Spelling suggestions: "subject:"data processing service centers"" "subject:"mata processing service centers""
21 |
Teleport videoconferencing and computer centerHarris, Rodney Thibodeaux January 1991 (has links)
The task of this thesis is the design of a facility that will accommodate both a large workforce in an office environment and a large accompaniment of machines. The entire complex, located in a remote area of a large rural university campus, must establish harmony between the various functional components of the center. / Master of Architecture
|
22 |
Energy efficient thermal management of data centers via open multi-scale designSamadiani, Emad 20 August 2009 (has links)
Data centers are computing infrastructure facilities that house arrays of electronic racks containing high power dissipation data processing and storage equipment whose temperature must be maintained within allowable limits. In this research, the sustainable and reliable operations of the electronic equipment in data centers are shown to be possible through the Open Engineering Systems paradigm. A design approach is developed to bring adaptability and robustness, two main features of open systems, in multi-scale convective systems such as data centers. The presented approach is centered on the integration of three constructs: a) Proper Orthogonal Decomposition (POD) based multi-scale modeling, b) compromise Decision Support Problem (cDSP), and c) robust design to overcome the challenges in thermal-fluid modeling, having multiple objectives, and inherent variability management, respectively. Two new POD based reduced order thermal modeling methods are presented to simulate multi-parameter dependent temperature field in multi-scale thermal/fluid systems such as data centers. The methods are verified to achieve an adaptable, robust, and energy efficient thermal design of an air-cooled data center cell with an annual increase in the power consumption for the next ten years. Also, a simpler reduced order modeling approach centered on POD technique with modal coefficient interpolation is validated against experimental measurements in an operational data center facility.
|
23 |
Cooling analysis of data centers CFD modeling and real-time calculators /Shrivastava, Saurabh K. January 2008 (has links)
Thesis (Ph. D.)--State University of New York at Binghamton, Thomas J. Watson School of Engineering and Applied Science, Department of Mechanical Engineering, 2008. / Includes bibliographical references.
|
24 |
Dynamic thermal response of the data center to cooling loss during facility power failureShields, Shawn 01 July 2009 (has links)
Provisioning for power failure is an important element of data center design. It is
important to assess both tangible and intangible costs of unplanned data center downtime.
These costs must be compared with the capital cost of providing various levels of backup
power infrastructure to compute and cooling equipment. Various levels of backup power
infrastructure each lead to a most probable transient scenario after utility power failure.
Because of differences between facilities, the level of risk that unacceptable compute
equipment inlet temperature associated with each level of backup power infrastructure is
not standardized; in particular, facilities with differing compute equipment power
densities may require different levels of backup power infrastructure to maintain safe
operation. Choosing one level of backup power infrastructure above another is not
necessarily obvious for every facility, as there may be large gaps in costs and unknown
levels of risk for lower levels of provisioning.
A first order model is also used to compare inclusion of various thermal
capacitance values with experimental results. Room level experiments also illustrate the
relative level of risk associated with various levels of provisioning for the same control
volume and compute equipment. Although provisioning to back up as much equipment as
possible remains the "safest" solution, cost will continue to play a factor in facility design
decisions. This work offers a step toward appropriate modeling of data center power
failure events and suggests further steps to continue the process.
|
25 |
Modeling, characterization, and optimization of web server power in data centers = Modelagem, caracterização e otimização de potência em centro de dados / Modelagem, caracterização e otimização de potência em centro de dadosPiga, Leonardo de Paula Rosa, 1985- 11 August 2013 (has links)
Orientadores: Sandro Rigo, Reinaldo Alvarenga Bergamaschi / Tese (doutorado) - Universidade Estadual de Campinas, Instituto de Computação / Made available in DSpace on 2018-08-24T00:17:07Z (GMT). No. of bitstreams: 1
Piga_LeonardodePaulaRosa_D.pdf: 5566406 bytes, checksum: 5fcce79bb9fc83646106c7580e0d77fc (MD5)
