Data Center Management Research Articles

A critical analysis of an IoT—aware AAL system for elderly monitoring

A growing number of elderly people (65+ years old) are affected by particular conditions, such as Mild Cognitive Impairment (MCI) and frailty, which are characterized by a gradual cognitive and physical decline. Early symptoms may spread across years and often they are noticed only at late stages, when the outcomes remain irrevocable and require costly intervention plans. Therefore, the clinical utility of early detecting these conditions is of substantial importance in order to avoid hospitalization and lessen the socio-economic costs of caring, while it may also significantly improve elderly people’s quality of life. This work deals with a critical performance analysis of an Internet of Things aware Ambient Assisted Living (AAL) system for elderly monitoring. The analysis is focused on three main system components: (i) the City-wide data capturing layer, (ii) the Cloud-based centralized data management repository, and (iii) the risk analysis and prediction module. Each module can provide different operating modes, therefore the critical analysis aims at defining which are the best solutions according to context’s needs. The proposed system architecture is used by the H2020 City4Age project to support geriatricians for the early detection of MCI and frailty conditions.

ECRA: An Encounter-aware and Clustering-based Routing Algorithm for Information-centric VANETs

The vehicular ad hoc networks (VANETs) have recently received much attention. However, the efficiency of network transmission is low because of the dynamic characteristics. To improve transmission performance, many studies introduce Content Centric Networks (CCN) into VANETs forming information-centric VANETs. Due to dynamic of network topology, the efficient design of data transmission is a great challenge. In this paper, we propose the encounter-aware and clustering-based routing algorithm (ECRA) for information-centric VANETs. The awareness of encounter allows vehicles to record the movement track of other vehicles to find the destination node quickly. The clustering can better manage data exchange among nodes in urban traffic networks. The cluster heads exchange historical encounter information (e.g., encounter node, encounter time and encounter location) with each other to maintain a g-ECRA table, which is used to determine the position of the next anchor. The experimental results verify the effectiveness of ECRA in information-centric VANETs.

Comparison of data driven modeling approaches for temperature prediction in data centers

Energy-efficient thermal management of data centers based on dynamic optimization and provisioning of cooling resources requires rapid (nearly real-time) predictions of temperatures within data centers. This work for the first time compares multiple Data-Driven Models (DDMs) to achieve such rapid temperature predictions. DDM typically employs statistical or machine learning-based tools, in combination with physics-based modeling and/or experimental data to predict system behavior. In general, DDM approaches are well-suited to systems that have multiple operational states based on interactions between the many electrical, mechanical and control parameters typical of data centers.This study compares the performance of three different DDM methods, namely Artificial Neural Networks (ANN), Support Vector Regression (SVR), Gaussian Process Regression (GPR) in predicting both steady-state and transient rack inlet air temperature distributions in data centers. Additionally, Proper Orthogonal Decomposition (POD) was considered for transient modeling. The data used for training and analysis were obtained by performing 300 offline numerical simulations with a room-level, experimentally validated computational fluid dynamics/heat transfer (CFD/HT) model.The performance of the four data-driven models was evaluated based on the absolute mean error for interpolation and extrapolation, and the adaptability of the models to changes in physical domain (data center room) configuration. Additionally, the impact of the size of the training data set on prediction accuracy is also compared for the four models.For the steady-state case study, the predictions for ANN, SVR and GPR models are in good agreement with CFD/HT simulations, with the GPR model having the smallest overall average prediction error of 0.6 °C in rack inlet air temperature, corresponding to a relative error of 2.7% with respect to rack inlet temperature measured in °C. It was found that for all the frameworks the prediction error increases when the size of training data set was less than 300 samples. The GPR model had the best accuracy for smaller training data sets compared with the other models, with an average prediction error for rack inlet temperatures <1 °C when trained on only 50 simulations. For the transient case study, the interpolative prediction error for all the models is very low (<0.3 °C); however, the extrapolative prediction errors are much greater, and appear to be directly proportional to the (here, temporal) “distance” from the interrogation point to the input parameter space.

