A static VM placement and hybrid job scheduling model for green data centers.

Reducing energy consumption has become a critical issue in today data centers. Reducing the number of required physical and Virtual Machines results in energy-efficiency. In this paper, to avoid the disadvantages of VM migration, a static VM placement algorithm is proposed which places VMs on hosts in a Worst-Fit-Decreasing (WFD) fashion. To reduce energy consumption further, the effect of job scheduling policy on the number of VMs needed for maintaining QoS requirements is studied. Each VM is modeled by an M/M/* queue in space-shared, time-shared, and hybrid job scheduling policies, and energy consumption of real-time as well as non-real-time applications is analyzed. Numerical results show that the hybrid policy outperforms space-shared and time-shared policies, in terms of energy consumption as well as Service Level Agreement (SLA) violations. Moreover, our non-migration method outperforms three different algorithms which use VM migration, in terms of reducing both energy consumption and SLA Violations.

Cost-effective stochastic MAC circuits for deep neural networks.

Stochastic computing (SC) is a promising computing paradigm that can help address both the uncertainties of future process technology and the challenges of efficient hardware realization for deep neural networks (DNNs). However the impreciseness and long latency of SC have rendered previous SC-based DNN architectures less competitive against optimized fixed-point digital implementations, unless inference accuracy is significantly sacrificed. In this paper we propose a new SC-MAC (multiply-and-accumulate) algorithm, which is a key building block for SC-based DNNs, that is orders of magnitude more efficient and accurate than previous SC-MACs. We also show how our new SC-MAC can be extended to a vector version and used to accelerate both convolution and fully-connected layers of convolutional neural networks (CNNs) using the same hardware. Our experimental results using CNNs designed for MNIST and CIFAR-10 datasets demonstrate that not only is our SC-based CNNs more accurate and 40∼490× more energy-efficient for convolution layers than conventional SC-based ones, but ours can also achieve lower area-delay product and lower energy compared with precision-optimized fixed-point implementations without sacrificing accuracy. We also demonstrate the feasibility of our SC-based CNNs through FPGA prototypes.

Adaptive Resource Utilization Prediction System for Infrastructure as a Service Cloud.

Infrastructure as a Service (IaaS) cloud provides resources as a service from a pool of compute, network, and storage resources. Cloud providers can manage their resource usage by knowing future usage demand from the current and past usage patterns of resources. Resource usage prediction is of great importance for dynamic scaling of cloud resources to achieve efficiency in terms of cost and energy consumption while keeping quality of service. The purpose of this paper is to present a real-time resource usage prediction system. The system takes real-time utilization of resources and feeds utilization values into several buffers based on the type of resources and time span size. Buffers are read by R language based statistical system. These buffers' data are checked to determine whether their data follows Gaussian distribution or not. In case of following Gaussian distribution, Autoregressive Integrated Moving Average (ARIMA) is applied; otherwise Autoregressive Neural Network (AR-NN) is applied. In ARIMA process, a model is selected based on minimum Akaike Information Criterion (AIC) values. Similarly, in AR-NN process, a network with the lowest Network Information Criterion (NIC) value is selected. We have evaluated our system with real traces of CPU utilization of an IaaS cloud of one hundred and twenty servers.

A systematic review of the cost and cost-effectiveness of electronic discharge communications.

BACKGROUND: The transition between acute care and community care can be a vulnerable period in a patients' treatment due to the potential for postdischarge adverse events. The vulnerability of this period has been attributed to factors related to the miscommunication between hospital-based and community-based physicians. Electronic discharge communication has been proposed as one solution to bridge this communication gap. Prior to widespread implementation of these tools, the costs and benefits should be considered. OBJECTIVE: To establish the cost and cost-effectiveness of electronic discharge communications compared with traditional discharge systems for individuals who have completed care with one provider and are transitioning care to a new provider. METHODS: We conducted a systematic review of the published literature, using best practices, to identify economic evaluations/cost analyses of electronic discharge communication tools. Inclusion criteria were: (1) economic analysis and (2) electronic discharge communication tool as the intervention. Quality of each article was assessed, and data were summarised using a component-based analysis. RESULTS: One thousand unique abstracts were identified, and 57 full-text articles were assessed for eligibility. Four studies met final inclusion criteria. These studies varied in their primary objectives, methodology, costs reported and outcomes. All of the studies were of low to good quality. Three of the studies reported a cost-effectiveness measure ranging from an incremental daily cost of decreasing average discharge note completion by 1 day of $0.331 (2003 Canadian), a cost per page per discharge letter of €9.51 and a dynamic net present value of €31.1 million for a 5-year implementation of the intervention. None of the identified studies considered clinically meaningful patient or quality outcomes. DISCUSSION: Economic analyses of electronic discharge communications are scarcely reported, and with inconsistent methodology and outcomes. Further studies are needed to understand the cost-effectiveness and value for patient care.

