Power and Cooling

Managing older data centers in a dynamically growing healthcare organization provides many challenges. UPMC is an $8 Billion, 50,000 employee health system supporting more than 20 hospitals in the US and Europe along with a 1.5 Million Member Insurance Plan. This session will detail the planning and results of three major ongoing projects and their impact on the organization’s main data centers: remote data center consolidation and hardening; server and storage virtualization; and energy saving Green projects.

Generator sets are vital parts of the Data Center, but since they are not used 99.9% of the time, some issues can be missed. Additionally, suppliers do not always understand what types of detailed monitoring requirements and protocols would best support the data center infrastructure. This presentation will include the Ethernet network infrastructure, various protocol options such as Modbus, BACnet, and SNMP, and building management system integration. In addition, alarm notification via email, text, or SNMP traps will be discussed in detail.

Trinity Health recently developed a simple method to identify and compare the four basic components of data center capacity (UPS power, generator power, cooling, and space) and use that information to drive capacity management.  This case study will review how Trinity Health, using this technique,  was able to assess and compare its data centers (ranging from 1,000sf up to 15,000sf) on an apples-to-apples basis, and how that enabled them to increase capacity in UPS power, generator power, cooling, and space.

Direct current can impact the data center in numerous ways, including: reducing energy consumption from both the electrical and mechanical load; lowering capital costs; lowering installation costs;  improving reliability; and lowering  maintenance costs. This session will review how direct current comes together to deliver on these impressive benefits.

Electric grid emergencies are rare, but if they happen, there are ways you can work with your power providers to lessen the risk. With increasing demand straining power grids across North America, many local utilities and regional grid operators have created free, voluntary programs that pay data centers for agreeing to switch to standby power BEFORE blackouts occur. Learn how and why hundreds of data centers are participating in these programs, and how you can decrease risk, test standby generator availability, and improve cash flow at your facility through demand response programs.

Traditional air conditioners positioned up against or on the walls of data centers use up much of their capacity cooling areas that don’t require it. A more efficient and effective method of cooling is to remove heat at the device. This session reviews how to employ an energy efficient intelligent cooling approach for data centers with a definable and immediate reduction in cost of ownership.

Changes in the ASHRAE guidelines for temperature and humidity control in data centers has opened up new methods of cooling for many areas of the country using adiabatic evaporative systems. These are cooling systems that pass  warm air through a wet media and then emit the vapors to cool the air This presentation will provide an analysis of the different technologies that can be used as well as cover the pros and cons of each.

Most existing data centers are not using efficient cooling techniques. Overall energy savings opportunities exceed 20% in most existing data centers. Most of the programs being developed today are focused on new installations leaving incredible opportunities in existing legacy facilities. This presentation provides practical, real world examples of savings opportunities for cooling systems. Most celebrate success on the raised floor but the benefits extend back into the central cooling system and these will be explored. Attendees will leave with at least three practical ideas they can bring back to their sites.

The key to driving down data center operational cost is directly related to optimizing the cooling system design. Understanding the impact of decisions made during the initial design phase, prior to implementation, can pay large dividends over the long term. Using predictive modeling tools to help understand the impact of design decisions upfront is becoming easier and more cost effective than ever before. In this session participants will see several real world designs analyzed and operational costs predicted and compared against one another.

Power management features within IT equipment have evolved substantially over the last few years. When not needing full performance, servers can scale back their power usage to 40%, for instance, and save substantial amounts of energy. When needing to cap power at the server, at the rack, or at the pod level, it is now possible to do this intelligently by assigning priority levels to differing applications. However, many data centers fail to take advantage of these powerful features, perhaps due to concerns that performance will be adversely affected or because some do not know they exist. Learn about the power savings options available and how you can implement them in your data center.

Reducing the energy consumed in maintaining proper temperature and humidity levels can be achieved by using several innovative designs and specialized equipment available today. These include controlling a fan’s operation via static pressure, using fan speed controllers (inverters & electrically commutated motors), determining the proper locations of control sensors and taking advantage of several “free cooling” options.  These methods can be incorporated into any new or pre-existing centers looking to reduce their overall energy consumption.