Energy Efficiency in Data Centers: How to Reduce Power Consumption in Data Centers by Optimum UPS Loading
N.P. Zinjad1, S. S. More2
1N.P. Zinjad, Research Scholar, Department of Electrical Engineering, BVDU COE, Pune, India.
2Mrs. S. S. More, Professor & Head Department of Electrical Engineering, BVDU COE, Pune, India.
Manuscript received on May 17, 2012. | Revised Manuscript received on June 11, 2012. | Manuscript published on June 30, 2012. | PP: 453-456 | Volume-1 Issue-5, June 2012. | Retrieval Number: E0460061512/2012©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Computation and data center has a huge value to modern enterprise. This has resulted in the installation of millions of data centers in business around the globe. Historically, the cost to power and cool these facilities was small relative to the investment in servers, storage units and other equipments. Today, however, the annual power and cooling costs of typical data centers are almost equal to the cost of hardware. In the past decade, India has witnessed an exponential increase in the demand for digital storage, from 1 petabyte in 2001 to more than 34 petabytes by 2007. They also continue to grow at a compounded rate of 25-30%. Datacenter growth is basically driven by increasing requirements from the sectors such as financial institutions, telecom operators, manufacturing and services. While large financial institutions and telecom companies are likely to build captive Datacenters for hosting their growing data storage needs. Datacenter service providers are expected to invest significantly to multiply their capacities, so as to fulfill the demand arising from small and midsize users. Datacenter is highly energy intensive. With the increasing energy cost, the increase in operational cost is inevitable. Therefore it becomes necessary to reduce the energy consumption to offset the increasing operational cost and to maintain competitiveness. Existing Datacenters need to adopt the best practices in design, operation and maintenance to achieve operational excellence. The increasing IT business process outsource from foreign countries has resulted in phenomenal growth of Datacenters in India. The total datacenter capacity in India is growing at a rapid pace and is expected to exceed 5.1 million square feet by 2012. The primary scope of this paper is to provide a framework in which data centers, large and small, could analyze and reduce their power consumption. This paper provides a quantitative approach to understanding energy efficiency within a server and within a data center. A panorama for power minimization and energy efficiency beginning with the basics of dual in line memory modules (DIMM) selection, configuring processors with reduced power states, options for constantly spinning disks, power management features in operating systems and other internal equipments.
Keywords: Loading optimization, Harmonics, Flywheel plus converter, Loading, Efficiency, Five “nines”, MTTF, MTBR, MTBF.