Previous issue date: 2013 / Resumo: Para acompanhar uma demanda crescente pelos recursos computacionais, empresas de TI precisaram construir instalações que comportam centenas de milhares de computadores chamadas centro de dados. Este ambiente é altamente dependente de energia elétrica, um recurso que é cada vez mais caro e escasso. Neste contexto, esta tese apresenta uma abordagem para otimizar potência e desempenho em centro de dados Web. Para isto, apresentamos uma infraestrutura para medir a potência dissipada por computadores de prateleiras, desenvolvemos modelos empíricos que estimam a potência de servidores Web e, por fim, implementamos uma de nossas heurísticas de otimização de potência global em um aglomerado de nós de processamento chamado AMD SeaMicro SM15k. A infraestrutura de medição de potência é composta por: uma placa personalizada, que é capaz de medir potência e é instalada em computadores de prateleira; um conversor de dados analógico/digital que amostra os valores de potência; e um software controlador. Mostramos uma nova metodologia para o desenvolvimento de modelos de potência para servidores Web que diminuem a quantidade de parâmetros dos modelos e reduzem as relações não lineares entre medidas de desempenho e potência do sistema. Avaliamos a nossa metodologia em dois servidores Web, um constituído por um processador AMD Opteron e outro por processador Intel i7. Nossos melhores modelos tem erro médio absoluto de 1,92% e noventa percentil para o erro absoluto de 2,66% para o sistema com processador Intel i7. O erro médio para o sistema composto pelo processador AMD Opteron é de 1,46% e o noventa percentil para o erro absoluto é igual a 2,08%. A implantação do sistema de otimização de potência global foi feita em um aglomerado de nós de processamento SeaMicro SM15k. A implementação se baseia no conceito de Virtual Power States, uma combinação de taxa de utilização de CPU com os estados de potência P e C disponíveis em processadores modernos, e no nosso algoritmo de otimização chamado Slack Recovery. Propomos e implementamos também um novo mecanismo capaz de controlar a utilização da CPU. Nossos resultados experimentais mostram que o nosso sistema de otimização pode reduzir o consumo de potência em até 16% quando comparado com o governador de potência do Linux chamado performance e em até 6,7% quando comparado com outro governador de potência do Linux chamado ondemand / Abstract: To keep up with an increasing demand for computational resources, IT companies need to build facilities that host hundreds of thousands of computers, the data centers. This environment is highly dependent on electrical energy, a resource that is becoming expensive and limited. In this context, this thesis develops a global data center-level power and performance optimization approach for Web Server data centers. It presents a power measurement framework for commodity servers, develops empirical models for estimating the power consumed by Web servers, and implements one of the global power optimization heuristics on a state-of-the-art, high-density SeaMicro SM15k cluster by AMD. The power measuring framework is composed of a custom made board, which is able to capture the power consumption; a data acquisition device that samples the measured values; and a piece of software that manages the framework. We show a novel method for developing full system Web server power models that prunes model parameters and reduces non-linear relationships among performance measurements and system power. The Web server power models use as parameters performance indicators read from the machine internal performance counters. We evaluate our approach on an AMD Opteron-based Web server and on an Intel i7-based Web server. Our best model displays an average absolute error of 1.92% for the Intel i7 server and 1.46% for AMD Opteron as compared to actual measurements, and 90th percentile for the absolute percent error equals to 2.66% for Intel i7 and 2.08% for AMD Opteron. We deploy the global power management system in a state-of-the-art SeaMicro SM15k cluster. The implementation relies on the concept of Virtual Power States, a combination of CPU utilization rate to the P/C power states available in modern processors, and on our global optimization algorithm called Slack Recovery. We also propose and implement a novel mechanism to control utilization rates in each server, a key aspect of our power/performance optimization system. Experimental results show that our Slack Recovery-based system can reduce up to 16% of the power consumption when compared to the Linux performance governor and 6.7% when compared to the Linux ondemand governor / Doutorado / Ciência da Computação / Doutor em Ciência da Computação
|
26 |
Designing a geodetic research data management system for the Hartebeeshoek radio astronomy observatoryCoetzer, Glenda Lorraine 11 1900 (has links)
The radio astronomy and space geodesy scientific instrumentation of the Hartebeesthoek Radio Astronomy Observatory (HartRAO) in Gauteng, South Africa, generate large volumes of data. Additional large data volumes will be generated by new geodesy instruments that are currently under construction and implementation, including a lunar laser ranging (LLR) system, seismic and meteorological systems and a Very Long Baseline Interferometry (VLBI) global observing system (VGOS) radio telescope. The existing HartRAO data management and storage system is outdated, incompatible and has limited storage capacity. This necessitates the design of a new geodetic research data management system (GRDMS). The focus of this dissertation is on providing a contextual framework for the design of the new system, including criteria, characteristics, components, an infrastructure architectural model and data structuring and organisation. An exploratory research methodology and qualitative research techniques were applied. Results attained from interviews conducted and literature consulted indicates a gap in the literature regarding the design of a data management system, specifically for geodetic data generated by HartRAO instrumentation. This necessitates the development of a conceptual framework for the design of a new GRDMS. Results are in alignment with the achievement of the research questions and objectives set for this study. / Information Science / M.A. (Information Science)
|
Page generated in 0.1304 seconds