Optimizing Data-Center Performance in 2017

Optimizing Data-Center Performance in 2017

If your goal is to increase the uptime, reduce the energy use, and minimize the operational costs of a mission-critical facility, here is some advice.

If your goal is to increase the uptime, reduce the energy use, and minimize the operational costs of a mission-critical facility, here is some advice:

Certify What You Are Buying

Sacramento, Calif.-area mechanical contractor Cooper Oates Air Conditioning lowers a Liebert unit through a roof.

Historically, data-center operators had no way of knowing the actual capacity and efficiency of the cooling solutions they were purchasing—unless they spent months and tens of thousands of dollars on independent testing. As a result, cooling often was overprovisioned and overdesigned.

To address this issue, Air-Conditioning, Heating, and Refrigeration Institute (AHRI) created ANSI/AHRI Standard 1360, Performance Rating of Computer and Data Processing Room Air Conditioners. Now that certification for data-center and telecommunications cooling equipment exists, specifying engineers, mechanical contractors, and their customers can be assured they are getting the capacity and efficiency they expect without the need to overprovision or perform expensive independent testing.

Be sure your equipment is AHRI-certified.

Autotune Your Data Center

Whether it is a direct-expansion (DX) compressor short-cycling because of low load or a chilled-water system “hunting” for the right valve position, data-center cooling equipment regularly over-responds to changes in load, outdoor temperature, and equipment configuration. These irregular changes waste energy, increase system wear, and result in temperature inconsistency across data centers.

Typical data-center cooling systems have separate control systems for mechanical components such as chillers, computer-room air-conditioning units, and economizers. These control systems generally are not coordinated and often fight each other. As equipment is added, controls are modified, or loads change, the problem is compounded and components oscillate. This issue either goes unnoticed or requires time-intensive manual tuning to resolve.

One way to address these issues is with system autotuning, which applies machine learning to regulate DX compressors or chilled-water valves in a way that brings the rest of the system components into balance and stabilizes cooling capacity.

For DX systems, it means harmonizing compressors, fans, and condensers to eliminate costly short-cycling. For chilled-water systems, it means eliminating valve hunting to balance fan speeds, water temperatures, and flow rates. Autotuning does this in real time, responding to changes in setpoint, data-center load, outdoor temperature, and other conditions.

System autotuning constantly identifies and monitors unwanted oscillation patterns in components and uses historical data from internal and external system sensors to automate retuning of control timing to bring a system into a stable steady-state condition. Its dynamic intelligence provides just the right amount of tuning needed to accommodate rooms of different sizes, which can respond very differently to environmental changes.

Benefits of system autotuning include:

  • Up to a 15-percent improvement in cooling-system efficiency.
  • More stable supply-air temperatures.
  • More stable cooling capacity.
  • Reduced startup and commissioning time.
  • Extended cooling-system life.
  • Operational-cost savings.

Take Full Control

The newest thermal controls, operating at both individual-unit and system levels, utilize machine-to-machine communication and advanced algorithms to provide powerful analytics and automated self-healing routines. They monitor hundreds of unit and component points to eliminate single points of failure while simplifying operations to save time and reduce human error. Not only does this provide greater system protection and increase energy efficiency by up to 50 percent compared with legacy technologies, it provides continuous insight into and tracking of thermal conditions and system operations.

Here is an example of potential payback from adding thermal-system controls to harmonize the operation of multiple cooling units and eliminate cooling and heating conflicts: An enterprise data center with 500 kW of information-technology load and energy costs of 10 cents per kilowatt-hour can lower annual thermal-energy consumption from 380 kW to 184 kW, yielding $171,690 in savings. That will lower mechanical power-usage effectiveness by more than 20 percent, from 1.76 to 1.37.

Conclusion

These are just a few of the thermal innovations for enhancing the performance and reliability of mission-critical applications available today. In the coming months and years, you will see new technologies and strategies for eliminating downtime, reducing energy use, and minimizing operating costs.

 

JP Valiulis is vice president, North America thermal-management marketing for Vertiv.

Hide comments

Comments

  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Publish