• Overview
• Benefits
  • Features
• Purchasing
• Services
  • Scope of Maintenance Services
  • System Requirements
• Support
• Partners
•  

These two acronyms have become benchmarks and performance indicators for data centres. They are efficiency measures and it doesn't matter which is used as they are both interrelated being calculated from the same measurements of power used by a data centre. Their values are calculated as follows:

PUE	=	Total Facility Power
		IT Equipment Power
DCiE	=	IT Equipment Power		X100%
		Total Facility Power

These efficiency metrics are therefore determined by the ratio of the total energy used by a data centre, including IT equipment, and the energy consumed just by the IT equipment alone. The Total Facility Power includes lighting, cooling, air movement equipment and inefficiencies in electricity distribution within the data centre (UPS losses etc.). The IT equipment portion is the electrical energy that reaches IT equipment to carry out computational tasks. This measurement is usually taken before the power supplies on IT hardware so does not allow for losses in IT platform power supplies.

Much has been written and discussed recently about the value of these ratios. There are flaws as mentioned above with the measurements not accounting for IT platform design inefficiencies and no account is made to determine if the electrical power is actually being used for real computational work. So, as true efficiency metrics they have some limitations.

However, they are becoming widely quoted by data centre operators and they do provide a good indication of the modularity and effectiveness of cooling systems serving data centres. Large open room facilities with a "just to be sure" approach to cooling will show poorly using these metrics unless they are fully occupied. Contained cooling supplying the areas of space occupied by IT systems with active management of set points will perform better.

From this, it is clear that to improve performance against these benchmarks much tighter control over the utilisation of data centre space and cooling will be required.

Benefits

From the above explanation you can see how a virtualisation project with an objective to consolidate server hardware could adversely affect these efficiency metrics if no further action is taken around cooling and space management. As well as decommissioning old servers after virtualisation, moving other servers to optimise space use vs. cooling may also be required as part of the consolidation process. This is where a Data Centre Infrastructure Management system such as dcTrack™ should provide benefits as follows:

• Assist in identifying location of equipment vs. space, rows, aisles and any air containment that may be deployed such as contained hot aisle/cold aisle or perforated floor tiles.
• Provide a central information point where additional temperature measurements in real time can be viewed in relation to IT assets assisting in management of temperature set points.
• Assist in placement of new systems to maintain cooling optimisations.
• Provide a central point to manage and automate required work orders to ensure planned changes are carried out efficiently.

ROI

Tighter control of cooling and management of set points will deliver cost savings see 'Blowing Hot or Cold' or download Raritan's own data centre optimisation study (PDF). To capitalise on these savings better space management and placement of IT systems will be required. dcTrack™ includes the views and measurement capabilities to help you optimise your data centre.

Ask us for more information on how dcTrack™ can help as part of your data centre space management programme