The Science of Measurement
Improving Data Center Performance with Continuous Monitoring and Measurement of Site Infrastructure
Modius is pleased to make available for download The Science of Measurement, an independent research report, which aims to help data center managers (both IT and facilities managers) better understand the business value of measuring data center performance characteristics and the considerations that are important to developing a measurement scheme.
This report is written by two of the thought leaders in data center efficiency:
- Jonathan Koomey is a Project Scientist at Lawrence Berkeley National Laboratory and a Consulting Professor at Stanford University. Dr. Koomey is one of the leading international experts on electricity used by computers, office equipment, and data centers, and is the author or co-author of eight books and more than one hundred and fifty articles and reports on energy and environmental economics, technology, forecasting, and policy. Learn more at http://www.koomey.com.
- John Stanley currently works at a firm specializing in data center energy efficiency research and consulting. He holds a Master’s degree from the Energy and Resources Group at the University of California, Berkeley, where his thesis research focused on energy efficiency in data centers
Executive Summary from the Report:
This report explains some of the key considerations surrounding performance monitoring and measurements in data centers. The key lessons for data center managers are:
- Measurement is conducted to inform decisions on actions that can enhance a data center’s business value by better accommodating growth, reducing costs, or increasing uptime.
- There are opportunities to improve efficiency through a continuous process of incremental changes, as well as through the more traditional “one shot” project-based approach.
- The business or strategic objectives of the measurement effort should drive measurement decisions, and having “the right data” is more important than simply having “more data”
- Data center managers should think carefully before beginning a monitoring program, so that the data collected will be maximally useful for judging operational performance and making decisions.
- After collecting data, managers should take action based on what they have learned. Managers should then continue collecting data to get feedback on the effectiveness of their actions.
- When evaluating potential monitoring systems, managers should examine multiple factors, including:
- Ability to collect data from all desired devices
- Granularity of data collection
- User friendliness and ease of integrating data across devices and time scales
- Scalability for mass deployment and multi-site capability
- Adaptability to new measurement needs
- Trending and analysis of data
- Integration with control systems
- Ability to detect problems and notify data center operators
Read the entire 18-page report.