse and disposal. When an unused server is discovered, a decision needs to be made about what to do with it.  Electronics – data center equipment included – cannot be thrown in the trash. Every electronic device contains hazardous materials like lead, cadmium and sometimes even mercury. If electronics end up in landfills, these hazardous materials will eventually begin to contaminate our environment    Options for an unused server include:
    • Decommissioning it by taking it out of the rack and recycling it
    • Turning it off but leave it in place, so it is not consuming power but is ready to be re-purposed.
    • Re-using it by re-imaging the server and redeploying it in another capacity.

1 30% Of Servers Are Sitting "Comatose" According To Research, by Ben Kepes, Forbes, June 3, 2015.

2 Power, Pollution and the Internet, New York Times, by James Glanz, 9/22/2012.

3 Unused Servers Survey Results Analysis, The Green Grid, 2010.

4 A service level agreement (SLA) is a contract between a service provider (either internal or external) and the end user that defines the level of service expected from the service provider.

5 New Strategies for Cutting Data Center Energy Cost and Boosting Capacity, Emerson Network Power presentation, 2012, p.8. (PDF, 1.4 MB)

6 Decommissioning as a Discipline: Server Roundup Winners Share Success, by Matt Stansberry, Uptime Institute, undated.

7 Decommissioning as a Discipline: Server Roundup Winners Share Success, by Matt Stansberry, Uptime Institute, undated.

8 Decommissioning as a Discipline: Server Roundup Winners Share Success, by Matt Stansberry, Uptime Institute, undated.

9 Section was developed using the following white paper as a guide: Unused Servers Survey: Results Analysis, by Mark Blackburn, Green Grid, 2010.