Smart-Grid-Aware Load Regulation of Multiple Datacenters towards the Variable Generation of Renewable Energy

Recently, as renewable and distributed power sources boost, many such resources are integrated into the smart grid as a clean energy input. However, since the generation of renewable energy is intermittent and unstable, the smart grid needs to regulate the load to maintain stability after integrating the renewable energy source. At the same time, with the development of cloud computing, large-scale datacenters are becoming potentially controllable loads for the smart grid due to their high energy consumption. In this paper, we propose an appropriate approach to dynamically adjust the datacenter load to balance the unstable renewable energy input into the grid. This could meet the demand response requirements by taking advantage of the variable power consumption of datacenters. We have examined the scenarios of one or more datacenters being integrated into the grid and adopted a stochastic algorithm to solve the problem we established. The experimental results illustrated that the dynamic load management of multiple datacenters could help the smart grid to reduce losses and thus save operational costs. Besides, we also analyzed the impact of the flexibility and the delay of datacenter actions, which could be applied to more general scenarios in realistic environments. Furthermore, considering the impact of the action delay, we employed a forecasting method to predict renewable energy generation in advance to eliminate the extra losses brought by the delay as much as possible. By predicting solar power generation, the improved results showed that the proposed method was effective and feasible under both sunny and cloudy/rainy/snowy weather conditions.

CropSight: a scalable and open-source information management system for distributed plant phenotyping and IoT-based crop management.

BackgroundHigh-quality plant phenotyping and climate data lay the foundation for phenotypic analysis and genotype-environment interaction, providing important evidence not only for plant scientists to understand the dynamics between crop performance, genotypes, and environmental factors but also for agronomists and farmers to closely monitor crops in fluctuating agricultural conditions. With the rise of Internet of Things technologies (IoT) in recent years, many IoT-based remote sensing devices have been applied to plant phenotyping and crop monitoring, which are generating terabytes of biological datasets every day. However, it is still technically challenging to calibrate, annotate, and aggregate the big data effectively, especially when they were produced in multiple locations and at different scales.FindingsCropSight is a PHP Hypertext Pre-processor and structured query language-based server platform that provides automated data collation, storage, and information management through distributed IoT sensors and phenotyping workstations. It provides a two-component solution to monitor biological experiments through networked sensing devices, with interfaces specifically designed for distributed plant phenotyping and centralized data management. Data transfer and annotation are accomplished automatically through an hypertext transfer protocol-accessible RESTful API installed on both device side and server side of the CropSight system, which synchronize daily representative crop growth images for visual-based crop assessment and hourly microclimate readings for GxE studies. CropSight also supports the comparison of historical and ongoing crop performance while different experiments are being conducted.ConclusionsAs a scalable and open-source information management system, CropSight can be used to maintain and collate important crop performance and microclimate datasets captured by IoT sensors and distributed phenotyping installations. It provides near real-time environmental and crop growth monitoring in addition to historical and current experiment comparison through an integrated cloud-ready server system. Accessible both locally in the field through smart devices and remotely in an office using a personal computer, CropSight has been applied to field experiments of bread wheat prebreeding since 2016 and speed breeding since 2017. We believe that the CropSight system could have a significant impact on scalable plant phenotyping and IoT-style crop management to enable smart agricultural practices in the near future.

The CARICOMP Network of Caribbean Marine Laboratories (1985–2007): History, Key Findings, and Lessons Learned

