Search

The energy demands of data centers (DCs) worldwide are rapidly increasing, as are their environmental and economic costs. This paper presents a study conducted at Sannazzaro de' Burgondi (Po Valley), Italy, specifically aimed at optimizing the operating conditions of a DC designed for the Italian Oil and Gas Company (Eni) (5200 m(2) of Information Technology installed, 30 MW) and based on a direct free cooling (DFC) system. The aim of the study was to save the largest possible quantity of energy, while at the same time preventing aerosol corrosion. The aerosol properties (number size distribution, chemical composition, deliquescence relative humidity (DRH), acidity) and meteorological parameters were monitored and utilized to determine the potential levels of aerosol entering the DC (equivalent ISO class), together with its DRH. These data enabled us both to select the DC's filtering system (MERV13 filters) and to optimize the cooling cycle through calculation of the most reliable humidity cycle (60% of maximum allowed RH) applicable to the DFC. A potential energy saving of 81%, compared to a traditional air conditioning cooling system, was estimated: in one year, for 1 kW of installed information technology, the estimated energy saving is 7.4 MWh, resulting in 2.7 fewer tons of CO2 being emitted, and a financial saving of € 1100.

Vertical profiles of aerosol number-size distribution and black carbon (BC) concentration were measured between ground-level and 500m AGL over Milan. A tethered balloon was fitted with an instrumentation package consisting of the newly-developed micro-Aethalometer (microAeth® Model AE51, Magee Scientific, USA), an optical particle counter, and a portable meteorological station. At the same time, PM(2.5) samples were collected both at ground-level and at a high altitude sampling site, enabling particle chemical composition to be determined. Vertical profiles and PM(2.5) data were collected both within and above the mixing layer. Absorption coefficient (b(abs)) profiles were calculated from the Aethalometer data: in order to do so, an optical enhancement factor (C), accounting for multiple light-scattering within the filter of the new microAeth® Model AE51, was determined for the first time. The value of this parameter C (2.05±0.03 at λ=880nm) was calculated by comparing the Aethalometer attenuation coefficient and aerosol optical properties determined from OPC data along vertical profiles. Mie calculations were applied to the OPC number-size distribution data, and the aerosol refractive index was calculated using the effective medium approximation applied to aerosol chemical composition. The results compare well with AERONET data. The BC and b(abs) profiles showed a sharp decrease at the mixing height (MH), and fairly constant values of b(abs) and BC were found above the MH, representing 17±2% of those values measured within the mixing layer. The BC fraction of aerosol volume was found to be lower above the MH: 48±8% of the corresponding ground-level values. A statistical mean profile was calculated, both for BC and b(abs), to better describe their behaviour; the model enabled us to compute their average behaviour as a function of height, thus laying the foundations for valid parametrizations of vertical profile data which can be useful in both remote sensing and climatic studies., (Copyright © 2011 Elsevier B.V. All rights reserved.)