Accurate data center power monitoring is critical for managing modern data center environments. With many data centers rapidly running out of capacity and energy costs increasing, it’s more important than ever to know how much power is being used, when you expect to run out of capacity, and where you can make improvements to boost data center efficiency.
Data center power monitoring can help you with all these things and more. However, with the massive amounts of data points generated by your PDUs, UPSs, CRACs, and other devices, you need data center monitoring software to help collect the data and then transform it into information. That’s where Data Center Infrastructure Management (DCIM) software comes in.
How DCIM Software Enhances Power Monitoring in Your Data Center
Most modern data centers house a significant number of devices, including rack and floor PDUs, UPSs, CRACs, power panels, standalone meters, environmental sensor aggregators, and other facility items. The wide range of devices results in similarly large amounts and variety of data. Each of these devices contributes to your power usage, so it’s ideal to be able to collect data from all of them at the most granular level possible for the highest level of accuracy.
DCIM software helps you rapidly discover (through autodiscovery of supported devices on a network) and poll these devices and then use the information to glean insights about power and energy usage in your data center. For example, outlet-level metering is available with some intelligent rack PDUs (iPDUs). When combined with DCIM software, you’re able to know not only the exact amount of power consumed but also who is using it.
In addition to monitoring the power usage of the devices in your data center, DCIM software takes the data collected directly from your devices and displays it in dashboards and visualization charts. You don’t need to worry about inaccurate data or poor data quality with DCIM software since the data is taken directly from the devices. Based on this data, visual analytics are created automatically by the software to help you rapidly engage with and analyze your data for deeper, more meaningful insights—if you know what to look for.
Key Metrics for Data Center Power Monitoring
Without knowing which metrics to track and how to leverage your data, the visualizations and dashboards in your DCIM or data center monitoring software are nothing more than pretty pictures. So, what are some of the metrics and analytics you should consider for data center power monitoring? Your dashboards and Key Performance Indicators (KPIs) will vary based on your objectives, but most modern data centers can benefit from tracking and analyzing the following:
- Power Usage Effectiveness. Although some organizations are moving away from PUE as a key energy metric, data center power monitoring experts still consider it useful. Power Usage Effectiveness (PUE) is a measurement of the total energy of the data center divided by the IT energy consumption. The Department of Energy spotlights PUE as a key metric for data center energy management and performance assessment. For the same reason, PUE is included among five key efficiency metrics measured in the Data Center Optimization Initiative (DCOI).
- Power capacity trends and analysis. A lack of visibility into when you need more power capacity can make it difficult to keep energy costs down while maintaining uptime. Trending power capacity over time can help you forecast your power consumption more accurately, so you’ll have a more accurate window of when you’ll run out of capacity and won’t have to order more at a premium price or on short notice. Combine this with failover simulation analysis to ascertain whether you have enough resources to handle a failover situation and eliminate outage concerns.
- Active power by month and device. Tracking active power over time can help you quickly identify spikes, prevent potential power issues, and maximize uptime. Leverage power monitoring at the server level to identify power consumption by server, which will allow you to decrease energy costs by shutting down ghost servers and replacing power hogs with more efficient devices or virtual servers. Tracking active power by device or even more granular levels can also help you more accurately billback customers for their energy usage.
Using DCIM software to monitor power usage is a key step to boosting efficiency, decreasing costs, and getting a better handle on your data center energy management. Once you have power monitoring software in place, you can transform your information into actionable insights that will enable you to optimize your data center.
Published May 12, 2017; Updated August 3, 2018.