Design for Efficient Operational and Energy Management
While most data centers have some basic form of Build Management System (BMS), any new design needs to include a highly granular network of sensors in virtually all of the systems and sub-systems of the facility power and cooling infrastructure. This is the fourth article in a series that is titled, the DCK Executive Guide to Data Center Design.
May 14, 2013
![Design for Efficient Operational and Energy Management Design for Efficient Operational and Energy Management](https://eu-images.contentstack.com/v3/assets/blt8eb3cdfc1fce5194/bltd080b25559b9bd65/66212642632720393ade31af/emc-durham-racks.jpg?width=1280&auto=webp&quality=95&format=jpg&disable=upscale)
This is the fourth article in a series on DCK Executive Guide to Data Center Designs.
While most data centers have some basic form of Building Management System (BMS), any new design needs to include a highly granular network of sensors in virtually all of the systems and sub-systems of the facility power and cooling infrastructure. Older, general purpose BMS systems typically had simple alarms to warn of equipment failures and perhaps a moderate amount of basic information on energy use. In recent years, it became clear that as data centers grew larger and used more complex systems, it became more difficult for operators to keep track of all the critical infrastructure system operational details on maintenance requirements and energy efficiency.
A newer more sophisticated class of systems designed specifically for data centers have been developed which are known as Data Center Infrastructure Management (DCIM). They not only encompass monitoring energy usage and efficiency optimization, they can improve operational reliability, by early detection of operational anomalies. DCIM systems also can help track and schedule preventive maintenance and spot any trends of recurring problems.
While DCIM software varies widely based on vendor offerings and continues to evolve, the data center design should include pre-installed or provision for sensors at all the critical systems and sub-systems. You need to in review a variety of DCIM vendor offering to see which product features offers what you need. Regardless of your choice, make sure that you include the pre-installation of sensors in the design phase. Adding sensors after building the facility is both costly and can be intrusive if the systems are already operational. Examples would be energy monitoring for every CRAC/CRAH and if a chiller system is involved Chilled water flow metering, as well as energy monitoring all the individual components such as compressors, pumps and fans. This information will allow you to optimize cooling system operation and energy efficiency. Moreover, with real-time monitoring and maintenance management you can detect trends and anomalies and proactively address potential issues before they become critical problems.
In addition, every point in power distribution system to the IT equipment should have Branch Circuit Monitoring pre-installed. This will allow you to integrate real-time information from IT systems energy usage and correlate it to computing activity to provide for better capacity planning, resource optimization and avoid islands of stranded capacity and improve facility side provisioning of IT equipment deployments. No new data center should be designed or built without some form of DCIM system as part of the base infrastructure system. While DCIM requires additional investment, it can ultimately lower the TCO by improving operational and energy efficiency, while reducing the number of data center and IT support staff.
The complete Data Center Knowledge Executive Guide on Data Center Design is available in PDF complements of Digital Realty. Click here to download.
About the Author
You May Also Like