Fast Outlier Detection Using a Grid-Based Algorithm.

As one of data mining techniques, outlier detection aims to discover outlying observations that deviate substantially from the reminder of the data. Recently, the Local Outlier Factor (LOF) algorithm has been successfully applied to outlier detection. However, due to the computational complexity of the LOF algorithm, its application to large data with high dimension has been limited. The aim of this paper is to propose grid-based algorithm that reduces the computation time required by the LOF algorithm to determine the k-nearest neighbors. The algorithm divides the data spaces in to a smaller number of regions, called as a "grid", and calculates the LOF value of each grid. To examine the effectiveness of the proposed method, several experiments incorporating different parameters were conducted. The proposed method demonstrated a significant computation time reduction with predictable and acceptable trade-off errors. Then, the proposed methodology was successfully applied to real database transaction logs of Korea Atomic Energy Research Institute. As a result, we show that for a very large dataset, the grid-LOF can be considered as an acceptable approximation for the original LOF. Moreover, it can also be effectively used for real-time outlier detection.

A Simple, Low-Cost Platform for Real-Time Isothermal Nucleic Acid Amplification.

Advances in microfluidics and the introduction of isothermal nucleic acid amplification assays have resulted in a range of solutions for nucleic acid amplification tests suited for point of care and field use. However, miniaturisation of instrumentation for such assays has not seen such rapid advances and fluorescence based assays still depend on complex, bulky and expensive optics such as fluorescence microscopes, photomultiplier tubes and sensitive lens assemblies. In this work we demonstrate a robust, low cost platform for isothermal nucleic acid amplification on a microfluidic device. Using easily obtainable materials and commercial off-the-shelf components, we show real time fluorescence detection using a low cost photodiode and operational amplifier without need for lenses. Temperature regulation on the device is achieved using a heater fabricated with standard printed circuit board fabrication methods. These facile construction methods allow fabrications at a cost compatible with widespread deployment to resource poor settings.

Computer glitch is a 'blow to morale'.

Delays and spiralling costs of an IT upgrade for Scotland's NHS 24 helpline is a blow to nursing staff, the RCN says.

Costs of health IT: beginning to understand the financial impact of a dental school EHR.

Health Information Technology (Health IT) constitutes an integral component of the operations of most academic dental institutions nowadays. However, the expenses associated with the acquisition and the ongoing maintenance of these complex systems have often been buried among costs for other electronic infrastructure systems, distributed across various cost centers including unmeasured central campus support, covered centrally and therefore difficult to quantify, and spread over years, denying school administrators a clear understanding of the resources that have been dedicated to Health IT. The aim of this study was to understand the financial impact of Health IT at four similar U.S. dental schools: two schools using a purchased Electronic Health Record (EHR), and two schools that developed their own EHR. For these schools, the costs of creating ($2.5 million) and sustaining ($174,000) custom EHR software were significantly higher than acquiring ($500,000) and sustaining ($121,000) purchased software. These results are based on historical data and should not be regarded as a gold standard for what a complete Health IT suite should cost. The presented data are intended to inform school administrators about the myriad of costs associated with Health IT and give them a point of reference when comparing costs or making estimates for implementation projects.

Cost-effective GPU-grid for genome-wide epistasis calculations.

