Simplified and Detailed Analysis of Data Center Particulate Contamination At Server and Room Level Using CFD

Continuous rise in cloud computing and other web-based services propelled the data center proliferation seen over the past decade. Traditional data centers use vapor-compression-based cooling units that not only reduce energy efficiency but also increase operational and initial investment costs due to involved redundancies. Free air cooling and airside economization can substantially reduce the IT Equipment (ITE) cooling power consumption, which accounts for approximately 40% of energy consumption for a typical air-cooled data center. However, this cooling approach entails an inherent risk of exposing the IT equipment to harmful ultrafine particulate contaminants, thus, potentially reducing the equipment and component reliability. The present investigation attempts to quantify the effects of particulate contamination inside the data center equipment and ITE room using CFD. An analysis of the boundary conditions to be used was done by detailed modeling of IT equipment and the data center white space. Both 2-D and 3-D simulations were done for detailed analysis of particle transport within the server enclosure. An analysis of the effect of the primary pressure loss obstructions like heat sinks and DIMMs inside the server was done to visualize the localized particle concentrations within the server. A room-level simulation was then conducted to identify the most vulnerable locations of particle concentration within the data center space. The results show that parameters such as higher velocities, heat sink cutouts, and higher aspect ratio features within the server tend to increase the particle concentration inside the servers.

Different contributions of efferent and reafferent feedback to sensorimotor temporal recalibration

Adaptation to delays between actions and sensory feedback is important for efficiently interacting with our environment. Adaptation may rely on predictions of action-feedback pairing (motor-sensory component), or predictions of tactile-proprioceptive sensation from the action and sensory feedback of the action (inter-sensory component). Reliability of temporal information might differ across sensory feedback modalities (e.g. auditory or visual), which in turn influences adaptation. Here, we investigated the role of motor-sensory and inter-sensory components on sensorimotor temporal recalibration for motor-auditory (button press-tone) and motor-visual (button press-Gabor patch) events. In the adaptation phase of the experiment, action-feedback pairs were presented with systematic temporal delays (0 ms or 150 ms). In the subsequent test phase, audio/visual feedback of the action were presented with variable delays. The participants were then asked whether they detected a delay. To disentangle motor-sensory from inter-sensory component, we varied movements (active button press or passive depression of button) at adaptation and test. Our results suggest that motor-auditory recalibration is mainly driven by the motor-sensory component, whereas motor-visual recalibration is mainly driven by the inter-sensory component. Recalibration transferred from vision to audition, but not from audition to vision. These results indicate that motor-sensory and inter-sensory components contribute to recalibration in a modality-dependent manner.

Evaluation of reliability index and probability of failure for the improvement of the Nigerian empirical mechanistic flexible pavement analysis and design system (Nempads)

The aim of this work was to evaluate reliability index (RI) with respect to fatigue and rutting within the different seasons peculiar to Nigeria, in order to improve Empirical-Mechanistic flexible pavement design approach, using First Order Reliability Method (FORM). Flexible pavement design involves many uncertainties, variabilities, and approximations regarding the input parameters like material properties, traffic loads. Others include subgrade strength, drainage conditions, construction, compaction procedures and climatic factors such as temperature, rainfall, and snowfall, etc. The combination of the variances associated with input parameters contributes to components and system uncertainty, and this combination of variances can have a significant effect on the predicted performance of the pavement. Reliability in pavement design is introduced to consider these uncertainties. Layers thicknesses, material properties, and Equivalent Standard Axle Load (ESAL) were entered into a multi-layer elastic theory software, ELSYM-5, which in turn were used to calculate strains and stresses for different seasons. The results obtained were entered into Nigerian fitted transfer function distress models to compute allowable ESALS. Miner’s hypothesis theory equation was used to calculate the cumulative damage due to stress and strains generated. A Framework was generated for finding individual reliability index (RI), systemic reliability index (SRI), and probability of failure. The findings showed that Season I (Winter) recorded the highest component reliability index for fatigue (5.63 for Normal Distribution). Season II (Summer) recorded the lowest component reliability index (β) for rutting (5.4 for Normal Distribution). Season III (Spring) recorded the lowest component reliability index for fatigue (1.85 for Normal Distribution)

Recovering Missing Component Dependence for System Reliability Prediction via Synergy between Physics and Data

Predicting system reliability is often a core task in systems design. System reliability depends on component reliability and dependence of components. Component reliability can be predicted with a physics-based approach if the associated physical models are available. If the models do not exist, component reliability may be estimated from data. When both types of components coexist, their dependence is often unknown, and the component states are therefore assumed independent by the traditional method, which can result in a large error. This work proposes a new system reliability method to recover the missing component dependence, thereby leading to a more accurate estimate of the joint probability density (PDF) of all the component states. The method works for series systems whose load is shared by its components that may fail due to excessive loading. For components without physical models available, the load data are recorded upon failure, and equivalent physical models are created; the model parameters are estimated by the proposed Bayesian approach. Then models of all component states become available, and the dependence of component states, as well as their joint PDF, can be estimated. Four examples are used to evaluate the proposed method, and the results indicate that the method can produce more accurate predictions of system reliability than the traditional method that assumes independent component states.

Unmanned Platform Design Methodology

Remotely operated and unmanned facilities offer significant safety, environmental and economic benefits over conventional facilities. This paper describes the key elements for successful design and an approach for evaluating the reliability, availability and TOTEX of unmanned facilities. The approach was developed during the concept and FEED phases of a wellhead platform project and forms the basis of the unmanned strategy going forwards but can also be used for facilities with partial processing topsides. During the design of a recent platform it became clear that normal FMEA/RAM analysis was not suitable for assessing the reliability and availability of unmanned facilities with low visit frequency. Drawing on previous experience, a new approach was developed to address the specific challenges of low maintenance intervals and provide a methodical approach to proving reliability. The new approach improved confidence in the predicted availability by identifying key components and appropriate reliability data. The process adds some extra steps to typical reliability and availability assessment, which are designed to address the specific demands of unmanned operations. The result of this work has given a clearer understanding of how reliability can be assessed and managed for low-manned or unmanned applications. The methodology helps to identify unmanned /low manned opportunities and provides guidance on design and reliability assessment It is observed that system reliability is usually driven by a few key components and that whilst many components have good overall reliability data this may not be applicable for the proposed specific operating environment and maintenance regime of an unmanned platform. It is therefore essential to evaluate components individually for their specific applications. It is concluded that to achieve the unmanned goal it is vital to fully understand the system and component reliability early in the project. The proposed methodology can be applied at any stage to validate the design, confirm assumptions, or identify gaps.

A Bayesian Approach to Recovering Missing Component Dependence for System Reliability Prediction via Synergy Between Physics and Data

Predicting system reliability is often a core task in systems design. System reliability depends on component reliability and dependence of components. Component reliability can be predicted with a physics-based approach if the associated physical models are available. If the models do not exist, component reliability may be estimated from data. When both types of components coexist, their dependence is often unknown, and the component states are therefore assumed independent by the traditional method, which can result in a large error. This work proposes a new system reliability method to recover the missing component dependence, thereby leading to a more accurate estimate of the joint probability density (PDF) of all the component states. The method works for series systems whose load is shared by its components that may fail due to excessive loading. For components without physical models available, the load data are recorded upon failure, and equivalent physical models are created; the model parameters are estimated by the proposed Bayesian approach. Then models of all component states become available, and the dependence of component states, as well as their joint PDF, can be estimated. Four examples are used to evaluate the proposed method, and the results indicate that the proposed method can produce more accurate predictions of system reliability than the traditional method that assumes independent component states.