Caribbean Coastal Marine Productivity (CARICOMP) was a basin-wide cooperative, international network of marine laboratories established in 1985. Recognizing major trends of change in coastal ecosystems and the importance of the linkages among them, our goal was to monitor synoptically with standardized methods the physical environment and to document trends in measures of the structure and functioning of coral reefs, seagrasses and mangroves. Between 1985 and 1993, the CARICOMP Steering Committee established a data management center and wrote a methods manual. Marine laboratories joined the program by appointing a Site Director and signing an agreement specifying the cost sharing and responsibilities of the laboratory. With significant outside funding in 1993, the program became fully functional and ultimately more than 30 institutions in 21 Caribbean countries participated. Monitoring lasted from 1992 to 2007, spanning many technological advances including the internet, automated in situ data logging and remote sensing. Annual CARICOMP meetings, organized at a different laboratory each year, were essential in standardization of methods and maintaining interest. Open access to the data was a goal from the start, although the members imposed an embargo to allow time to publish major results. At some of the sites, monitoring continues to this day, generating among the longest coastal monitoring data sets in the Caribbean, and possibly in the world. Over time, multi-authored papers were prepared for the Proceedings of the International Coral Reef Symposia and other journals, and independent scientists drew on the open database for regional analyses of ecosystem trends. Recently, active members have written summary papers based on the monitoring data covering physical parameters, coral reefs, seagrasses and mangroves. Overall, the data reveal major differences across the region and changing rates and trends showing the dynamism and vulnerability of coastal ecosystems. The longer the monitoring continues, the more valuable the dataset becomes as a tool to discern the underlying factors driving the structure and functioning of Caribbean coastal ecosystems. Several recent workshops have concluded that the need for regionally cooperative monitoring and research has never been greater.

Topology-aware multi-objective virtual machine dynamic consolidation for cloud datacenter

With help of virtualization technology, cloud datacenters provide affordable computation power to numerous enterprises and business users. Virtual machine dynamic consolidation (VMDC) plays an important role in the energy consumption management of datacenters. Previous solutions on VMDC either fail to consider impact resulting from the interaction between datacenter network topology and communication topology, or handle this issue in coarse granular level. In this paper, we propose a new approach to coordinate datacenter network topology and communication topology. The proposed VMDC approach adopts a multi-objective genetic algorithm to simultaneously minimize energy consumption and communication traffic causing performance bottlenecks, while meeting SLA constraint on migration time. In this approach, both SLA constraint and objective function on communication traffic are formulated with topology awareness in switch link level. A comprehensive simulation study shows that the proposed approach can significantly save energy consumption of datacenters and provide SLA assurance for different types of datacenter topologies. Simulation results demonstrate that the proposed approach outperforms prior approaches in terms of reducing energy consumption, bottleneck ratio and SLA violation ratio.

Receiver function imaging of mantle transition zone discontinuities and the origin of volcanism beneath Libya

The origin of magma at Libya’s volcanic provinces is still poorly understood. In order to constrain these sources we report results of the first P-to-S receiver function investigation of the 410 km and 660 km depth discontinuities bounding the mantle transition zone (MTZ) beneath Libya. The dataset used by this study is a combination of eighteen seismic stations belonging to the Libyan Center for Remote Sensing and Space Science and the Incorporated Research Institutions for Seismology Data Management Center. The average thickness of the MTZ is found to be 249 ± 14 km, which is similar to the global average. A 10 km thinning of the MTZ was observed beneath the Miocene – Holocene volcanic provinces in central Libya, suggesting higher-than-normal temperatures in those parts of the MTZ. Our preferred model suggests that the origin of volcanism in Libya is due to higher temperatures in the MTZ beneath this region. However, in Eastern Libya, the thickness of the MTZ increases from 249 km to 270 km which indicates a colder-than-normal MTZ which may relate to ancient subducted slabs due to the Africa-Europe convergence since the late Mesozoic.

PPHA-Popularity Prediction Based High Data Availability for Multimedia Data Center

