I would like to measure the power efficiency of servers (virtualized) using an opensource management tool like Nagios or OpenNMS. Cacti also might be useful for it's historical graphing capabilities, but it doesnt seem to be as flexible as Nagios with regards to plugins etc.

I do not have physical access to the server, only remote access. How do I capture the required info and report on using these tools? I am from a network management background so not too familiar with server power management.



closed as off-topic by HopelessN00b Jan 22 '15 at 3:32

This question appears to be off-topic. The users who voted to close gave this specific reason:

If this question can be reworded to fit the rules in the help center, please edit the question.

locked by HopelessN00b Jan 22 '15 at 3:32

This question exists because it has historical significance, but it is not considered a good, on-topic question for this site so please do not use it as evidence that you can ask similar questions here. This question and its answers are frozen and cannot be changed. See the help center for guidance on writing a good question.

Read more about locked posts here.

  • Also need to add that I am considering ZenOSS Core as well..... Any suggestions will be appreciated! :) – KateLibby Feb 16 '10 at 20:13

I'm not sure what you're asking, and I'm not 100% sure if you are either :-) Some general information that may be helpful:

Power consumption needs to be metered at the distribution point (either via the CDU in the rack, or a meter in-line with the power cords) -- I don't know of any server hardware that reports power supply current draw, though I could be missing something.

"Power efficiency" is a fuzzy value for a virtual server as electricity usage is inherently physical -- If you must have a number for a specific VM your virtual machine could be said to consume 1/Nth of the power being drawn by the physical equipment hosting it (where N=Number of virtual servers hosted).
Note that this needs to take into account all supporting infrastructure (e.g. SANs) used in the virtualization environment, and even then it's an extremely rough value since it doesn't take into account what percentage of the load on the virtualization environment is created by a specific VM: It just splits the power consumption up evenly.

Relative efficiency of virtualized servers as opposed to the equivalent physical hardware is usually expressed as a power savings - The difference between the power draw of the fully-physical environment and the virtualization environment. If you never had a fully-physical environment you can estimate that value (number of virtual servers * average power consumption of a server).

  • Thanks - that makes sense. I understand it needs to be measured at the distribution point, but was hoping I could get that info though some type of MIB or through the use of a daemon etc. Ultimately I would like to take measurements (like you said you could divide the total power consumption by the number of VM's) or load more VM's and see what the effect is on the power usage vs just adding more servers. Maybe use Matlab for doing these graphs. Thanks again! – KateLibby Feb 17 '10 at 8:23