BACKGROUND: Until recently, genotype studies were limited to the investigation of single SNP effects due to the computational burden incurred when studying pairwise interactions of SNPs. However, some genetic effects as simple as coloring (in plants and animals) cannot be ascribed to a single locus but only understood when epistasis is taken into account [1]. It is expected that such effects are also found in complex diseases where many genes contribute to the clinical outcome of affected individuals. Only recently have such problems become feasible computationally. OBJECTIVES: The inherently parallel structure of the problem makes it a perfect candidate for massive parallelization on either grid or cloud architectures. Since we are also dealing with confidential patient data, we were not able to consider a cloud-based solution but had to find a way to process the data in-house and aimed to build a local GPU-based grid structure. METHODS: Sequential epistatsis calculations were ported to GPU using CUDA at various levels. Parallelization on the CPU was compared to corresponding GPU counterparts with regards to performance and cost. RESULTS: A cost-effective solution was created by combining custom-built nodes equipped with relatively inexpensive consumer-level graphics cards with highly parallel GPUs in a local grid. The GPU method outperforms current cluster-based systems on a price/performance criterion, as a single GPU shows speed performance comparable up to 200 CPU cores. CONCLUSION: The outlined approach will work for problems that easily lend themselves to massive parallelization. Code for various tasks has been made available and ongoing development of tools will further ease the transition from sequential to parallel algorithms.

Unsupervised markerless 3-DOF motion tracking in real time using a single low-budget camera.

Motion tracking is a critical task in many computer vision applications. Existing motion tracking techniques require either a great amount of knowledge on the target object or specific hardware. These requirements discourage the wide spread of commercial applications based on motion tracking. In this paper, we present a novel three degrees of freedom motion tracking system that needs no knowledge on the target object and that only requires a single low-budget camera that can be found installed in most computers and smartphones. Our system estimates, in real time, the three-dimensional position of a nonmodeled unmarked object that may be nonrigid, nonconvex, partially occluded, self-occluded, or motion blurred, given that it is opaque, evenly colored, enough contrasting with the background in each frame, and that it does not rotate. Our system is also able to determine the most relevant object to track in the screen. Our proposal does not impose additional constraints, therefore it allows a market-wide implementation of applications that require the estimation of the three position degrees of freedom of an object.

Remote real-time monitoring of subsurface landfill gas migration.

The cost of monitoring greenhouse gas emissions from landfill sites is of major concern for regulatory authorities. The current monitoring procedure is recognised as labour intensive, requiring agency inspectors to physically travel to perimeter borehole wells in rough terrain and manually measure gas concentration levels with expensive hand-held instrumentation. In this article we present a cost-effective and efficient system for remotely monitoring landfill subsurface migration of methane and carbon dioxide concentration levels. Based purely on an autonomous sensing architecture, the proposed sensing platform was capable of performing complex analytical measurements in situ and successfully communicating the data remotely to a cloud database. A web tool was developed to present the sensed data to relevant stakeholders. We report our experiences in deploying such an approach in the field over a period of approximately 16 months.

Radio Frequency Identification (RFID) and communication technologies for solid waste bin and truck monitoring system.

This paper deals with a system of integration of Radio Frequency Identification (RFID) and communication technologies for solid waste bin and truck monitoring system. RFID, GPS, GPRS and GIS along with camera technologies have been integrated and developed the bin and truck intelligent monitoring system. A new kind of integrated theoretical framework, hardware architecture and interface algorithm has been introduced between the technologies for the successful implementation of the proposed system. In this system, bin and truck database have been developed such a way that the information of bin and truck ID, date and time of waste collection, bin status, amount of waste and bin and truck GPS coordinates etc. are complied and stored for monitoring and management activities. The results showed that the real-time image processing, histogram analysis, waste estimation and other bin information have been displayed in the GUI of the monitoring system. The real-time test and experimental results showed that the performance of the developed system was stable and satisfied the monitoring system with high practicability and validity.

[Cost-effectiveness in the detection of influenza H1N1: clinical data versus rapid tests]. / Costo-efectividad en la detección de influenza H1N1: datos clínicos versus pruebas rápidas.

OBJECTIVE: Evaluate the performance of clinical data and the rapid influenza diagnostic test (RIDT) in diagnosing influenza H1N1, and analyze the cost-benefit of using this diagnostic tool. METHODS: The RIDT was used for patients who came to four hospitals in Mexico City with an influenza-like illness (ILI) in October and November 2009. The diagnostic performance of the ILI clinical data and the RIDT was compared to that of the real-time reverse transcription polymerase chain reaction (rRT-PCR) test. The rRT-PCR test was conducted in a reference laboratory and blinded to the results of the RIDT. An economic evaluation also was conducted to estimate the budgetary impact of using the RIDT. RESULTS: The study included 78 patients, 39 of whom tested positive for influenza H1N1 and 6 tested positive for seasonal influenza A, according to the results of the rRT-PCR. The ILI clinical data yielded a sensitivity of 96% and specificity of 21%; the RIDT yielded a sensitivity of 76% and specificity of 82%; and the ILI clinical data and RIDT together yielded a sensitivity of 96% and specificity of 100%. The positive likelihood quotient for ILI-headaches was 31.5 and that of ILI-odynophagia, 330. The use of RIDT yielded savings of US$12.6 per each suspected case. CONCLUSIONS: Use of the RIDT to aid in the diagnosis of influenza H1N1 increases certainty and lowers the average cost per suspected and infected patient.