Due to the growth of internet-connected devices and extensive data analysis applications in recent years, cloud computing systems are largely utilized. Because of high utilization of cloud storage systems, the demand for data center management has been increased. There are several crucial requirements of data center management, such as increase data availability, enhance durability, and decrease latency. In previous works, a replication technique is mostly used to answer those needs according to consistency requirements. However, most of the works consider full data, popular data, and geo-distance-based replications by considering storage and replication cost. Moreover, the previous data popularity based-techniques rely on the historical and current data access frequencies for replication. In this article, the authors approach this problem from a distinct aspect while developing replication techniques for a multimedia data center management system which can dynamically adapt servers of a data center by considering popularity prediction in each data access location. Therefore, they first label data objects from one to ten to track access frequencies of data objects. Then, they use those data access frequencies from each location to predict the future access frequencies of data objects to determine the replication levels and locations to replicate the data objects, and store the related data objects to close storage servers. To show the efficiency of the proposed methods, the authors conduct an extensive simulation by using real data. The results show that the proposed method has an advantage over the previous works in terms of data availability and increases the data availability up to 50%. The proposed method and related analysis can assist multimedia service providers to enhance their service qualities.

Allocation of energy-efficient task in cloud using DVFS

Nowadays, the expanding computational capabilities of the cloud system rely on the minimisation of the consumed power to make them sustainable and economically productive. Power management of cloud data centres received a great attention from industry and academia as it consumes high energy and thus increases the operational cost. One of the core approaches for the conservation of energy in the cloud data centre is the task scheduling. This task allocation in a heterogeneous environment is a well known NP-hard problem due to which researchers pay attention for proposing various heuristic techniques for the problem. In this paper, a technique is proposed based on dynamic voltage frequency scaling (DVFS) for optimising the energy consumption in the cloud environment. The basic idea is to address the trade-off between energy consumption and makespan of the system. Here, we formally introduce a model that includes various subsystems and assess the implementation of the algorithm in the heterogeneous environment.

Context Aware VM Placement Optimization Technique for Heterogeneous IaaS Cloud

Ever increasing demand for cloud adoption is prompting researchers and engineers around the world to make cloud computing more efficient and beneficial for cloud service providers and users. Cloud computing brings profits for all when the cloud infrastructure is used efficiently, and its services are made affordable to businesses of all scales. Managing cloud data center incurs a significant cost, which includes investing in IT infrastructure at the beginning and data center management costs for power, repair, space, and so on at later stages. The power costs are contributing to a significant share in overall data center management costs, and saving in power consumption can help reduce management costs for data center owners. This paper proposes an efficient context-aware adaptive heuristic-based solution for the virtual machine (VM) placement optimization in the heterogeneous cloud data centers. The proposed VM placement technique takes into the account of physical machine characteristics and load (peak and non-peak) conditions in the heterogeneous data centers to save power and also improve performance efficiency for data center owners. The experiments conducted with real cloud workloads and also synthetic workloads against a well-known adaptive heuristic-based technique indicate significant performance improvements and energy saving with our proposed solution.

U-LITE, a Private Cloud Approach for Particle Physics Computing

The computing infrastructure of Laboratori Nazionali del Gran Sasso (LNGS) is the primary platform for data storage, analysis, computing and simulation of the LNGS-based experiments, which are part of the research activities of the Istituto Nazionale di Fisica Nucleare (INFN). Scientific collaborations running such experiments have diverse needs and adopt different approaches in developing the computing frameworks that support their activities. The emergence of resource virtualization and the cloud paradigm has allowed a complete change to the way of operating and managing the LNGS computing infrastructure, and realize U-LITE, a private versatile environment apt at hosting such varied ecosystem and providing LNGS scientific users a familiar computing interface which hides all the complexities of a modern data center management. Since 2011, U-LITE has proven to be a valuable tool for the LNGS experiments, and provides an example of effective use of a private Cloud computing approach in a real scientific context.

Deep Learning-Based Sustainable Data Center Energy Cost Minimization With Temporal MACRO/MICRO Scale Management

