This page contains a Flash digital edition of a book.
Issue 1, December 2008


FOCUS ENERGY ASSURANCE


EYP suggest an ‘onion’ approach to Data Center Energy Strategy, starting from the core IT process and moving outwards through air management to mechanical, electrical and mains supply.


• Mechanical systems including cooling, fans, humidifiers and fuel


• IT systems and equipment including fans, drives and processors


• Electrical systems including UPS, generators and lighting.


From this it is clear that the priorities for data center energy saving can be ranked in order of magnitude as IT energy, followed by cooling energy, then fan/ UPS energy.


It is important to recognise that for each of these key areas of energy (and cost) saving that the tools and metrics exist for achieving those efficiencies, namely,


1) Server input power and CPU activity enabling load reduction through rationalisation or consolidation (e.g., the Electric Power Research Institute have identified that input power to servers can remain virtually constant irrespective of workload, while data center servers typically have 10% utilization).


2) Server electrical performance


3) Data center air stream and air management modeling and metrics to enable the more effective targeting of air cooling to the IT equipment (cool air bypassing the IT directly back to CRAC/air handling units, coupled with hot air re-circulating directly to the server load results in the increased possibility of ‘hotspots’ and the reduced potential for energy savings by setting higher set points).


4) Metrics for evaluating the mechanical systems performance for different chilled water and DX CRAC cooling methods relative to one another and to develop effective cooling strategies through increased free-cooling and the identification of the highest air supply temperature set points possible.


DEVELOPING EFFECTIVE ENERGY STRATEGY


EYP suggest an ‘onion’ approach to Data Center Energy Strategy, starting from the core


IT process and moving outwards through air management to mechanical, electrical and mains supply (figure 2).


The strategy involves:


IT; greater utilization of IT equipment, increased business output (i.e. “work” or compute cycles) per energy consumed and a wider range of temperature and relative humidity to allow more flexible data hall conditions.


Air Management; reduce negative pressure flow and air velocities near CRAC units; reduce bypass flow and reduce recirculation (using physical elements such as blanking plates to segregate hot and cool air streams); ensure sufficient cool air reaches server inlets (i.e., not too much, not too little); ideal air management will physically segregate hot and cold air streams, varying the supply to match the requirement of the servers and enable adequate control of interfaces between server and CRAC fans.


Mechanical; increase chilled water/ air supply set points; use free cooling and increase temperature set points to improve the energy operation of a system.


Electrical; increase part load efficiency of UPS and enable energy saving features; eliminate redundant plant and partial loads, rightsize electrical system to load requirement; use more efficient lighting and automatic operation.


Lastly: Mains Energy; consider energy supply options such as renewable options as a means to reduce overall CO2


emissions.


Embodied Energy; the impact of embodied energy is considerable in the building services industry, in fact it has been estimated that it could be of the same order of magnitude as IT or mechanical energy for the data center. This source of inefficiency requires more detailed consideration in the future.


FURTHER THINKING ABOUT RELIABILITY AND BUSINESS STRATEGY


In tandem with efficiency strategies reliability strategies need to be developed which permit systems to meet both objectives. It is possible, for example, to design reliable and energy efficient systems, but it may require a paradigm shift to design more in terms of part-load operation, for example, the use of


mechanical cooling systems with all-variable pumps, fans and compressors which enable higher efficiencies at part loads.


The information provided by energy assessments will ultimately prove valueless unless supported by financially prioritised recommendations for improvement. Business strategy needs to include the business growth in terms of data center infrastructure systems (M&E or power and cooling systems) together with IT systems efficiency. Figure 3 gives an example of the interplay between these factors in strategy development. It indicates utility power (or average power which is energy) in terms of IT power. The series of lines that start at the origin are constant DCiE or PUE lines.


The graph indicates several scenarios (0 to 5). The full line (3, 0, 1) represents how the data center energy will vary with IT energy and it allows for parasitic constant M&E loads. The dashed line (5, 4) represents how an improved M&E systems loads will vary with IT loads. The graph provides an example of how the data center load could increase from ‘0’ to ‘2’ without doing any M&E and IT improvements and allowing the business to grow. The IT improvements could include rationalisation, consolidation and virtualisation. Alternatively the business could grow as much from ‘0’ to ‘5’ but the data center energy and power consumption would be considerably reduced if M&E improvements together with IT improvements were introduced.


Approaches aligned with those outlined in the White Paper suggest that the ideal data center with negligible power and cooling losses to provide power to IT equipment is conceptually possible and therefore feasible in the future. 


Robert Tozer Ing MSc MBA PhD CEng MCIBSE MInstR MASHRAE,


(1) Data centerDynamics research into Western Europe and North America 2008.


MORE DETAILS


HP Critical Facilities Services, delivered by EYP Mission Critical Facilities, London, UK


Tel +44 (0)20 7796 5000, rtozer@eypmcf.com


The full White Paper ‘Data Center Energy Strategy’ will be available shortly on datacenterdynamics.com The EYP White Paper discussing air conditioning design for data centers is also available at: www.datacenterdynamics.com


www.datacenterdynamics.com


45


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68  |  Page 69  |  Page 70  |  Page 71  |  Page 72