Okay, I get forgetting to pick up that sixteen cents if you’re in a hurry.
I don’t get giving up tens or hundreds of thousands of dollars, even if it isn’t yours.
Some of the losses can be easily prevented. Installing blanking panels in the rack, for example, helps to prevent the mixing of hot and cold air and reduces your energy usage at a minimal cost. Purchasing more energy efficient hardware may cost more up front, but will save you in energy costs down the line. Raising the temperature in the data center by 1 degree C can save as much as 4% in energy cost. Eliminating ghost servers can reduce server energy use by 15-30%. Knowing where equipment is located and how it is configured can save time in troubleshooting issues.
So why aren’t these things being done in every data center?
One of the reasons is fear of downtime. What happens if I increase the temperature and one of my CRAC units fails and systems go down as a result?
Another reason is lack of information. How can I find out which servers might not be running any applications? How do I know how many systems are running this software?
In many cases, the real reason is that the data center management team doesn’t have a tool which can help them. That’s where DCIM (Data Center Infrastructure Management) comes in. DCIM provides an efficient way to quickly find the information you need. DCIM can tell you where a server is located, its hardware configuration, what software is installed, what virtual machines it is hosting, and how it is connected (both power and network). Wait, that sounds like a lot of work! How is that information going to get into the DCIM tool and what is my assurance the data will remain accurate? That’s where auto-discovery comes in. Auto-discovery can quickly collect the detailed data you need and automatically update the information when it changes (Note: be careful because Not All Auto-Discovery is Created Equal).
How else can DCIM help us recover those lost pennies and nickels?
- By monitoring CPU usage and other server parameters, DCIM can help to identify ghost servers.
- By tracking software installations, DCIM can help you to purchase the number of software licenses you actually need.
- By tracking temperatures throughout the data center (particularly the inlet temperatures at the rack), DCIM can help you to identify hot spots.
- By monitoring facilities equipment, DCIM can quickly alert to system events such as a CRAC failure or a UPS going on battery, allowing you to avert potential downtime by quickly responding.
- By tracking energy usage and efficiency, DCIM allows you to see the results of energy efficiency initiatives.
- By automating the change management process, DCIM increases availability.
By the way, if you’re wondering where that loose change goes, it remains with the TSA. The TSA Loose Change Act - which would have required the TSA to give the money to private charities that provide travel-related services to the members of the U.S. military and their families - passed in the House of Representatives but died in committee in the Senate.