Recently, distributed sustainable data centers based on renewable power generators have been deployed in order to efficiently reduce both the energy cost and carbon emission. Dynamic right-sizing (DRS) and frequency scaling (FS) have been considered as promising solutions to tune the computing capacity corresponding to the dynamic renewable power capacity. However, in existing works, the inaccurate power prediction and the uncoordinated decision making of DRS/FS still lead to low service quality and high energy cost. In this paper, we propose a novel joint optimization method for energy efficient distributed sustainable data centers. The proposed method adopts long short-term memory approach to improve the prediction accuracy of renewable power capacity for a long period, and unsupervised deep learning (DL) solver to resolve the coordinated DRS/FS optimization. Furthermore, we present the MACRO/MICRO (MAMI) time scale-based data center management technique to achieve both high energy efficiency and low wake-up transition overhead of DRS. To evaluate the proposed DL-based MAMI optimizer, we use the real trace data of renewable power capacity from the U.S. Measurement and Instrumentation Data Center. The experimental results demonstrate that our method reduces the energy cost by 25% compared with conventional metaheuristics while guaranteeing the service response delay requirement.

Developing a monitoring system for Cloud-based distributed data-centers

Nowadays more and more datacenters cooperate each others to achieve a common and more complex goal. New advanced functionalities are required to support experts during recovery and managing activities, like anomaly detection and fault pattern recognition. The proposed solution provides an active support to problem solving for datacenter management teams by providing automatically the root-cause of detected anomalies. The project has been developed in Bari using the datacenter ReCaS as testbed. Big Data solutions have been selected to properly handle the complexity and size of the data. Features like open source, big community, horizontal scalability and high availability have been considered and tools belonging to the Hadoop ecosystem have been selected. The collected information is sent to a combination of Apache Flume and Apache Kafka, used as transport layer, in turn delivering data to databases and processing components. Apache Spark has been selected as analysis component. Different kind of databases have been considered in order to satisfy multiple requirements: Hadoop Distributed File System, Neo4j, InfluxDB and Elasticsearch. Grafana and Kibana are used to show data in a dedicated dashboards. The Root-cause analysis engine has been implemented using custom machine learning algorithms. Finally, results are forwarded to experts by email or Slack, using Riemann.

Regional Seismic Activity after 2012 M8.6 Sumatra Earthquake

The Sumatra subduction zone, one of the most active plate tectonic margins in the world, is characterised by the Indo-Australia Plate subducting beneath the Sunda plate and Andaman micro plate, causing seismic activity along the plate boundary. There had been five major earthquakes of magnitude greater than 8.0 in this region from 2004 to 2014. Three of them are dip-slip and rest of the two is strike-slip type events. Regional earthquake activity after the occurrence of those five events was analyzed. Hypocentral data obtained from the Data Management Center at the Incorporated Research Institutions for Seismology for the period from January 2000 to December 2014 of magnitude 3.0 were used for the analysis. A statistical analysis was carried out to know whether earthquake activity has increased after the major five events and the analysis was carried out both qualitatively and quantitatively. The results of the analysis show that the number of earthquakes in the region has increased considerably after the occurrence of April, 2012 magnitude 8.6 and 8.2 strike-slip events. Further results show that there is no change in the regional earthquake activity after the occurrence of other three major dip-slip type events. Present study results reasonably agree with the results obtained by the other studies carried out with different methods. In the context of Sri Lanka, strike-slip type focal mechanism of the 2012 two major events may be the reason for increasing of activity in the region, especially in Eastern part of Sri Lanka near Maduruoya, Highland-Wijayan boundary and Wadinagala area of Ampara District. Keywords: seismology, Sumatra subduction zone, regional seismicity, earthquake, Sri Lanka

Centralized Clinical Trial Imaging Data Management: Practical Guidance from a Comprehensive Cancer Center's Experience.

Medical imaging is an integral part of clinical trial research and it must be managed properly to provide accurate data to the sponsor in a timely manner (Clune in Cancer Inform 4:33-56, 2007; Wang et al. in Proc SPIE Int Soc Opt Eng 7967, 2011). Standardized workflows for site qualification, protocol preparation, data storage, retrieval, de-identification, submission, and query resolution are paramount to achieve quality clinical trial data management such as reducing the number of imaging protocol deviations and avoiding delays in data transfer. Centralization of data management and implementation of relational databases and electronic workflows can help maintain consistency and accuracy of imaging data. This technical note aims at sharing the practical implementation of our centralized clinical trial imaging data management processes to avoid the fragmentation of tasks among various disease centers and research staff, and enable us to provide quality, accurate, and timely imaging data to clinical trial sponsors.