Costo-efectividad en la detección de influenza H1N1: datos clínicos versus pruebas rápidas / Cost-effectiveness in the detection of influenza H1N1: clinical data versus rapid tests

OBJETIVO: Evaluar el desempeño de los datos clínicos y la prueba rápida (PR) en el diagnóstico de influenza H1N1, y analizar el costo-beneficio que representa el uso de esta herramienta diagnóstica. MÉTODOS: Se aplicó la PR a pacientes que acudieron a cuatro hospitales en la ciudad de México con sintomatología similar a influenza (SSI) durante el período octubre y noviembre de 2009. Se comparó el desempeño diagnóstico de la SSI más la PR contra el de la reacción en cadena de la polimerasa en transcripción reversa en tiempo real (rRT-PCR). La rRT-PCR fue procesada en un laboratorio de referencia y cegado al resultado de la PR. Además, se llevó a cabo una evaluación económica a partir de la cual se estimó el impacto presupuestal relacionado con la utilización de la PR RESULTADOS: Se incluyó a 78 pacientes, de los cuales 39 fueron positivos para influenza H1N1 y 6 para influenza A estacional, de acuerdo al resultado de la rRT-PCR. La SSI mostró una sensibilidad de 96 por ciento y una especificidad de 21 por ciento, la PR de 76 por ciento y 82 por ciento y el conjunto de SSI más PR de 96 por ciento y 100 por ciento, respectivamente. El Cociente de Verosimilitud positivo de la SSI-cefalea fue de 31,5 y el de SSI-odinofagia fue de 330. El uso de PR mostró un ahorro de US$ 12,6 por cada caso sospechoso. CONCLUSIONES: El uso de la PR como auxiliar en el diagnóstico de influenza H1N1 incrementa la certeza y reduce el costo promedio por paciente sospechoso e infectado.

OBJECTIVE: Evaluate the performance of clinical data and the rapid influenza diagnostic test (RIDT) in diagnosing influenza H1N1, and analyze the cost-benefit of using this diagnostic tool. METHODS: The RIDT was used for patients who came to four hospitals in Mexico City with an influenza-like illness (ILI) in October and November 2009. The diagnostic performance of the ILI clinical data and the RIDT was compared to that of the real-time reverse transcription polymerase chain reaction (rRT-PCR) test. The rRT-PCR test was conducted in a reference laboratory and blinded to the results of the RIDT. An economic evaluation also was conducted to estimate the budgetary impact of using the RIDT. RESULTS: The study included 78 patients, 39 of whom tested positive for influenza H1N1 and 6 tested positive for seasonal influenza A, according to the results of the rRT-PCR. The ILI clinical data yielded a sensitivity of 96 percent and specificity of 21 percent; the RIDT yielded a sensitivity of 76 percent and specificity of 82 percent; and the ILI clinical data and RIDT together yielded a sensitivity of 96 percent and specificity of 100 percent. The positive likelihood quotient for ILI-headaches was 31.5 and that of ILI-odynophagia, 330. The use of RIDT yielded savings of US$12.6 per each suspected case. CONCLUSIONS: Use of the RIDT to aid in the diagnosis of influenza H1N1 increases certainty and lowers the average cost per suspected and infected patient.

[Facing the challenges of ubiquitous computing in the health care sector]. / Herausforderungen durch ubiquitäres Computing im Gesundheitsbereich.

The steady progress of microelectronics, communications and information technology will enable the realisation of the vision for "ubiquitous computing" where the Internet extends into the real world embracing everyday objects. The necessary technical basis is already in place. Due to their diminishing size, constantly falling price and declining energy consumption, processors, communications modules and sensors are being increasingly integrated into everyday objects today. This development is opening up huge opportunities for both the economy and individuals. In the present paper we discuss possible applications, but also technical, social and economic barriers to a wide-spread use of ubiquitous computing in the health care sector.