CapNet

As the scale and density of data centers continue to grow, cost-effective data center management (DCM) is becoming a significant challenge for enterprises hosting large-scale online and cloud services. Machines need to be monitored, and the scale of operations mandates an automated management with high reliability and real-time performance. The limitations of today’s typical DCM network are many-fold. Primarily, it is a fixed wired network, and hence scaling it for a large number of servers increases its cost. In addition, with server densities increasing over recent years, this network also has to be cabled correctly and the management of this network parallels the complexity of managing a data network, since it needs to be networked with multiple switches and routers. In this article, we propose a wireless sensor network as a cost-effective networking solution for DCM while satisfying the reliability and latency performance requirements of DCM. We have developed CapNet, a real-time wireless sensor network for power capping, a time-critical DCM function for power management in a cluster of servers. CapNet employs an efficient event-driven protocol that triggers data collection only on the detection of a potential power capping event. We deploy and evaluate CapNet in a data center. Using server power traces, our experimental results on a cluster of 480 servers inside the data center show that CapNet can meet the real-time requirements of power capping. CapNet demonstrates the feasibility and efficacy of wireless sensor networks for time-critical DCM applications.

A prediction‐based and power‐aware virtual machine allocation algorithm in three‐tier cloud data centers

SummaryWith the increasing popularity of cloud computing services, the more number of cloud data centers are constructed over the globe. This makes the power consumption of cloud data center elements as a big challenge. Hereby, several software and hardware approaches have been proposed to handle this issue. However, this problem has not been optimally solved yet. In this paper, we propose an online cloud resource management with live migration of virtual machines (VMs) to reduce power consumption. To do so, a prediction‐based and power‐aware virtual machine allocation algorithm is proposed. Also, we present a three‐tier framework for energy‐efficient resource management in cloud data centers. Experimental results indicate that the proposed solution reduces the power consumption; at the same time, service‐level agreement violation (SLAV) is also improved.

Energy-aware dynamic resource management in elastic cloud datacenters

In clouds, placement of on-demand applications on heterogeneous machines, has turned out to be a crucial research problem, particularly, in terms of performance and energy consumption. Techniques like Dynamic Voltage and Frequency Scaling (DVFS), processor speed adjustment and features such as turning off displays, activating sleep modes, etc. are only useful for decreasing the energy consumption of a single machine, at marginal loss in performance. They cannot be used to achieve significant power optimization in High Performance Computing (HPC) systems such as grids, and cloud datacenters; because power saved by scaling down the processor voltage is far less than switching off a machine. Resource management, using dynamic consolidation of VMs, allows cloud service providers to optimize resource usability, performance and decrease power consumption. This paper investigates various resource management techniques, and suggests several heuristic approaches to optimise energy consumption and performance in elastic datacenters. Using real workload datasets, our evaluation suggests that a combination of the proposed VM allocation and consolidation with migration control technique could save approximately 1.96%–9.38% energy, and improve 0.32%–5.96% performance, as compared to its closest rivals.

Aspectos de inovação na implantação de um centro de digitalização e gestão de dados da pesquisa

Objetiva prover serviços de curadoria e gestão de dados da pesquisa com inovação em P&amp;D a um centro de digitalização e gestão de dados da pesquisa. Caracteriza o ecossistema de P&amp;D em gestão e curadoria de dados da pesquisa, e observa os tipos de processos de inovação que casam com características desse ecossistema. Propõe e cria um ambiente de inovação para o centro, desenvolvendo uma ferramenta fundamentada em trabalho colaborativo, corpos de conhecimento e processos de inovação. Apresenta resultados obtidos a partir do uso do ambiente.Palavras-chave: Inovação. Dados da Pesquisa. Curadoria Digital. Colaboração. Ciência da Informação.Link: https://periodicos.ufsc.br/index.php/eb/article/view/1518-2924.2018v23nespp1